T&T Clark Handbook oF Septuagint Research 9780567680259, 9780567680280, 9780567680273

Table of contents :
Cover
Contents
List of Figures
List of Tables
Preface
Foreword
List of Abbreviations and Sigla
Introduction William A. Ross
Part I Origins
1 The Origins and Social Context of the Septuagint James K. Aitken
2 Septuagint Translation Technique and Jewish Hellenistic Exegesis Marieke Dhont
Part II Language
3 Septuagint Transcriptions and Phonology Pete Myers
4 Septuagint Lexicography Patrick Pouchelle
5 The Septuagint and Discourse Grammar Christopher J. Fresch
6 The Septuagint and Greek Style Eberhard Bons
7 The Septuagint and Biblical Intertextuality Myrto Theocharous
Part III Text
8 The Septuagint and Textual Criticism of the Greek Versions José Manuel Cañas Reíllo
9 The Septuagint and Textual Criticism of the Hebrew Bible John Screnock
10 The Septuagint and Qumran Gideon R. Kotzé
11 The Septuagint and the Major Recensions Ville Mäkipelto
12 The Septuagint and the Secondary Versions Claude Cox
13 The Septuagint and Origen’s Hexapla Peter J. Gentry
14 The Septuagint and the Biblical Canon 207John D. Meade
Part IV Reception
15 The Septuagint and Second Temple Judaism Benjamin G. Wright III
16 The Septuagint in the New Testament Steve Moyise
17 The Septuagint in Patristic Sources Edmon L. Gallagher
18 The Septuagint in Byzantine Judaism Cameron Boyd-Taylor
19 The Septuagint in the Eastern Orthodox Tradition Mikhail G. Seleznev
20 The Septuagint in Early Modern Europe Scott Mandelbrote
Part V Theology, Translation, and Commentary
21 The Septuagint and Theology W. Edward Glenny
22 The Septuagint and Modern Language Translations William A. Ross
23 The Septuagint: The Text as Produced Robert J. V. Hiebert
24 The Septuagint: A Greek-Text-Oriented Approach Stanley E. Porter
Part VI Survey of Literature
25 The Septuagint and Contemporary Study 381Jennifer Brown Jones
Glossary
Contributors
References
Ancient Sources
Biblical/Extra-Biblical References
Modern Authors
Subjects

Citation preview

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

ii

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Edited by William A. Ross and W. Edward Glenny

T&T CLARK Bloomsbury Publishing Plc 50 Bedford Square, London, WC1B 3DP, UK 1385 Broadway, New York, NY 10018, USA 29 Earlsfort Terrace, Dublin 2, Ireland BLOOMSBURY, T&T CLARK and the T&T Clark logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2021 Copyright © William A. Ross, W. Edward Glenny and contributors, 2021 William A. Ross and W. Edward Glenny have asserted their right under the Copyright, Designs and Patents Act, 1988, to be identified as Author of this work. Cover design: Charlotte James Cover image: Lighthouse of Alexandria or Pharos of Alexandria. 19th Century depiction. Simpson Street Press Archives. © agefotostock / Alamy Stock Photo All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. Bloomsbury Publishing Plc does not have any control over, or responsibility for, any third-party websites referred to or in this book. All internet addresses given in this book were correct at the time of going to press. The author and publisher regret any inconvenience caused if addresses have changed or sites have ceased to exist, but can accept no responsibility for any such changes. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Names: Ross, William A, 1987- editor. | Glenny, W. Edward, editor. Title: T&T Clark handbook of Septuagint research / edited by William A. Ross and W. Edward Glenny. Description: New York : T&T Clark, [2021] | Series: T&T Clark Handbooks | Includes bibliographical references and index. | Summary: "A topical handbook to Septuagint scholarship and comprehensive guide for research in the discipline"– Provided by publisher. Identifiers: LCCN 2020044821 (print) | LCCN 2020044822 (ebook) | ISBN 9780567680259 (hardback) | ISBN 9780567680266 (epub) | ISBN 9780567680273 (pdf) Subjects: LCSH: Bible. Old Testament Greek–Versions–Septuagint–Handbooks, manuals, etc. Classification: LCC BS744 .T264 2021 (print) | LCC BS744 (ebook) | DDC 221.4/8–dc23 LC record available at https://lccn.loc.gov/2020044821 LC ebook record available at https://lccn.loc.gov/2020044822 ISBN: HB: 978-0-5676-8025-9 ePDF: 978-0-5676-8027-3 ePUB: 978-0-5676-8026-6 Typeset by Integra Software Services Pvt. Ltd. To find out more about our authors and books visit www.bloomsbury.com and sign up for our newsletters.

For Cantabrigian Septuagintalists of generations past, present, and future.

vi

CONTENTS

L i st

of

F i g u r e s

x

L i st

of

T a b l e s

xi

Pr e f ac e 

xii

Foreword L i st

of

Abbr e v i at i o n s

xiii an d

Si g l a 

xv

Introduction1 William A. Ross

Part I  Origins 1 The Origins and Social Context of the Septuagint James K. Aitken

9

2 Septuagint Translation Technique and Jewish Hellenistic Exegesis Marieke Dhont

21

Part II  Language 3 Septuagint Transcriptions and Phonology Pete Myers

37

4 Septuagint Lexicography Patrick Pouchelle

63

5 The Septuagint and Discourse Grammar Christopher J. Fresch

79

6 The Septuagint and Greek Style Eberhard Bons

93

7 The Septuagint and Biblical Intertextuality Myrto Theocharous

109

Part III  Text 8 The Septuagint and Textual Criticism of the Greek Versions José Manuel Cañas Reíllo

123

9 The Septuagint and Textual Criticism of the Hebrew Bible John Screnock

135

viii CONTENTS

10 The Septuagint and Qumran Gideon R. Kotzé

149

11 The Septuagint and the Major Recensions Ville Mäkipelto

161

12 The Septuagint and the Secondary Versions Claude Cox

175

13 The Septuagint and Origen’s Hexapla Peter J. Gentry

191

14 The Septuagint and the Biblical Canon John D. Meade

207

Part IV  Reception 15 The Septuagint and Second Temple Judaism Benjamin G. Wright III

231

16 The Septuagint in the New Testament Steve Moyise

243

17 The Septuagint in Patristic Sources Edmon L. Gallagher

255

18 The Septuagint in Byzantine Judaism Cameron Boyd-Taylor

269

19 The Septuagint in the Eastern Orthodox Tradition Mikhail G. Seleznev

283

20 The Septuagint in Early Modern Europe Scott Mandelbrote

299

Part V  Theology, Translation, and Commentary 21 The Septuagint and Theology W. Edward Glenny

313

22 The Septuagint and Modern Language Translations William A. Ross

329

23 The Septuagint: The Text as Produced Robert J. V. Hiebert

345

24 The Septuagint: A Greek-Text-Oriented Approach Stanley E. Porter

363

CONTENTS xi

Part VI  Survey of Literature 25 The Septuagint and Contemporary Study Jennifer Brown Jones

381

G l o ssa r y 

397

C o n t r i b u t o r s

407

R e f e r e n c e s

411

A n c i e n t So u r c e s

468

B i b l i c a l /E x t r a -B i b l i c a l R e f e r e n c e s 

470

M o d e r n A u t h o r s

477

S u b je c t s 

481

FIGURES

  3. 1   3.2    3.3   3.4    3.5   3.6 22.1

Proto-Semitic Vowel Sounds Tiberian Vowel Sounds Classical Attic Vowel Sounds Modern Greek Vowel Sounds Diachronic Vowel Change in Conservative Greek Pronunciation Diachronic Vowel Change in Innovative Greek Pronunciation Understanding of Relationship of Target Texts on Source Texts in Modern Language Translations of the Septuagint 22.2 The NETS Semantic “Spectrum”

55 56 58 58 58 59 330 335

TABLES

  3.1 Reflexes of Biblical Hebrew vowels in Tiberian and Greek transcriptions50   3.2 Hebrew and Aramaic consonants 53   3.3 Greek consonants 56 10.1 Details of the Septuagint Manuscripts from Qumran 150 10.2 Details of the Tobit Manuscripts from Qumran 152 10.3 Sample Lists of Qumran Scrolls that are closely related to the Source Texts of the Septuagint 153 12.1 The Main Secondary Versions and their Dates 181 12.2 Ziegler’s Collations for Job 1 with Revisions 183 13.1 Librarian Tenures in Ptolemaic Alexandria 192 13.2 Known Manuscripts of Hexapla Synopsis 194 13.3 Hexapla Fifth Column Manuscripts 195 13.4 Greek Colophon Manuscripts Related to the Hexapla 196 13.5 Syro-Hexapla Colophon Manuscripts 196 14.1 Philo’s Citation of Old Testament and Deuterocanonical Literature 209 14.2 Greek Canon Lists c. 100–400 211 14.3 Greek Canon Lists c. 401–850 214 14.4 Select Greek Manuscripts and Printed Editions 4th–16th 224 19.1 Septuagintal Additions to the Synodal Version 292 25.1 Comparison of Göttingen and Larger Cambridge Critical Editions 385 25.2 Published Volumes in BdA 391 25.3 Published Volumes in SEPT 392

PREFACE

This volume has been several full years in the making. It began its life as a very different sort of project and, as many projects will do, transformed early on into something quite different and more ambitious. As indicated by the title, this Handbook is meant to serve as a resource complementary to James K. Aitken’s T&T Clark Companion to the Septuagint (2015b), which provides a book-by-book overview of the corpus. Septuagint scholarship has benefited from Aitken’s volume for several years now, along with the valuable resource by Kreuzer (2016), which has similar aims as Aitken’s Companion and is now available in English (2019). Others will likely appear, but at the time of publication there is still no resource comparable to this Handbook, which treats the main research issues and methods within the discipline on a topical basis, in an accessible way that is above the introductory level. Our intention has been to provide a guide for specialists and non-specialists alike that helps to cut through the disparate secondary literature and to consolidate and illustrate the wide variety of scholarly approaches to research. We are honored to have worked toward this goal alongside leading scholars in the discipline, who were selected not only for their expertise but also to provide a balanced set of perspectives on the complex topics involved. A project of this scope is never completed without considerable help. As editors we would like to express our sincere thanks to those who labored with us to bring it to fruition. We owe much to Dominic Mattos and Sarah Blake of Bloomsbury T&T Clark for their able and responsive guidance. Jim Aitken also provided much-needed input and guidance at an early stage as the vision for this volume came into clearer focus. We are indebted to the contributors for their excellent work and constructive input as the project proceeded over the years, as well as for their patience through the final stages of completion. A special word of thanks goes to Travis Wright for his valuable assistance in the detailed matters of formatting and consistency over much of the final manuscript (while of course recognizing that any remaining shortcomings lie entirely with us), and to Ken McMullen, who prepared the indices. Important funding to assist with the final stages of the project came from within the framework of the research project Edición y estudio de textos bíblicos y parabíblicos (FFI2017-86726-P) funded by the Ministerio de Ciencia, Innovación y Universidades (Spain). We are also grateful for the support provided by Reformed Theological Seminary. Finally, we note that this volume appears almost exactly one century after the similarly titled work by R. R. Ottley, A Handbook to the Septuagint (1920). May scholars continue to pursue this fascinating and important field of research for the century to come after us and beyond. William A. Ross & W. Edward Glenny Autumn MMXX

FOREWORD

The Septuagint is well served with introductions and guides, but a handbook designed to function as a vade mecum for aspiring researchers and others wishing to work closely with the translation is a very welcome addition to resources. (The term “Septuagint” is a convenience term, as William Ross explains in the Introduction.) The editors, who themselves are prominently active in the field, have chosen a wide range of key topics for discussion and have enlisted a team of contributors well matched to their assignments. Translation is a challenging exercise at the best of times, and even when word-forword correspondence makes good sense in the target language it may still fall short of a perfect fit, as anyone knows who has compared English “I love you” and German “ich liebe dich.” Translation in general may involve gain as well as loss, in the sense that new nuances or inflexions, as well as honest errors, are frequently introduced in the act of translation. There are, to be sure, other reasons why the study of the Septuagint is technically very challenging. First and foremost, there is the number of manuscript witnesses with their multiple variants, groupings, and text-types. The organizing and interpreting of the data require the specialist type of skill sets that this Handbook aims to promote. Great names in textual and biblical criticism have devoted years, if not whole careers, to Septuagintal study—scholars of the caliber of Paul de Lagarde, Frederick Field, Henry St John Thackeray, Paul Kahle, Alfred Rahlfs, Henry Swete, Joseph Ziegler, Isaac Seeligmann, Dominique Barthélemy, and Takamitsu Muraoka. Happily, the Septuagint continues to command the attention of a large community of specialists who are aware of its indispensability in biblical studies and have equipped themselves to handle its complexities. For a long time, the Septuagint was valued in the Christian West principally as a witness to the text of the Old Testament/Hebrew Bible. This made good sense, for the translation, which came together over a couple of centuries, antedates the great majority of our Hebrew-Aramaic manuscripts of the Old Testament by many centuries. Already New Testament quotations from the Old Testament are mostly according to one or other version of the Septuagint. For that reason alone, the Septuagint is a principal resource for students of the New Testament. Moreover, as the earliest Christian communities continued to reach out into the Greek-speaking world of the first and subsequent centuries, the Septuagint, and not the original Hebrew and Aramaic, was their Old Testament. For several centuries it enjoyed in Christian circles a place comparable with that of the King James Bible in the English-speaking world. Today the Septuagint still has canonical authority in the Orthodox Churches, including, of course, the Greek Orthodox Church. As a translation made in the first instance for the needs of Greek-speaking Jewish communities in the Levant, and as the first translation of the Hebrew-Aramaic text of the Old Testament into another language, the Septuagint broke new ground. That meant, for example, that the translators had to weigh the merits of a more literal or more free approach to the work of translation. Were they to be “invisible” as translators, or “slaves to the letter,” as has been said of Aquila? We may surmise that translation theory as

xiv FOREWORD

such did not detain them for long. Interpretative renderings and contemporizing of the text were allowable, and these have contributed to a richly layered translation that also offers insights into the Judaism of the late Second Temple period. In thoroughly modern terminology, the Septuagint functions as a kind of hypertext, triggering connections and associations both within itself and with Jewish thought more widely—an uncovenanted plus that stricter canons of translation praxis would largely have suppressed. The Septuagint remains a text of supreme importance, and one with many questions still requiring answers. This Handbook will prove itself invaluable to anyone wishing to know the status quaestionis in key research areas, the problem “hot spots,” and the kinds of methods and approaches most likely to aid further, fruitful, research. Robert P. Gordon Emeritus Regius Professor of Hebrew, University of Cambridge

ABBREVIATIONS AND SIGLA

HEBREW BIBLE, SEPTUAGINT, AND NEW TESTAMENT 1 Chron.

1 Chronicles

1 Cor. 1 Corinthians 1 Jn 1 John 1 Kgdms 1 Kingdoms (= LXX 1 Samuel) 1 Kgs 1 Kings 1 Macc. 1 Maccabees 1 Pet. 1 Peter 1 Sam. 1 Samuel 1 Thess. 1 Thessalonians 2 Chron. 2 Chronicles 2 Cor. 2 Corinthians 2 Jn 2 John 2 Kgdms 2 Kingdoms (= LXX 2 Samuel) 2 Kgs 2 Kings 2 Macc. 2 Maccabees 2 Pet. 2 Peter 2 Sam. 2 Samuel 2 Thess. 2 Thessalonians 2 Tim. 2 Timothy 2 Tim. 2 Timothy 3 Jn 3 John 3 Kgdms 3 Kingdoms (= LXX 1 Kings) 4 Kgdms 4 Kingdoms (= LXX 2 Kings) Add. Est. Additions to Esther Bar. Baruch Cant. Song of Songs Col. Colossians Dan. Daniel Deut. Deuteronomy Eccl. Ecclesiastes (= Qoheleth) Ep. Jer. Epistle of Jeremiah Eph. Ephesians Esd. Esdras Est. Esther Exod. Exodus Ezek. Ezekiel Gal. Galatians Gen. Genesis Hab. Habakkuk Hab. Haggai Heb. Hebrew Hos. Hosea

xvi

ABBREVIATIONS AND SIGLA

Isa. Isaiah Jas James Jdt. Judith Jer. Jeremiah Jn John Jon. Jonah Josh. Joshua Judg. Judges Lam. Lamentations Lev. Leviticus Lk. Luke Mal. Malachi Mic. Micah Mk Mark Mt. Matthew Nah. Nahum Neh. Nehemiah Num. Numbers Obad. Obadiah Phil. Philippians Phlm. Philemon Pr. Man. Prayer of Manasseh (Odes 13) Prov. Proverbs Ps. Psalms Qoh. Qoheleth (= Ecclesiastes) Rev. Revelation Rom. Romans Sir. Sirach (= Ecclesiasticus) Song Song of Song Song 3 Childr. Song of Three Children (Dan. 3:24-90) Tit. Titus Tob. Tobit Wis. Wisdom of Solomon Zech. Zechariah Zeph. Zephaniah

OTHER ANCIENT SOURCES 1 Apol.

Apologia i (Justin)

Abr. Adv. Jud. Ag. Ap. Anab. Ant. Apol. Apol. Hier. Archid. Aristocr. Barn. Cat. Cels.

De Abrahamo (Philo) Adversus Judaeos (Tertullian) Against Apion (Josephus) Anabasis (Xenophon) Antiquities of the Jews (Josephus) Apologeticus (Tertullian) Apologia adversus Hieronymum (Rufinus) Archidamus (Or. 15) (Isocrates) In Aristocratem (Demosthenes) Barnabus Catechetical Lectures (Cyril of Jerusalem) Contra Celsum (Origen)

ABBREVIATIONS AND SIGLA xvii

Civ. Cohort. Graec. Comm. Cant. Comm. Ezech. Comm. Isa. Comm. Isa. Comm. Jer. Comm. Matt. Comm. Ps. Comm. Ps. Comm. Tit. Crat. De Mens. et Pond. Dial. Doctr. chr. Eccl. theol. Ep. ad Titum Ep. Afr. Ep. Arist. Ep. fest. Ep. Mar. Ep. Ors. 2 Epist. Geogr. Gos. Thom. Haer. Heir Hist. Hist. Hist. Hist. eccl. Hom. Matt. Hom. Ps. Inscr. Psa. Marc. Migr. Mot. an. Pan. Phaedr. Praef. in Pent. Praef. in Sol. Praep. ev. Prol. Sir. Pss. Sol. Ruf. Sel. Ps. Strom. Symb. Vir. ill. Vit. Const. Vit. Cont. Vit. Mos.

De civitate Dei (Augustine) Cohortatio ad Graecos (Pseudo-Justin) Commentarius in Canticum Canticorum (Theodoret of Cyrus) Commentarii in Ezechielem libri XVI (Jerome) Comentarii in Isaiam (Theodoret of Cyrus) Commentariorum in Isaiam libri XVIII (Jerome) Commentariorum in Jeremiam libri VI (Jerome) Commentarium in evangelium Matthaei (Origen) Commentarioli in Psalmos (Jerome) Commentarius in Psalmos (Theodoret of Cyrus) Commentariorum in Epistulam ad Titum (Jerome) Cratylus (Plato) De Mensuris et Ponderibus (Epiphanius) Dialogus cum Tryphone (Justin) De doctrina christiana (Augustine) De ecclesiastica theologica (Eusebius) Epistulam ad Titum (Jerome) Epistula ad Africanum (Origen) Letter of Aristeas Epistulae festales (Athanasius) Epistola ad Marcellinum (Athanasius) Epistula ad Orsisium ii (Athansius) Epistulae (Jerome) Geographica (Strabo) Gospel of Thomas Adversus haereses (Irenaeus) Who is the Heir? (Philo) Historiae (Herodotus) Eusebii Historia ecclesiastica a Rufino translata et continuata (Rufinus) Historiae (Thucydides) Historia ecclesiastica (Eusebius) Homiliae in Matthaeum (John Chrysostom) Homiliae in Psalmos (Origen) In Inscriptiones Psalmorum (Gregory of Nyssa) Adversus Marcionem (Tertullian) De migratione Abrahami (Philo) De motu animalium (Aristotle) Panarion (Adversus haereses) (Epiphanius) Phaedrus (Plato) Praefatio in Pentateuchum (Jerome) Praefatio in libros Salomonis (Jerome) Praeparatio evangelica (Eusebius) Prologue to Sirach Psalms of Solomon Adversus Rufinum libri III (Jerome) Selecta in Psalmos (Origen) Stromata (Clement of Alexandria) Commentarius in symbolum apostolorum (Rufinus) De virus illustribus (Jerome) Vita Constantini (Eusebius) De vita contemplativa (Philo) De vita Mosis (Philo)

xviii

ABBREVIATIONS AND SIGLA

MISCELLANEOUS ABBREVIATIONS A AASF AB ABD

Codex Alexandrinus (or ‫)א‬

Annales Academiae Scientiarum Fennicae Anchor Bible The Anchor Bible Dictionary, ed. David Noel Freedman, New York (1992) ʾAbot R. Nat. ʾAbot Rabbi Nathan AbrN Abr-Nahrain ad loc. ad locum, “at the place discussed” ADGE Anejo Diccionario Griego Español Aeth Ethiopic subversion AGLB Aus der Geschichte der lateinischen Bibel AJEC Ancient Judaism and Early Christianity AJSL American Journal of Semitic Languages and Literatures AJSLL The American Journal of Semitic Languages and Literatures AKM Abhandlungen für die Kunde des Morgenlandes ALGHJ Arbeiten zur Literatur und Geschichte des hellenistischen Judentums Apud “According to” Arm Armenian subversion ASFDHL Academiae Scientiarium Fennicae Dissertationes Humanarum Litterarum ASJT Amsterdam Studies in Jewish Thought ATA Alttestamentliche Abhandlungen Aug Augustinianum B Codex Vaticanus b. Babylonian (for Talmudic literature) B. Bat. Baba Batra B. Meṣ. Baba Meṣiʿa bar. Baraita BASP Bulletin of the American Society of Papyrologists BCE Before the Common Era BdA La Bible d’Alexandrie, Cerf (1986–) BDAG A Greek-English Lexicon of the New Testament and Other Early Christian Literature, 3rd edition, eds. W. Bauer, F. W. Danker, W. F. Arndt, and F. W. Gingrich, Chicago (2000) BDB A Hebrew and English Lexicon of the Old Testament, eds. Brown, F., S. R. Driver, and C. A. Briggs, Oxford (1907) BDF A Greek Grammar of the New Testament and Other Early Christian Literature, eds. Blass, F., A. Debrunner, and R. W. Funk., Chicago (1961) BEB Biblioteca de Estudios Bíblicos BETL Bibliotheca ephemeridum theologicarum lovaniensium BFBS British and Foreign Bible Society BGU Aegyptische Urkunden aus den Königlichen (later Staatlichen) Museen zu Berlin, Griechische Urkunden, Berlin (1895–) BH Biblical Hebrew BHQ Biblia Hebraica Quinta, Stuttgart (2004–) BHS Biblia Hebraica Stuttgartensia, eds. K. Elliger and W. Rudolph, Stuttgart (1983) BibInt Biblical Interpretation Series BIOSCS Bulletin of the International Organization for Septuagint and Cognate Studies (now JSCS)

ABBREVIATIONS AND SIGLA xix

BiPa

Biblia Patristica. Index des citations et allusions bibliques dans la littérature patristique, 7 vols., Paris (1975–2001) BiTS Biblical Tools and Studies BJRL Bulletin of the John Rylands University Library of Manchester BKAT Biblischer Kommentar: Altes Testament BMW The Bible in the Modern World BN Biblische Notizen Bo Bohairic subversion BSCHRC Brill’s Series in Church History and Religious Culture BSIH Brill’s Studies in Intellectual History BT The Bible Translator BWANT Beiträge zur Wissenschaft vom Alten und Neuen Testament BZ Biblische Zeitschrift BZAW Beihefte zur Zeitschrift für die alttestamentliche Wissenschaft C Codex Ephraemi Rescriptus c. circa, “approximately” CahRB Cahiers de la Revue biblique CahRB Cahiers de la Revue Biblique Cat. Catena manuscript CATSS Computer Assisted Tools for Septuagint Studies CBET Contributions to Biblical Exegesis and Theology CBET Contributions to Biblical Exegesis and Theology CBQMS Catholic Biblical Quarterly Monograph Series CBR Currents in Biblical Research CCL Corpus Christianorum, Series Latina CE Common Era CEJL Commentaries on Early Jewish Literature ch. Chapter number CJA Christianity and Judaism in Antiquity CM Classica et Mediaevalia ConBOT Coniectanea biblica: Old Testament Series CP Constantinople Pentateuch CPJ Corpus papyrorum judaicorum, ed. V. Tcherikover, 3 vols., Cambridge (1957–64) CQ The Classical Quarterly CREJ Collection de la Revue des Études Juives CRINT Compendia rerum iudaicarum ad Novum Testamentum CSCO Corpus scriptorum christianorum orientalium. ed. I. B.Chabot et al., Paris (1903–) CSEL Corpus Scriptorum Ecclesiasticorum Latinorum CSHB Critical Studies in the Hebrew Bible CSIC Consejo Superior de Investigaciones Cientificas CSL Cambridge Studies in Linguistics CSSTC Centre for Septuagint Studies and Textual Criticism CurBS Currents in Research: Biblical Studies D Codex Bezae d. Died DCH Dictionary of Classical Hebrew, ed. D. J. A. Clines, 8 vols., Sheffield (1993–2011) DCLS Deuterocanonical and Cognate Literature Studies DJD Discoveries in the Judean Desert DSD Dead Sea Discoveries

xx

ABBREVIATIONS AND SIGLA

DSI De Septuaginta investigationes DSS Dead Sea Scrolls DTS Descriptive Translation Studies e.g. exempli gratia, “for example” ECHC Early Christianity in its Hellenistic Context EMML Ethiopian Manuscript Microfilm Library ErIsr Eretz-Israel esp. especially EstBib Estudios bíblicos et al. et al., “and others” etc. et cetera, “and the rest” ETL Ephemerides theologicae lovanienses Fb Christian marginalia in Codex Ambrosianus FAT Forschungen Zum Alten Testament Fol. Folio frg. Fragment FRLANT Forschungen zur Religion und Literatur des Alten und Neuen Testaments G Codex Colberto-Sarravianus GBBJ Greek Bible in Byzantine Judaism GCS Die griechischen christlichen Schriftsteller der ersten Jahrhunderte GE Brill Dictionary of Ancient Greek GELS Greek-English Lexicon of the Septuagint, T. Muraoka, Louvain (2009) Gen. Rab. Genesis Rabbah GKC Gesenius’ Hebrew Grammar, 2nd edition, ed. E. Kautzsch, transl. A. E. Cowley, Oxford (1910) GKC18 Hebräisches und aramäisches Handwörterbuch über das Alte Testament, Wilhelm Gesenius, 18th edition, rev. Herbert Donner et al., Berlin/ Heidelberg/New York (1987–2010) GOTR Greek Orthodox Theological Review Göttingen edition Septuaginta. Vetus Testamentum Graecum Auctoritate Academiae Scientiarum Gottingensis editum, Göttingen (1931–) Grk. Greek HALOT The Hebrew and Aramaic Lexicon of the Old Testament, L. Koehler, W. Baumgartner, and J. J. Stamm, transl. and ed. M. E. J. Richardson, 4 vols., Leiden (2001) HBCE The Hebrew Bible: A Critical Edition, ed. Ronald Hendel, Atlanta (2015–) Heb. Hebrew HeyJ Heythrop Journal HO Handbuch der Orientalistik HR A Concordance to the Septuagint and the Other Greek Versions of the Old Testament (Including the Apocryphal Books), 2nd edition, E. Hatch, and H. A. Redpath, Grand Rapids (1998) HSK Handbücher zur Sprach- und Kommunikationswissenschaft HSS Harvard Semitic Studies HTLS The Historical and Theological Lexicon of the Septuagint, Tübingen (2020–) HTR Harvard Theological Review HUAS Hebrew University Armenian Series HUCA Hebrew Union College Annual i.e. id est, “that is”

ABBREVIATIONS AND SIGLA xxi

ICC IECOT IOSCS IStratonikeia JAJ JAOS JBL JBR JCPS JGL JGRChJ JHS JIGRE

International Critical Commentary International Exegetical Commentary on the Old Testament International Organization for Septuagint and Cognate Studies Die Inschriften von Stratonikeia, M. Ç. Şahin, Bonn (1981–2010) Journal of Ancient Judaism Journal of the American Oriental Society Journal of Biblical Literature Journal of Bible and Religion Jewish and Christian Perspectives Series Journal of Greek Linguistics Journal of Greco-Roman Christianity and Judaism Journal of Hebrew Scriptures Jewish Inscriptions of Greco-Roman Egypt, eds. W. Horbury and D. Noy, Cambridge (1992) JIPA Journal of the International Phonetic Association JJS Journal of Jewish Studies JNES Journal of Near Eastern Studies JNSL Journal of Northwest Semitic Languages JQR Jewish Quarterly Review JQRSup Jewish Quarterly Review Supplement JSCS Journal of Septuagint and Cognate Studies JSJSup Supplements to the Journal for the Study of Judaism JSOT Journal for the Study of the Old Testament JSOTSup Journal for the Study of the Old Testament: Supplement Series JSP Journal for the Study of the Pseudepigrapha JSRC Jerusalem Studies in Religion and Culture JSRC Jerusalem Studies in Religion and Culture JSS Journal of Semitic Studies JTS Journal of Theological Studies L Lucianic Recension l. Line number Lat. Latin LCL Loeb Classical Library LEH Greek-English Lexicon of the Septuagint, 3rd edition, Lust, Johann, Eric Eynikel, and Katrin Hauspie, Stuttgart (2015) LEIEDS Leiden Indo-European Etymological Dictionary Series LES The Lexham English Septuagint, eds. Penner, Ken M., and Rick Brannan et al., Bellingham (2019) Lĕš Lĕšonénu LHBOTS Library of Hebrew Bible/Old Testament Studies LLT Library of Latin Texts LLT Library of Latin Texts LNTS Library of New Testament Studies LSAWS Linguistic Studies in Ancient West Semitic LSJ A Greek-English Lexicon. 9th edition, eds. H. G. Liddell, R. Scott, H. S. Jones, and R. McKenzie, Oxford (1996). Other editions specified as necessary. LXX.D Septuaginta Deutsch: Das Griechische Alte Testament in Deutscher Übersetzung, eds. W. Kraus, and M. Karrer, Stuttgart (2009) LXX.H Handbuch zur Septuaginta LXXSA Association for the Study of the Septuagint in South Africa m. Mishnah tractate

xxii

ABBREVIATIONS AND SIGLA

Meg. Megillah Mek. Mekilta MGWJ Monatschrift für Geschichte und Wissenschaft des Judentums MH Mishnaic Hebrew Midr. Midrash MS(S) Manuscript(s) MSU Mitteilungen des Septuaginta-Unternehmens MT Masoretic Text MVEOL Mededelingen en verhandelingen van het Vooraziatisch-Egyptisch Genootschap “Ex Oriente Lux” n. Footnote number n.d. no date n.p. no page n.s. New Series NA28 Novum Testamentum Graece, Nestle-Aland, 28th edition Ned. Nedarim NETS A New English Translation of the Septuagint and the Other Greek Translations Traditionally Included under That Title, eds. A. Pietersma and B. G. Wright, Oxford (2007) NewDocs New Documents Illustrating Early Christianity, 10 vols., eds. G. H. R. Horsley, and S. Llewelyn, North Ryde (1981–2012). Second ed. (2009) NHMS Nag Hammadi and Manichaean Studies NIGTC New International Greek Testament Commentary NJahrb Neue Jahrbücher für das klassiche Altertum (1898–1925); Neue Jahrbücher für Wissenschaft und Jugendbildung (1925–36) NKJV New King James Version NovT Novum Testamentum NovTSup Novum Testamentum Supplements NRSV The New Revised Standard Version NRSV New Revised Standard Version NT New Testament NTTS New Testament Tools and Studies O A recension of the text of the fifth column in Origen’s Hexapla (ο΄) O.Deiss. Griechische Texte aus Aegypten, vol. 2, Ostraka der Sammlung Deissmann, ed. P. M. Meyer, Berlin (1916). Papyri nos. 1–45; ostraca (O.Deiss.) nos. 1–92 OBO Orbis biblicus et orientalis OECS Oxford Early Christian Studies OG Old Greek translation of the LXX, sometimes superscripted in a reference after the abbreviation of a biblical book, e.g., Dan.OG 3:9 OGI Orientis Graeci Inscriptiones Selectae, 2 vols., ed. W. Dittenberger, Leipzig (1903–5) OL Old Latin subversion (Vetus Latina) OLA Orientalia Lovaniensia Analecta OSB Orthodox Study Bible OT Old Testament OtSt Oudtestamentische Studiën P.Amh. The Amherst Papyri, Being an Account of the Greek Papyri in the Collection of the Right Hon. Lord Amherst of Hackney, F.S.A. at Didlington Hall, Norfolk, ed. B. P. Grenfell and A. S. Hunt, London: (1900–1) P.Ant. The Antinoopolis Papyri, 3 vols., London (1950–67)

ABBREVIATIONS AND SIGLA xxiii

P.Bod. P.Cair.Zen.

Papyrus Bodmer, Cologny-Génève (1954–97) Zenon Papyri, Catalogue général des antiquités égyptiennes du Musée du Caire, ed. C. C. Edgar, Cairo (1925–40) P.Col. Columbia Papyri (1929–98) P.Count Counting the People in Hellenistic Egypt, eds. W. Clarysse and D. J. Thompson, I: Population Registers (P.Count), Cambridge (2006). Nos. 1–54 P.Fouad. Les Papyrus Fouad I, ed. A. Bataille, O. Guéraud, P. Jouguet, N. Lewis, H. Marrou, J. Scherer and W. G. Waddell, Cario (1939) P.Grenf. 1. An Alexandrian Erotic Fragment and other Greek Papyri chiefly Ptolemaic, ed. B. P. Grenfell, Oxford (1896); 2. New Classical Fragments and Other Greek and Latin Papyri, ed. B. P. Grenfell and A. S. Hunt, Oxford (1897) P.Oxy. The Oxyrhynchus Papyri, London (1898–) P.Petr. The Flinders Petrie Papyri, Dublin (1891–1905) P.Polit.Jud. Urkunden des Politeuma der Juden von Herakleopolis (144/3 ‚ 133/2 v. Chr.), eds. K. Maresch and J. M. S. Cowey, Wiesbaden (2001). (Pap. Colon. XXIX). Nos. 1–20 P.Rev. Revenue Laws of Ptolemy Philadelphus, ed. B. P. Grenfell, Oxford (1896) P.Ryl. Catalogue of the Greek and Latin Papyri in the John Rylands Library, Manchester, Manchester (1911–52) P.Tebt. The Tebtunis Papyri, London (1902–2005) P13 = P.Oxy. 657 P45 See Chester Beatty Biblical Papyri; ed. F. G. Kenyon, London (1933) P46 See Chester Beatty Biblical Papyri; ed. F. G. Kenyon, London (1933) PAAJR Proceedings of the American Academy for Jewish Research para. paragraph PFES Publications of the Finnish Exegetical Society PG Patrologia graeca [= Patrologiae cursus completus: Series graeca], 162 vols., ed. J.-P. Migne, Paris (1857–86) PL Patrologia latina [= Patrologiae cursus completus: Series latina], 217 vols., ed. J.-P. Migne, Paris (1844–64) pl. Plural PLO Porta Linguarum Orientalium PS Proto-Semitic R recto R. Rabbi Ra Indicates a Rahlfs manuscript number (see Rahlfs 2012) RB Revue biblique RBL Review of Biblical Literature RBLG Repertorio bibliográfico de la lexicografía griega, eds. P. Boned Colera, J. R. Somolinos, ADGE 3, Madrid (1998) RbSup Supplementi alla Rivista biblica REJ Revue des Études Juives RESS Review of Ecumenical Studies Sibiu RevQ Revue de Qumran RevScRel Revue des Sciences Religieuses RPLHA Revue de Philologie, de Littérature et d’Histoire anciennes RSV Revised Standard Version RTP Revue de théologie et de philosophie S Codex Sinaiticus SA Samaritan Aramaic

xxiv

Šabb. SBL SBLCS SBLDS SBLEJL SBLPS SBLSCS SBLSS SBLTT SC Sef SemClass SepH.

ABBREVIATIONS AND SIGLA

Šabbat Society of Biblical Literature SBL Commentary on the Septuagint SBL Dissertation Series SBL Early Judaism and Its Literature SBL Pseudepigrapha Series SBL Septuagint and Cognate Studies SBL Semeia Studies SBL Texts and Translations Sources Chrétiennes Sefarad Semitica et Classica “Septuagint Hebrew,” i.e., the Hebrew used by any given Septuagint translator SEPT Brill Septuagint Commentary Series (sometimes BSCS) SGLG Studia Graeca et Latina Gothoburgensia SH Samaritan Hebrew SJOT Scandinavian Journal of the Old Testament SLCTI Scientific and Learned Cultures and Their Institutions SNTSMS Society for New Testament Studies Monograph Series Sot. Soṭah SP Samaritan Pentateuch Sp Spanish SSTC Groupe de Recherche ‘Septante et Critique Textuelle’ SSU Studia Semitica Upsaliensia STDJ Studies on the Texts of the Desert of Judah SubBi Subsida biblica SVTGSup Septuaginta. Vetus Testamentum Graecum Supplements SyH Syro-Hexapla subversion Tan. Tanna/Tannaim TBN Themes in Biblical Narrative TDNT Theological Dictionary of the New Testament, eds. G. Kittel and G. Friedrich, transl. G. W. Bromiley, Grand Rapids (1964) TDOT Theological Dictionary of the Old Testament, 16 vols., eds. G. J. Botterweck, H. Ringgren, and H.-J. Fabry, Grand Rapids (1970–2019) TECC Textos y Estudios Cardinal Cisneros TEG Traditio Exegetica Graeca TENT Texts and Editions for New Testament Study Text Textus THB Textual History of the Bible, Leiden (2016–) THBSup Textual History of the Bible Supplements Them Themelios Tib. Tiberian Hebrew TLG Thesaurus Linguae Grecae TOB Traduction œcuménique de la Bible TQ TESOL Quarterly TRE Theologische Realenzyklopädie, 36 vols., eds. G. Krause and G. Müller, Berlin (1977–2004) TrinJ Trinity Journal TRu Theologische Rundschau TS Texts and Studies

ABBREVIATIONS AND SIGLA xxv

TSAJ TSMEMJ TZ UBS5 UPATS V v. VC VCSup VT VTSup WBC WMANT WU WUNT y. Yal. ZAW ZNW ZPE

Texte und Studien zum antiken Judentum Texts and Studies in Medieval and Early Modern Judaism Theologische Zeitschrift The Greek New Testament. Fifth Revised Edition, eds. B. Aland and K. Aland et al., Stuttgart (2014) University of Pennsylvania Armenian Texts and Studies Verso Verse number Vigiliae Christianae Vigiliae Christianae Supplements Vetus Testamentum Supplements to Vetus Testamentum Word Biblical Commentary Wissenschaftliche Monographien zum Alten und Neuen Testament Würzburger Universitätsreden Wissenschaftliche Untersuchungen zum Neuen Testament Jerusalem (for Talmudic literature) Yalquṭ Zeitschrift für die alttestamentliche Wissenschaft Zeitschrift für die neutestamentliche Wissenschaft und die Kunde der älteren Kirche Zeitschrift für Papyrologie und Epigraphik

SIGLA : ‹ text › « text » < text >

Indicates “variation between” two possibilities

Encloses graphemes in ancient script Encloses allographs in ancient script In a modern translation of ancient primary sources, encloses something not in the original text but implied by the ancient writer author. [text] Encloses phones expressed in the Latin alphabet ° With a numeral, refers to an instance of a word within a given reference (e.g., 1° = first, 2° = second, etc.) * Indicates a conjectural textual reconstruction that is unattested ** Indicates a form that is unattested and did not exist in the language system /text/ Encloses phonemes // Indicates a parallel text in the synoptic Gospels /text/ Indicates text written above the line in a manuscript C Indicates the work of a corrector in an ancient textual witness. May be followed by a superscripted number to distinguish correctors (e.g., BC2). Gö Indicates use of the relevant Göttingen critical edition → Indicates a diachronic linguistic change × Indicates frequency (e.g., 2× = two times) ~ Approximate equivalent αˊ Aquila εˊ The Quinta edition θˊ Theodotion οˊ The fifth column of Origen’s Hexapla (see also O) σˊ Symmachus 𝔐 Masoretic Text as represented by the model codex Firkovitch B19a

xxvi

Introduction WILLIAM A. ROSS

For more than fifty years it has been a favorite pastime of Septuagint scholars and enthusiasts everywhere to point out that interest in our area of study is growing. While this tradition may seem closer to an affirmation of academic relevance or personal cachet than anything else, it has nevertheless been and remains true. This Handbook is certainly part of that trend of real (if slow) disciplinary growth, even as it presents an overview of the fruits of the growth itself and equips readers to contribute to it. Before explaining the nature and purpose of this volume, it may prove helpful to understand the scholarly background from which it emerges.

A BRIEF HISTORY OF SEPTUAGINT SCHOLARSHIP In many ways, Septuagint scholarship coalesced as a distinct academic discipline within the broader academy of biblical scholarship as it is known today in the late nineteenth and twentieth centuries. This development occurred as scholars began (and occasionally even finished) producing textual editions and related resources for part or all of the corpus. To be sure, the Septuagint was well studied long before this period (see Ottley 1920: 62–71; Léonas 2008). As discussed in several chapters in this volume—such as those by Wright, Boyd-Taylor, and Mandelbrote—there is a rich scholarly tradition stretching back to the early modern period and beyond. Even at the beginning of the twentieth century, Swete (1900: vii) could say that the “literature of the subject is enormous.” In Swete’s day, most of that scholarly activity had been—and would continue to be— textual in nature. Several important editions of the Septuagint had appeared long before Swete, such as the Complutensian (1517), the Aldine (1518/19), and the Sixtine (1587). These older works provided a fundamental basis for the eventual appearance of the first critical edition of the Septuagint by Holmes and Parsons (1798–1827). That edition was then followed by the edition of Tischendorf (1850), the important work on the Hexapla by Field (1875), the partial edition of Lagarde (1883), and the concordance of Hatch and Redpath (1897–1906), to name but a few key publications.1 As early as 1875, the University of Cambridge had set out to produce a diplomatic edition based mostly on Vaticanus with a full textual apparatus. Swete produced the smaller, “portable” edition in three volumes (1887–94) as an advance upon the work of Holmes and Parsons. A far more exhaustive critical text—known informally as the Larger Cambridge Edition—was begun by Brooke and McLean prior to Swete’s death in 1917 and made good headway, but it never reached completion (1906–40). Other important scholarly work on the Septuagint began but went unfinished as well, such as the grammars by Helbing (1907) and Thackeray (1909). Also noteworthy are the Septuagint lexicons of Biel (1779–80) and Schleusner (1820–1).

1

2

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

The first half of the twentieth century saw even more concentrated and assiduous scholarship, the beginning of which might be marked by the appearance of Swete’s Introduction to the Old Testament in Greek, first published in 1900. Much of the scholarship in this period would further reinforce and improve the textual foundation upon which a distinct scholarly discipline would be built. The labor was mostly centered in Germany and was spurred on by the industry of Alfred Rahlfs, who assumed the vision of his mentor, Paul de Lagarde, for finally establishing the text of the entire Septuagint corpus. Rahlfs soon joined Rudolf Smend and Julius Wellhausen in heading up the Septuaginta-Unternehmen in Göttingen, where detailed textual research was carried out from 1908 all the way through 2015, at which point the work moved elsewhere. Rahlfs’s initial efforts to produce a new edition of the Septuagint were nearly destroyed by the First World War and the subsequent economic fallout, but he nevertheless managed to complete a two-volume edition of the entire Septuagint in 1935 that greatly improved upon Swete’s text. Rahlfs died soon after, knowing that his edition provided only the next rung on what would prove to be a very tall ladder for later scholarship to ascend. The work of Margolis on the text of Joshua (1931–8) provided yet another rung by which scholars like Montgomery and Gehman would join the climb in their own ways. But the most significant efforts continued at the Unternehmen, where a series of exhaustive, eclectic texts began to appear under the title Septuaginta: Vetus Testamentum Graecum. Prior to his death, Rahlfs himself produced an edition of the Psalter in 1931 as the first volume in this series, which has become known colloquially as the Göttingen edition. While certainly not unimpeachable in every detail, it is difficult to overstate the importance of the Göttingen edition as the best reconstruction of the original text of the Septuagint on the basis of all available evidence. It is the point of departure for all scholarship in the discipline, with the exception of studies that focus on textual reception at later periods in transmission history. With this historical background in view, the growth of Septuagint scholarship as a discipline in the latter half of the twentieth century—and the slowly increasing attention given to it by neighboring disciplines—should be linked with two other factors. The first is the work of Sidney Jellicoe, whose Septuagint and Modern Study (1968) set a new benchmark for the discipline insofar as it provided a much-needed update to Swete’s Introduction. But even more important for Septuagint scholarship at large was Jellicoe’s labor to help foster more collaboration among researchers by means of establishing a coordinated professional society. In 1967, Jellicoe recruited the help of his colleague Harry M. Orlinsky, who in a (handwritten) letter dated June 1st agreed “wholeheartedly” with Jellicoe’s “most happy suggestion to form a functional LXX circle.” With this goal in view, Orlinsky also shared Jellicoe’s desire to launch “an official Bulletin of LXX Studies, with articles, bibliography, reviews, etc.” Together with Charles T. Fritsch, they began to lay plans to coordinate their efforts with the annual meeting of the Society of Biblical Literature. They agreed upon the name “International Organization for Septuagint and Cognate Studies” (IOSCS), which met officially for the first time in Berkeley, California, on December 19th of 1968, and which has continued to meet annually ever since (although in 2020 it was of course a virtual meeting). In this way, Jellicoe not only made significant contributions in his research and writing, he also set into motion a scholarly society that over the past fifty years has become a critical center of gravity in the discipline. The second factor in the latter half of the twentieth century has been the continued— albeit slow—appearance of further volumes in the Göttingen edition. These have continued to emerge every handful of years ever since Rahlfs’s initial contribution in

INTRODUCTION 3

1931. By far the most prolific scholars in this regard have been Joseph Ziegler, John W. Wevers, and Robert Hanhart. The difficulty of producing these critical texts and the care that goes into them are most clearly evident in the fact that, although great progress has been made over the past century, the Göttingen edition still has eight volumes left before it reaches completion. Lacking a critical text that covers the entire Septuagint corpus has certainly presented theoretical challenges for the discipline over the years. For example, scholars have long debated whether and how a grammar or a lexicon of the Septuagint is possible without an established text for the full corpus. For better or worse in such cases, Rahlfs’s edition, which was eventually revised by Hanhart in 2006, has typically filled the lacunae of the Göttingen edition. With both the gradual appearance of rigorous critical texts and the regular meetings of the IOSCS, Septuagint scholarship has had more reliable data to work with and regular occasions to collaborate in so doing. As the last fifty years have passed, these developments have provided increasingly better conditions for Septuagint scholars to address major desiderata in the secondary literature that in parallel areas of biblical studies like New Testament scholarship have long been complete. For example, in the past thirty years alone, Septuagint scholarship has produced (or begun to produce) new or fresh translations into modern languages (e.g., BdA, NETS, LXX.D, LBGS), two lexicons (LEH, GELS), a vocabulary guide (Lanier and Ross 2019), a syntax (Muraoka 2016), a reader’s edition (Lanier and Ross 2018), two commentary series (SBLCS, SEPT), several introductions (e.g., Fernández Marcos 2000b; Dines 2004; Jobes and Silva 2015), companion volumes (Aitken 2015b; Kreuzer 2019; LXX.H), and monograph series (e.g., SBLSCS, DSI). Such concentrated growth of the discipline has been slow in the grand scheme of scholarship over the past century, but its appearance has not gone unnoticed. This brief history of the discipline is certainly reductionistic in many ways. But to the extent that is the case, the space limitations of an introduction like this one have already helped to identify one more area for continued research. For further reading on the history of Septuagint scholarship, see especially Ottley (1920: 62–80), Katz (1956), Wevers (1968), Jellicoe (1969), Seeligmann (1990), Jobes and Silva (2015: 265–88), Meiser (2018), Pietersma (2019), and the geographically specific articles on the history of scholarship in JSCS volumes 51 and 52. Much more could also be said about recent tools and resources for Septuagint research, as is typical of a volume introduction like this one. Rather than rehearsing those here, the reader is directed to the substantive chapter by Jennifer Brown Jones in this Handbook.

THE PURPOSE OF THIS HANDBOOK As Septuagint scholarship has grown, so too has the awareness of its importance among scholars in parallel disciplines within biblical studies and even Classics. Simultaneously, as Septuagint scholarship has grown, it has become more complex, and so it has become increasingly difficult for non-specialists to approach the major areas of research. This problem is particularly acute given the small number of specialists who are actively training graduate and postgraduate students in the discipline relative to other areas of biblical scholarship. Even among specialists, the debates and developments within Septuagint research over the past fifty years have made it difficult to keep current. This  Handbook thus joins the broader trend within academia to help mitigate the proliferation of scholarship by providing an up-to-date overview of the discipline in a single volume.

4

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

This Handbook approaches that task with several subsidiary goals in view. While other resources exist that provide a book-by-book introduction to the Septuagint corpus (e.g., Aitken 2015b; Kreuzer 2019), to date no other resource exists that introduces and illustrates the major research topics in the discipline. This volume presents twenty-four such topics in the chapters that follow, each of which was written by an expert in the area of research under discussion. Generally speaking, the chapters have four sections: first, introducing the research topic; second, presenting an overview of views and/or debates related to it; third, discussing the relevant research methods, theories, or tools; and fourth, highlighting ongoing research questions. Each chapter concludes with a brief annotated bibliography pointing out and describing several key resources for that area of research. While most chapters in this volume adhere to this broad structure, in some cases the section headings or the structure itself may differ. The variation is a result of the fact that not all research topics can be handled in precisely the same way. As editors, we felt it was best to have contributors decide how best to present their topic, if it was necessary to depart from the structure outlined above. Even in such cases, the same overall goals are in mind. In particular, the chapters in this Handbook are intended to introduce the reader to key research topics, familiarize the major views and debates, and equip readers to undertake research in that area of Septuagint scholarship themselves. All contributors have labored to make the contents of these chapters accessible, so that graduate students and scholars in parallel fields could benefit from them as much as possible, yet without sacrificing scholarly rigor. Only the reader can say whether we have achieved that delicate balance.

APPROACHING THE MEANING OF “THE SEPTUAGINT” Throughout this introduction the concept of the Septuagint has been left entirely undefined, and intentionally so. But it is necessary at this point to spell out in more detail what exactly is the object of research in this Handbook. Scholars often point to the terminological problems involved in talking about “the Septuagint.” At its best, this term, along with its common abbreviation “LXX,” is a convenient way of referring to the ancient Greek translations of the canonical Hebrew Bible, along with works commonly called apocryphal or deuterocanonical, some of which were originally written in Greek rather than translated from Hebrew or Aramaic. When used this way, the term “Septuagint” typically implies the corpus of Greek writings collected in a modern edition like that of Rahlfs and Hanhart (2006). Indeed, it is this general sense that is intended in the title to this Handbook, as in most contemporary scholarship in the discipline. But in a very real sense there is no Septuagint. This is not at all to say that Greek translations of the Hebrew Bible were not made by Jews in the pre-Christian era. They most assuredly were. Rather, it is to point out that speaking of “the Septuagint” may give the erroneous impression of uniformity or synchronization in the production of the corpus, when in fact the reality was far different. The term “Septuagint” itself derives from the Latin numeral for seventy, septuaginta or LXX, which appears in ancient authors (sometimes in the Greek ε‛βδομήκοντα) as a shorthand reference to the legendary seventy-two translators purported to have miraculously produced the Greek Pentateuch in the Letter of Aristeas (see Josephus, Ant. 12.57, 86; Irenaeus, Haer. 3.21; Justin, Dial. 120; Augustine, Civ. 18.42). In time, particularly in early Christian communities, the term began to refer to the broader corpus mentioned above in similar terms (e.g., Justin Martyr, 1 Apol. 31, Dial. 71–3; Eusebius, Eccl. Hist. 6.16.1). As centuries passed, authors

INTRODUCTION 5

began to use the numeral seventy in various languages as a grammatically singular mass noun (e.g., “la Septante” in French), as Williams (2011) has aptly demonstrated. While it is very likely true that the Greek Pentateuch was produced first and in a coordinated way (Lee 2018), things are much less clear for the rest of the corpus, which emerged over the course of three or four centuries. Important aspects are not often firmly established, such as precisely when any given book of the Hebrew Bible was translated into Greek, by whom, where, and for what purpose(s). Scholars generally agree that in many, perhaps most, cases, it was not long after the original Greek translation was completed that copying and/or revision began, sometimes even full recension, although again the motivation for this activity is uncertain. What is clear is that, in antiquity, the Greek writings commonly referred to as “the Septuagint” were in fact not fixed in their precise wording or scope. The ambiguity of the term “the Septuagint” should be clear at this point. While there are reasonable ways to use it without qualification at a broad level, caution is always warranted in more detailed matters. In this connection, scholars tend to use the term “Old Greek” (OG) when discussing the original translation of a given book in distinction from any and all subsequent copies, revisions, or recensions. To deal with this complexity, each chapter in this Handbook specifies the meaning of its terminology wherever necessary.

6

PART ONE

Origins

8

CHAPTER 1

The Origins and Social Context of the Septuagint JAMES K. AITKEN

INTRODUCTION At a certain point in the Hellenistic period, someone sat down to write out a translation in Greek of at least one of the books of the Pentateuch. Contemporary accounts that might tell us anything more about this event are lacking. We have no account of who the translators were, how they worked, where they worked, what their fluency was in each language, or their very reasons for undertaking a translation in the first place. This gap in our knowledge has left a vacuum, to which many theories are drawn. While it may be true that we have very little knowledge of the translators or indeed of Jews in Egypt at the time, this does not mean we know nothing of the setting. The social and historical context has increasingly been revealed to us by the finds of papyri and inscriptions from Ptolemaic Egypt that shed light on the circumstances around the translation. This allows for evaluation of the validity of theories and the premises that they presuppose, and for the investigation of new areas that have been neglected despite the finds. Here the focus will primarily be on the first translations, namely, of the Pentateuch; every Septuagint book may have a separate history, but once the first books had begun to be translated, their influence on later translation practice is clear. Our earliest ancient accounts of the translation found in Aristobulus and the Letter of Aristeas (see Wright’s chapter in this volume), although only deriving from the second century BCE, locate the translation within Ptolemaic Egypt, and specifically under Ptolemy I or Ptolemy II (305/4–282 and 283–246 BCE). Details in their accounts, especially Aristeas’s presentation of royal sponsorship for the translation and the placement of the translation in the library of Alexandria, are more likely attempts at bolstering the status of the Jewish community in Alexandria in the later second century than reliable historical information (Wright 2011). Nonetheless, some modern theories accept an element of historical memory in these legends and begin their reconstructions from this evidence (see, e.g., Honigman 2003; Rajak 2009). Even if the reliability of such historical narratives is doubted, the internal evidence within the Septuagint translation itself provides data that at times corroborate the picture learned from those sources. We are therefore confident regarding the larger contours into which the translation may be placed. The translation technique and language of the five books of the Greek Pentateuch are consistent, indicating that they were all translated within a similar time period, if not by collaboration between translators (Lee 2018). Each

10

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

book, however, also presents subtle differences in technique that suggest that there was a different translator for each. The language has been placed within the early period of post-Classical Greek, confirming the date of the third century (Lee 1983), or at the latest the early second century, for the translation of these books. Confirmation that it could not have been translated any later comes from citation in second-century sources, even if those sources cannot be precisely dated within the century, and the existence of secondcentury fragments of the Septuagint (for Deuteronomy P.Ryl. 458 and 4QLXXDeut; for Leviticus 4QLXXLeva). The specific location of Alexandria cannot be supported from this internal evidence, but Egypt certainly can, owing to the presence of Egyptian loan-words (e.g., θίβις “container” and οἰφί, a unit of measurement), local administrative terminology (ἀρχισωματοφύλαξ “chief bodyguard” [a court title], τοπάρχης “governor of a district”), and perhaps even theology (Görg 2001). Beyond these small pieces of evidence little more can be said with confidence about the origins. Theorizing by scholars is dependent upon whether greater priority is given to the scenarios envisaged in Aristeas and elsewhere or to features detected within the Septuagint itself. The scenario in Aristeas implies an impetus for the translation that is external to the Jewish community, motivated by Ptolemaic interest in the people and their writings. The majority of theories imply a degree of internal Jewish need, and especially how Jews accommodated to their new home in Egypt. To each of these starting points are then added assumptions about the status of Jews in Ptolemaic Egypt and their social environment, which now can be illuminated by literary and papyrological data from Egypt. The theories that have been offered on origins shed light on three broad categories: the need for the translation, the social background of the translators, and the mechanisms of producing the translation.

KEY DEBATES In the nineteenth century, scholars sought to reconstruct the setting of the translators from the limited information on Jews in Egypt that they had access to. Frankel (1841: 9–11), for example, supposed that the translators must have come from among the Jews living in the chora of Egypt, while Grinfield similarly argued that they were descended from Egyptian Jews, most likely soldiers who now formed a proportion of the commercial population (Grinfield 1850: 23). They both adduced from the sparse evidence collected in Josephus and other historians that Jews served in the Egyptian and Ptolemaic armies. Swete’s popular Introduction (1914: 7–9) draws upon similar data from Egypt but already presumes a location in Alexandria. It is an irony that scholars in the twentieth and twentyfirst centuries, with access to a vast range of new data, have tended to offer explanations without any attention to such new evidence. There have been occasional exceptions that will be noted, but theories regarding the context have largely circumvented the actual evidence.

THE NEED FOR A TRANSLATION The starting point for many has been the assumption that ignorance of Hebrew in the second generation of immigrants to Egypt led to a need for the translation. Porter is one of the few to express this in terms of bilingualism, proposing that the Jews were either native speakers of Greek (L1) or, if Greek had been their second language (L2), that their first language (L1) was a local language (Porter 2014: 288). He does not provide

THE ORIGINS AND SOCIAL CONTEXT OF THE SEPTUAGINT 11

substantiation for his point, but the implication is that Jews of Egypt lacked knowledge of Hebrew while being fluent in Greek. The language of the translation does suggest they attained the level of native speakers of Greek, in their natural rendering of Greek syntax that did not arise from adherence to Hebrew word order (Janse 2002: 380–1). This attainment reflects competence in the language but does not provide a clear direction on the ethnicity or native language of the translators. Apparent misunderstanding of the Hebrew text also could suggest a lack of familiarity with Hebrew. However, traces of the influence of Aramaic and contemporary Late Biblical Hebrew on the translators (Loiseau 2016) suggest a native speaking ability of Hebrew. The translators read the text according to their own knowledge of the language rather than having been schooled in the classical Hebrew tradition. Nonetheless, some strands of scholarship have assumed that Jews were not proficient in Greek, the proof for this lying in the translation itself. In earlier generations the apparently lower level of Greek was contrasted to Classical Greek models and taken as evidence of the lack of social integration of Jews in the diaspora (see Aitken 2014c). This recently has been expressed in more positive terms. The translation features, such as matching Hebrew word order and syntax and the creation of new religious vocabulary, have been taken as indicative of an internal Jewish need for the translation. It provided opportunity for Jews to express their identity in the Hellenistic environment, to maintain their identity as Jews, and articulate their attachment to the Hebrew language (Rajak 2009). The application of contemporary translation theory to the language of the Pentateuch has also concluded that the translation technique is indicative of a countermovement to Hellenistic acculturation (Ziegert 2017). Certainly the Septuagint as a self-conscious statement of Jewish identity has often been compared to Manetho’s Aegyptiaca, a translation of Egyptian king-lists into Greek from the third century BCE. The difference, though, is that Manetho’s work seems to have been a rewriting for a Greek audience, and in a literary form to appeal to Greeks (Moyer 2011: 131), not a translation adhering closely to its source text. Such theories on the language are predicated on a particular understanding both of the linguistic makeup of the Greek and on the social environment of the translators. In contrast, research has demonstrated how, in addition to native competency already noted, the vocabulary is typical of post-Classical Greek (Lee 1983) and how the Greek is not to be dismissed for its poor quality. The language reflects a particular stage in the history of Greek and at the same time is composed in a register that is not of the highest literary standards but is not without literary features. It is not as peculiar as it might first seem. Interference from the translation process does result in unexpected idioms and a higher degree of particular forms, but this need not be a specific Jewish strategy of the translators. Examination of other translations in antiquity, including in Egypt (Aitken 2014), implies the Septuagint translators were following conventional translation methods and not expressing a unique identity through language. Internal Jewish need for the translation is still the most likely option, even if conclusions cannot so easily be drawn from translation technique. A popular suggestion has been that the translation stems from the liturgical use of the Pentateuch (Thackeray 1921: 9–15), although we know little of the liturgical life at this time. Jewish prayer houses do appear in Egypt in the third century BCE, as attested by dedicatory inscriptions (e.g., JIGRE 22; 117), but there is no certainty whether reading Torah was part of that liturgy. An alternative scenario is that of the school room. For Jews in Egypt wishing to learn about their heritage, the Greek translation would have provided a guide to reading the Hebrew original. The style of the Greek translation, following very closely word order

12

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

and syntax of the Hebrew, has been seen as indicative of this educational use (Pietersma 2002; cf. Brock 1972: 16) perhaps even for elementary learning of the Hebrew language. If the translation technique is a standard method for translation in Hellenistic times, it cannot necessarily serve as a sure guide to the purpose (Aitken 2016). Nonetheless, the schoolhouse as a setting is possible, especially if we move away from the focus on learning Hebrew. The agreement between the translators for consistent rendering of Hebrew words (such as διαθήκη for “covenant,” εὐλογέω “to bless,” and δόξα for “glory”) implies that the translations began in an oral context where the equivalents had been established over time (Aejmelaeus 2013). This could well have been in a context where the Hebrew biblical scrolls were being discussed in a Greek medium, and the schoolhouse is one such possible location. One need not imagine, as does Pietersma (2002) or Brock (1972), that the Septuagint served for Jewish education the same function as Homer in Greek education. For Greek-speaking Jews, Homer would have been a text for study since memorizing the Greek poet was the only means of learning Geek literacy in antiquity (Cribiore 2001). Homer was also the model of poetic work and would have held a high status alongside the Septuagint, each serving their separate purposes; one for literary and poetic models, the other for national history and theology. The suggestion that there was an oral tradition behind the translation can be illuminated by the consistency in translation choices. Agreement over how to translate key religious terms (see above) implies decisions being made before the translation has been penned (Aejmelaeus 2013). Likewise, subtle translation choices such as ὀσμὴ εὐωδίας (“a fragrant odor”; e.g., Gen. 8:21; Exod. 29:18), θυσιαστήριον, ὁλοκαύτωσις require thoughtful consideration and could not appear in more than one book by chance (cf. Lee 2018). However, this oral background is a natural development of the familiarity of any text within its community. It is not support for a larger theory, proposed by Kahle (1959), that the Septuagint began in the manner of an oral Targum. Kahle built upon the existence of translation variants and the allusion in Aristeas to possibly defective translations (§ 30), that there was no one original translation since it began in a context of oral synagogue translation. As already noted, we cannot reconstruct anything of the early synagogue liturgy, but Kahle should not be entirely rejected. In reality there is some early variation between manuscripts and, coupled with the revision movement, it is possible to show that textual variation appears early in the tradition (Fernández Marcos 1994a). While the larger contours of Kahle’s theory have not been accepted, recognition of some oral transmission is feasible, and this would conform to recent discussion over the open-ended nature of writing in antiquity (Mroczek 2016). The theory of a legal need for the translation combines elements of both Jewish and Ptolemaic support. Mélèze-Modrzejewski (1996) made the case that the Jews were governed by their own laws in Egypt, and, on the model of Demotic Egyptian laws requiring an accompanying Greek translation, a Greek translation of the laws was produced. His view has been largely rejected owing to lack of evidence for Jewish legal self-governance, his one case of papyri from the third century that refer to Jewish litigants proving to be tentative. Apparent supporting evidence has come to light more recently in the second-century BCE Jewish archive from Heracloepolis in the Egyptian Fayum, where the language of a “bill of divorce” (βιβλίον ἀποστασίου) is derived from Deut. 24:1-4 (P.Polit.Jud. 4). It may not confirm, however, that the Septuagint was first translated for legal reasons, but rather show how it gradually gained importance and influence by the later second century (Rajak 2009: 85). The legal theory should not be too readily dismissed, nonetheless, since it has the advantage of working from data that comes from

THE ORIGINS AND SOCIAL CONTEXT OF THE SEPTUAGINT 13

Egypt in the same time period. Indeed, a recent study, while not fully confirming the legal theory, substantiates portions of the argument, by using papyri for the reconstruction of the social context. A comparison of linguistic features in the Septuagint with documentary papyri has led to the conclusion that the translators were most likely trained among the Egyptian scribal class. Their translation technique, vocabulary, and register all display features found among the Greek-language-trained scribes of Egypt. As such the translators were working in similar ways to Greek scribes in Egypt, and we cannot infer that the translation was conceived as a literary work if it displays the same features as documentary texts (Aitken 2014; cf. Honigman 2017). Mélèze-Modrzejewski’s legal theory implies a mild, almost indirect, support from Ptolemy, and therefore is poised between Jewish internal and external reasons for the translation. Attempts to associate the translation with the library, if not under direct sponsorship, also sit between the two explanatory models. Had we not the Letter of Aristeas, would scholars have associated the translation with the library of Alexandria? While recognizing that placement in the library was unlikely, some suggest the “bookish air” of Alexandria would have encouraged the translation (Aejmelaeus 2013: 13; Kreuzer 2015a: 19–20). This approach places it within a scholarly environment, if only by association. However, it overlooks the wide dispersion of books in antiquity: it was possible to be educated in many different circles especially when Greek literary works are found in an array of archives in Egypt (van Minnen 1998). Egyptian temples, Greek gymnasia, and household archives have all left behind collections of papyri that we may justifiably call libraries. To be scholarly did not require access to a state-sponsored library. Potential evidence of Byzantine authors recording the existence of translated books in the library (Hatzmichali 2013) should be doubted, since such accounts are themselves derived from Aristeas. Nevertheless, by paying attention to the diversity of content in such archives, and by breaking down the mythical image of the Alexandrian library, Honigman still leaves open the possibility that the Septuagint could have been deposited in the Alexandrian library (Honigman 2017). The important point from her work is that the library is not the elite cultural center envisaged in the ancient sources and, therefore, would not have been opposed to having the Septuagint on deposit. Ptolemaic support, therefore, for the translation is still maintained as a possibility, even if not to the degree envisaged in Aristeas. The suggestion that the translators are to be found among documentary scribes does not rule out access to literature and a high literate education but, rather, supports the idea of their not being close to the library in Alexandria. A highly educated environment for the translators has also been surmised by Honigman (2003) from the Letter of Aristeas, wherein the translation is compared to the textual work of Homeric scholarship. However, Honigman has been bold among scholars in later retracting her own view, doubting that Aristeas is informative for Septuagint origins (2011). Some have, even so, gone further and suggested that Ptolemy II’s known interest in minorities might have meant there is some truth in the king’s wish to include the book in his library (Carleton Paget 2014). Indeed, the inclusion of the translation is seen as reflective of his interest in a people whose land was at the time a matter of political rivalry with the Seleucids (Honigman 2003: 117). Without Aristeas, royal sponsorship for the translation is unlikely to have been an obvious derivation from the evidence. Prayer house dedications indicate a concern for recognition from the Ptolemies within the community (e.g., JIGRE 22: “On behalf of King Ptolemy and queen Berenice his sister and wife and their children”), but this does not seem sufficient to prove actual sponsorship. While some would point to the large size

14

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

of the translation and therefore the cost of its production that would require sponsorship, this is dependent on the understanding of the mechanisms of translation (see below). With each book being undertaken by a different translator, the size is reduced. Even if sponsorship were required, a Jewish community that could sponsor the building of prayer houses could probably afford to pay for a translation. Language has again been called into play in this debate to oppose the idea that the translation would have to have been undertaken under royal sponsorship. On this reasoning the Greek is not appropriate for a king as it reflects a popular level of Greek or excessive interference from the Hebrew source text (Wright 2011). While it is probably correct that the translation was not made for the Ptolemy, the case cannot be mounted so easily from language. Such reasoning places too high an expectation on the part of the monarch, especially for a work that is not dedicated in its preface to the king. Petitions addressed to the Ptolemy (such as P.Cair. Zen. 1.59075 and 1.59076) are written in a similar register to the Septuagint and yet are explicitly intended for the king.

THE SOCIAL BACKGROUND OF THE TRANSLATORS The identity, social background, and education of the translators are important indicators for the origins of the Septuagint, even though we cannot offer firm conclusions. While a strand of scholarship portrays them as struggling with the language of both Greek and Hebrew, the presentation of their incompetency should not be overstated. Their knowledge of Hebrew and especially their apparent engagement with scholarly readings of the Hebrew text (e.g., Frankel 1841) indicate a well-educated group of people trained in Greek literacy and in the reading of the Hebrew text. They are occasionally mistaken in their understanding of features of classical Hebrew (Tov 1999a) and therefore do not have the same level as the scribes of the Qumran Hebrew manuscripts, but their competence is in a reading knowledge of Hebrew and a writing level in Greek. Van der Kooij, in a number of studies (e.g., 1998), has argued that the translators were scribes and that they would have had to have come from Jerusalem to have the knowledge and understanding of the Hebrew text. This coincides with the account of Aristeas but is derived from a reading of the translation and indicating the interpretative elements within it. It has to be admitted that “scribe” is a broad term that does not assist in clarifying precisely who they were, and the reconstruction works from the assumption that biblical exegesis could not be found in Egypt. Other theories that see a scholarly context for the translation locate it in Alexandria by contrast, close to the Alexandrian library (e.g., Fernández Marcos 2009b). Whether in Jerusalem or in Alexandria, these theories do not admit of scholarly activity elsewhere and do not take into the account the distribution of libraries in many localities in antiquity and especially in Egypt, discussed above. A variation on this theme is Tov’s suggestion, while admitting that the Pentateuch stems from Egypt, that translations after the Pentateuch must have come from Jerusalem since that is where Hebrew scrolls are to be found (Tov 2010). In opposition to this scholarly or Jerusalem model, the nineteenth-century portrayals have their reflex in more recent literature. Theories are sometimes built on the fact that the Jewish population in Egypt was primarily soldiers or mercenaries (cf. Fernández Marcos 2009b: 85, “mainly composed of soldiers, mercenaries and peasants”). This therefore requires the translators to have some connection with Jerusalem and the educated elite, and to be separate from their local populations in Egypt. There is a danger

THE ORIGINS AND SOCIAL CONTEXT OF THE SEPTUAGINT 15

in such an argument that soldiers are presumed to be of a lower level of education. The publication of the documentary material from Egypt shows that this is not the case. The sources indicate that Jews in many locations across Egypt were working in various strata of society. Accordingly, another hypothesis examines linguistic features of the Septuagint and compares them to papyri (Aitken 2014, 2016). This leads to the hypothesis that the translators learned both their Greek and their translation method in the same milieu as trained Greek scribes in Egypt. It does not presume a level of education and cannot identify a precise social status, but it does suggest they might have earned their employ as administrators. The importance of knowledge of Greek for social preferment in Egypt would support this. Tax lists (P.Count) as early as the second half of the third century BCE indicate many ethnic groups, including Jews, being classed as Hellenes (Aitken 2014). This tax class seems to have been sought out by many, presumably acquired by going to the gymnasium or participating in Greek cultural activities, since knowledge of Greek became important under the Ptolemies for employment in many administrative functions. Such social preferment would also mean that the translators could be located in various districts of Egypt where they would find employment and where we now know there were local libraries and access to books. Such a hypothesis distances the translators from a purely Jewish environment and places them in the world of Ptolemaic Egypt. It also removes them from the shadow of the Alexandrian library without separating them fully from contact with Greek literature and educational circles.

THE MECHANICS OF TRANSLATION One aspect of origins that has only received passing attention is the actual mechanism and method of translation. In one sense it is implied in the study of translation technique, but it also underlies assumptions that people draw about the significance and sponsorship of the translation. Thus, if the translation is conceived of as a major undertaking, then it would have required sponsorship. Such sponsorship could be sought either from wealthy Jewish benefactors or from the Jerusalem temple, as partially implied in Aristeas, or even from the Ptolemaic king himself, as fully stated in Aristeas. The wealth of the kingdom is implied in Aristeas (§27) by the donation of 660 talents for the liberation of the slaves. It is clear that closer attention to the mechanics of translation is fundamental to understanding the origins. The linguistic environment of Ptolemaic Egypt sheds some light on the writing processes of the time and can be used alongside suggestions from common sense by scholars. Talk of the uniqueness of the Septuagint has tended to lead to conclusions regarding its large scale and cost. However, if the translators are school teachers or administrative scribes earning their daily keep through their work, they would have the Greek literary skills and a modest income in order to undertake the translation with little trouble. Additional funding could have come from the Egyptian Jewish community, whose building of prayerhouses attested by the surviving dedicatory inscriptions (e.g., JIGRE 22; 117, both from the third century BCE) demonstrates access to public funds. As each book was written by a different translator, any one translator would have translated no more than fifty chapters (the number in Genesis), and given the evidently straightforward translation method and apparently pre-agreed list of terms, this could have been completed in a short space of time. Van der Louw has experimented with modern translators, using the

16

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

model of three persons—one reading the Hebrew, one translating into Greek, and a third writing the Greek down—and calculated it would have taken 10.2 weeks (van der Louw 2010). Even this may be too long if we do not follow the complex three-way writing method that he envisages. The method of working does seem largely to have been phrase by phrase, which would have speeded up the process. We should not presume, however, a rigid method of segmented translation that has been inferred by some from the work of Soisalon-Soininen (1987). The latter identifies four features that suggest the translation was undertaken in small phrasal units without attention to the larger syntax of the sentence leading to repetition, the apodotic καί (representing Hebrew wāw), the stilted translations of the ‫ב‬ (bêt) preposition when followed by the infinitive construct, and the pleonastic pronoun in relative clauses. There is inconsistency throughout the translation indicating both segmented translation and also sensitivity to context. Contextual awareness can be seen both on the smaller level of occasional idiomatic renderings (e.g., Πῶς ἔχετε, “How are you?” in Gen. 43:27) and at times in the sensitivity to the larger discourse through careful use of particles (Fresch 2015). Harmonization beyond the level of the verse and across larger segments might also suggest the translation was undertaken with some forethought. We see, for example, in Gen. 2:1 the insertion of the word “book” (ἡ βίβλος γενέσεως, “the book of the creation”) in anticipation of the same full expression some chapters later (Gen. 5:1). If the translation still bears the mark of segmentation it implies that the translator did not return and correct his work afterward (cf. van der Louw 2019). However, examination of writing methods in the Hellenistic period and other features within the Septuagint make the scenario of segmentation without any editing afterward unlikely. It is clear that many works were dictated, as evidenced by references to the practice: on the one hand the social status of the intellectual elite meant it was below their office to write even when they could (Harris 1989: 248–9), while on the other correspondents in letters often admit that they are unable to write (e.g., O.Deiss. 58, second century BCE). Furthermore, a change in handwriting for the signature of a letter establishes that the correspondent or “author” is not the scribe writing (e.g., P.Cair. Zen. 2.59287, third century BCE). This allowed the dictator and/or the scribe to check and read through the document after it had been written. Ptolemaic papyri letters also show emendations and corrections, in many cases since the surviving letter is a first draft before the finalized version is sent, but also even in the final version (Aitken 2014: 131–2). Editing, therefore, was a natural part of the writing process such that writing and editing according to how it sounds to a Greek are all one and the same activity. Within the Septuagint Pentateuch the particles (discourse markers) appear sporadically and inconsistently in the translation, but this might itself be a sign of editing. In surviving documents from Egypt we find that, after the drafting of a text, particles are one of the most frequent additions made between the lines or in the margins. They are not added consistently throughout, especially as they were probably becoming rare in spoken Greek of the post-Classical period, but added at moments where the author felt emphasis was needed or a stock phrase required a particle. Such inconsistency in papyri might explain the inconsistency in the Septuagint, and therefore be an indicator of editorial work. As already noted, dictation was common in antiquity. This sheds light on the process of translation and might explain the physical layout of the translators’ “desk.” Since desks were not the norm, we cannot imagine someone reading a scroll while writing down a translation on another scroll. Scribes usually worked on their laps while sitting upon the floor, but could consult another scroll by having it on a stand nearby (Askin 2018). The

THE ORIGINS AND SOCIAL CONTEXT OF THE SEPTUAGINT 17

person reading the Hebrew would be able to translate as he reads it, allowing for a second person to write it down in Greek. This would mean at least two people would be involved, but not necessarily the three of van der Louw’s reconstruction. Translation would then be a process of reading and of writing down, two separate stages that would not require particular training as a translator on the part of the scribe. The skill of writing Greek would be sufficient. It is even not impossible that the person writing down the Greek is an Egyptian, given the wide employment of Egyptian scribes in the Ptolemaic bureaucracy, but on this we can only speculate. An older theory holds that one can reconstruct from the Septuagint traces that the translation was made from a transliteration of Hebrew into Greek characters (Wutz 1925–33). Wutz’s theory has not held sway in scholarship, even though some proper names and rare words are transliterated in the Septuagint. It does not prevent us from thinking of an intermediary translator of the Hebrew and the writer of the Greek version, while also indicating the potential for reconsideration of our assumptions.

RESEARCH METHODS As scholars have become more cautious over using ancient accounts on the origins of the Septuagint, greater emphasis in recent years has instead been placed upon philological evidence. In previous scholarship, reconstructions of the social context have been supported by attention to translation technique, but often circumscribed within a narrow framework. The focus was largely on questions of competency of the translators, whether in Hebrew or Greek (e.g., Tov 1999a). This is not to say that earlier studies did not pay attention to translation features as indicators of social context. The identification of exegetical renderings was often used as a clue for the social background of the translators (e.g., Seeligman 1948; Gerleman 1956). We might describe the turn to philology in recent years as a renewed quest for the social background of the translators. As already indicated, attempts to raise sociolinguistic questions about the Greek have now become common and conclusions about the identity of the translations are drawn from them. This is rooted in the recognition since Deissmann (1901) and his contemporaries that the papyri confirm that the language of the Septuagint is the everyday language of its time and not an artificial language or a peculiar dialect of Greek. For Deissmann this led to the conclusion that the translators were ordinary people rather than men of letters, a derivation from the fact that comparison with the Septuagint was made with documentary papyri rather than literary Greek. More recently, without assigning the translators to a social class, it has been shown through philological evidence that the translators had native competency in Greek (Janse 2002; Lee 2018). Translation technique is still a focus for some (e.g., Rajak 2009; Ziegert 2017) from which broad conclusions are derived of the translators’ intent from this method. Such a method can only take us a limited way as it must assume a general intent of the translators from a specific method. It offers only a broad general picture and requires modern presumptions about language translation to prevail in antiquity. A more fruitful exploration is to compare specifics of the language register or of the vocabulary choice to what we see in other ancient documents, especially from Egypt. Egyptian influence in the vocabulary has been identified (Görg 2001; cf. Aitken 2015c) to confirm that the translators did not come from Jerusalem as implied by Aristeas. The register of the Greek, which has frequently been classed as colloquial (Thackeray 1909: 21) or affected as a result of the translation process (Wright 2011), might appear to be inappropriate for a

18

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Ptolemaic king. While the method to use the vocabulary is an important application in determining clues for the background of translators, we once more need to be cautious of generalizations. The classification of registers in post-Classical Greek is still in its infancy, and texts that display a range of features cannot be classed easily into one category such as “street Greek” or “uneducated,” or be dismissed as writing just how they spoke. The Septuagint itself, in any one book, can be said to be an amalgam of features, using terms of both literary and nonliterary Greek, and while it displays a style that is unpretentious it can also reveal moments of literary flair (Aitken 2014). There is a need to avoid such blanket terms that have been applied since Deissmann (1901) until today, especially when no attempt is made to define them. The full range of features that comprise standard post-Classical Greek needs to be determined from a systematic analysis of documents of a sufficient size to demonstrate individual styles and norms (Evans 2010; cf. Dhont 2019). From this, substandard, popular, or lower register features can be identified and accounted for, placing the Septuagint within a diachronic framework. For a sociolinguistic description of the Greek it is necessary to engage with the many individual features before moving to general conclusions. A marker of higher education is the ability to vary style and expression according to the context or purpose of the document. The translation technique of regular equivalence and of adhering to the Hebrew word order, however, limits any opportunity to display such ability. It is, therefore, hazardous to draw strong conclusions on the translators’ social background. Only indicative markers can be found. It is notable that the translation seems to have been carefully thought out and as any written communication has been considered and composed with care. The same can be said for any of the letters in Ptolemaic Egypt, with which the Septuagint is compared. None of them straightforwardly contains spoken Greek, except where we can show for sure that they aim to present a conversational style. An example might be BGU 4.1079 (= CPJ 2.152, from 41 CE) where the words are given to pass on as an order. The spoken words are more identifiable by the staccato short sentences than by the register. While it may be difficult to determine from the Greek the precise sociolinguistic context, the philological method is a very important step in developing our understanding of the origins and social background of the Septuagint. Research should continue into post-Classical Greek and especially as attested in papyri and inscriptions. Comparison of individual translation elements in the Septuagint with what is found in the papyri helps us to locate the world in which they moved. For the Pentateuch, it is not only an Egyptian world, but one in which they are clearly learning in the same “bookish” circles as Egyptians. The translation method is that used by others at the time (Aitken 2016; cf. Bickerman 1959), and their vocabulary and stylistic features are consistent with those trained in Greeks scribal schools (Aitken 2014). Greater attention to comparison between translation features and the world of Ptolemaic Egypt may shed further light on this. Another recent development is the application of theory from translation studies. This has been seen in the so-called interlinear paradigm (Pietersma 2002), and in the writings especially of van der Louw and Ziegert, discussed above. Translation studies as a movement of the second half of the twentieth century certainly has informed much about the relation between the translation method and the genre of the text, the translator’s intention, and the target community (see van der Louw 2007). It has challenged us to think further about why the ancient translators chose their particular techniques, which indeed vary from the consistent method of the Pentateuch to the more literary style of Job and Proverbs and to the extreme precision of the Kaige group and its successors. There are some fundamental challenges, however, as to how far to apply modern theory

THE ORIGINS AND SOCIAL CONTEXT OF THE SEPTUAGINT 19

to ancient translations. In contemporary studies it is already known who the translator is, the purpose of the translation, and the community or target audience for which it is intended. From this, the translation process is then described and explained. For the Septuagint none of this is known and we must work in reverse from the character of the translation toward the translator and the translation’s purpose. We might know more if a detailed historical study of the Egyptian context and the role of Greek in ancient Egypt were studied in order to gain a picture of the translators first. However, translation study in Septuagint studies tends to work exclusively with the textual data without anything but a cursory glance at historical context. The two areas need to be brought together. Just as in historical sociolinguistics where the historical limitations and gulf in knowledge do not prevent some conclusions, historical translation studies have begun to achieve the same (Pym 1998). For Septuagint studies to integrate translation studies, attention to the data from other ancient translations and to the sociohistorical context of the Septuagint translators needs to be granted greater prominence in the discussion.

ONGOING RESEARCH QUESTIONS It remains true that we will never know the full reasons for the translation, and indeed, like many historical explanations of origins, there may be no one theory to account for the whole. Nonetheless, as has been made clear, our knowledge of the social context in Egypt and of writing and language use at the time is ever increasing. Therefore, new hypotheses can be advanced, while theories that do not undertake in-depth investigation of this material should be treated judiciously. Although language has been an area of study for some time, bringing in insights from sociolinguistic theory (see, e.g., Bentein 2013) will provide a better means of describing the Greek of the Septuagint. We have not yet reached an adequate description of the social status and level of Greek, and therefore no agreement has been attained on the social position and linguistic competence of the translators. The function of the translation is also dependent on how the technique is evaluated, on the degree to which the translation is seen as literary, and on how far the translators intended to write in the manner they did. All this can be appreciated only once we have an improved method of describing the Greek. Further comparison of translation technique with the papyri should also bring new evidence to bear on the questions. So far, such studies have largely focused on the nature of the Greek, but comparison can be made with word choice and indications of register in papyri. Marginal corrections in papyri will also give an insight into the working methods of the translators, indicating where they might have learned techniques for stylistically improving their Greek and used them in the translation (Aitken 2014: 131–2). As the publications from Egypt proceed apace, Septuagint scholars should also be learning more about the history and social context of Ptolemaic and early Roman Egypt. This can only help in our understanding for contextualizing the production of the Septuagint. One fundamental question still remains outstanding. Why translate a text in the first place? Within the history of religion translation has not always been a preferred option, and in some traditions it is considered anathema. If the Pentateuch was viewed as in some way religious in antiquity, what does translation mean for the understanding of the text and its status? While the translation legends such as Aristeas and that of Philo may not tell us about the origins themselves, they do tell us much about the reaction to this first translation. They are grappling with this fundamental question of the role of

20

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

a translation within a religious system and at the same time providing justifications for translation (see Matusova 2018). They therefore still perform a role in the investigation of the origins of the Septuagint. The story of origins continues too with the story of the revisions, which on the one hand suggest a desire to correct the earlier translation(s) and on the other support the notion of using translations. The appearance of translations after the Pentateuch, both of other books later becoming the canon and also of works not appearing in the canon (e.g., 1 Enoch), confirms the importance of translation in ancient Judaism, and afterward adopted by Christianity. Understanding the origins, then, is to appreciate a pivotal moment in the history of religions and to raise questions concerning the role of text and language in ancient religion.

RECOMMENDED RESOURCES 1. Aejmelaeus (2013). A proposal for some oral circulation of translations before the written version, visible especially in the agreed choice of lexical equivalents. 2. Aitken (2014c). A review of earlier attempts to relate the Septuagint to its social context, and a demonstration of the complexity of describing the language. The paper proposes that the translators are to be found among the administrative class of Egyptian scribes. 3. Bickerman (1959). A classic paper that has been neglected owing to its tendency to assert without providing supporting evidence, but the paper does propose that the translators were more professional translators of everyday documents than learned scribes. 4. Dorival (2010). A useful overview of the many theories on the origins of the Septuagint and a suggestion that papyri finds may help in providing the answers. 5. Van der Kooij (1998). A defense of van der Kooij’s position that the translators were learned scribes and would most likely have come from Jerusalem.

CHAPTER 2

Septuagint Translation Technique and Jewish Hellenistic Exegesis MARIEKE DHONT

INTRODUCTION The Septuagint, defined here as the Old Greek translation of the Hebrew Bible, offers a window into the world of the translators.1 The Greek text is the product of a translation process that comprises two components: the interpretation of the source text and its rendering in the target language. Sometimes, the result of this process is a translation in which the Greek reflects the Hebrew word by word. Deut. 5:42

‫פנים בפנים דבר יהוה עמכם בהר מתוך האש‬ The Lord spoke with you face to face on the mountain from the midst of the fire πρόσωπον κατὰ πρόσωπον ἐλάλησεν κύριος πρὸς ὑμᾶς ἐν τῷ ὄρει ἐκ μέσου τοῦ πυρός The Lord spoke with you face to face on the mountain from the midst of the fire

The target language allows for a precise matching of every element in the source text in such a way that the meaning of the source text and the target text is the same. Yet, differences between the Hebrew and the Greek texts may arise in the course of the translation process. The author of the prologue to Sirach already shows awareness of this reality when he writes οὐ γὰρ ἰσοδυναμεῖ αὐτὰ ἐν ἑαυτοῖς Ἑβραϊστὶ λεγόμενα καὶ ὅταν μεταχθῇ εἰς ἑτέραν γλῶσσαν (“For what was originally expressed in Hebrew does not have exactly the same sense when translated into another language”). Such differences between

I wish to thank Jim Aitken and the editors for their feedback and generosity in sharing their insights. At this point, I would like to make a note regarding citations from primary texts. For the Hebrew, I use BHS. Translations are cited from NRSV, at times revised. For the Greek, I use the editions in the Göttingen-series; in cases where this edition is not available for a specific book, recourse is taken to the edition of Rahlfs and Hanhart (2006). Translations are cited from NETS, at times revised. 1

2

22

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the Hebrew and the Greek may occur for various reasons.3 Since translation is a transition between languages, and no two languages are exactly the same in terms of grammatical, syntactic, and semantic structure, some deviations between the Hebrew and the Greek are the result of structural differences between the two languages. A straightforward example of this type of deviation is the rendering of the conjunctive wāw as either καί or δέ. The use of the latter equivalent leads to a change in word order vis-à-vis the Hebrew, since δέ occurs in the second position in a clause (Aejmelaeus 1982: 34–47). Another example pertains to the rendering of Hebrew idiom into Greek. To illustrate the latter, we may look at the expression of greetings. In Hebrew, the idiom “to greet someone” consists of a verb ‫“( ׁשאל‬to ask”) and the noun ‫“( ׁשלום‬peace”) introduced by a prepositional lamed. The person who is greeted is introduced also by a prepositional lamed. The Septuagint rendering of this idiom varies. In 1 Kgdms 10:4, we find καὶ ἐρωτήσουσίν σε τὰ εἰς εἰρήνην (“and they will ask you about matters regarding peace”). The translator has rendered the elements of the Hebrew idiom individually. The result is a Greek construction that reflects the Hebrew source text in a way that is not idiomatic in Greek. In such a case, scholarship speaks of interference.4 Alternatively, when a translator aims to express the meaning of the Hebrew in Greek idiom, we encounter renderings such as ἠρώτησεν δὲ αὐτούς Πῶς ἔχετε (“they asked them, ‘How do you do?’ ”) in Gen. 43:27 (Lee 1983: 25) or καὶ ἠσπάσαντο αὐτόν (“and they greeted him”) in Judg. 18:15. In essence, the meaning of the Hebrew and the Greek remains the same, but the translators show different attitudes regarding what they consider the most desirable rendering in Greek. The differences between the Hebrew and the Greek may go beyond the linguistic realm and have a sociocultural or theological dimension. A classic example in this regard is the Septuagint rendering of ‫“( מזבח‬altar”). In the Greek Pentateuch, we find two different equivalents: βωμός and θυσιαστήριον. The former is used consistently in reference to pagan altars, the latter for an altar of Israel. No aspect of the semantic range of these Greek words themselves makes one necessarily more appropriate for one type of altar or another. The consistent use of different Greek terms for the same Hebrew source word in function of the religious context suggests that the translators wanted to indicate a distinction between Jewish and pagan cultic practice through lexical choice (Van der Kooij 2003: 606–7). All of the above examples illustrate different ways in which the Septuagint translators could render their source text. The abilities and methods of a translator in rendering the Hebrew into Greek are commonly referred to as the “translation technique” (Sollamo 2016: 153–5). The crucial question this chapter seeks to address is, “How do we understand the relationship between translation technique on the one hand, and the theological and (by extension) sociocultural background of the translators on the other?” The answer to this question has often been sought specifically in those renderings that were thought to attest to exegesis on the part of the translator. In the field of Septuagint

For the sake of completeness, I add here that deviations between the Hebrew and the Greek can have other origins as well, such as variation between the source text of the translator and the Masoretic text, or a translator’s possible misreading of the Hebrew (see Tov 2012). In what follows, I focus on deviations that occur in the process of rendering the Hebrew into Greek and that can be ascribed to the translator. 4 Interference refers to linguistic features of the source text (mainly lexical and syntactic patterning) copied in the target text. Interference can either be “negative” (i.e., the pattern in the target text is non-normal in the target language) or “positive” (i.e., the pattern in the target text is normal in the target language but used disproportionally often, because it happens to resemble a feature in the source language) (Toury 2012: 311). 3

TRANSLATION TECHNIQUE AND JEWISH EXEGESIS 23

research, exegesis has generally not been understood as the process of interpretation of a source text, as in expressions like “biblical exegesis” or “Homeric exegesis.” Rather, it is seen as a translation-technical category that describes differences between the Hebrew and the Greek, particularly when the differences have a bearing on theological matters. In recent decades, however, the field of Septuagint studies has undergone two interrelated developments that warrant a new look at this question. First, the Greek translation of the Hebrew Bible is increasingly studied as a cultural artifact of Hellenistic Judaism, and not just as a witness to the biblical text. This has given rise to new questions and approaches in the field that move beyond a focus on what the Greek translation can tell us about the Hebrew source text. Secondly, Septuagint scholars have started to engage modern translation studies (e.g., Pietersma 2006c; Boyd-Taylor 2011). One of the main contributions of this engagement is that scholars have aptly come to think of the Septuagint translation as a complex, target-oriented phenomenon determined by sociocultural dynamics. The translators were engaged in their task within a context that provided them with a socioculturally determined awareness of the process and product of translation. This approach to translation implies that not just the deviations, but every aspect of the translation is considered to reflect the background of the translators, including the various ways in which the source text has been interpreted and rendered. Consequently, scholarship must reconsider what the Septuagint as a translation can tell us about the milieu and cultural (including theological) views of Jews in the Hellenistic era, and how to engage in this line of study.

VIEWS AND DEBATES Septuagint Translation Technique Research in Septuagint translation technique requires a specific methodology. Scholars operate with the consensus that the Septuagint is a collection of translations produced over a period of at least three centuries, by different translators, working alone or perhaps in groups. The translation technique is, therefore, studied by book, or by group of (presumably) related books, such as the Pentateuch (which is generally believed to have been translated first) or the Minor Prophets. Understanding a book’s translation technique requires awareness of key issues inherent to this field of research, including the text-critical questions related to Hebrew and the Greek texts of the book under investigation. When analyzing a deviation between the Hebrew and the Greek, all possible explanations for the deviation in question ought to be considered: a source text that potentially differed from the Masoretic text, changes that could have occurred in the transmission history of the Greek text, the unpointed state of the source text, the possibility of a graphic error or misreading on the part of the translator, and the translator’s working habits––that is, the translation technique (see Tov 2012). Traditionally, the translation technique of a book has been described along a single axis of literal to free. “Literal” is defined on the basis of set criteria: the seminal typology of James Barr was based on six criteria (Barr 1979: 294–323), whereas Emanuel Tov reformulated them as a set of five (Tov 2015a: 22–5): (1) Lexical consistency: the tendency to use standard word equivalents; (2) Segmentation: the tendency to divide Hebrew words into meaningful elements, each of which is represented by an individual Greek equivalent;

24

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

(3) Adherence to word order: the tendency to reflect the word order of the Hebrew in the Greek; (4) Quantitative representation: the tendency to represent each element in the Hebrew by an equivalent element in Greek; (5) Linguistic adequacy of lexical choices: Tov states that this last criterion is subjective and therefore cannot be used profitably to analyze translation units. “Freedom” has generally been understood as the relative absence of these features (see Lemmelijn 2001). Only rarely has it been understood positively: see van der Louw (2007) for a proposal aimed at a more systematic description of freedom. Tov’s category of “linguistic adequacy” hints toward a general trend in scholarly descriptions of the character of Septuagint translations: a “literal” translation is often associated with a high degree of interference from the Hebrew and, therefore, supposedly constitutes substandard Greek. Yet, scholarship has an implicit or explicit preference for literal translations, because of the presumed usefulness of literal renderings for textcritical studies. Theological bias may be involved as well: a “faithful” Jewish translator is presumably one who follows the form of the Hebrew text closely. “Free” translations, in turn, are often associated with “good” Greek usage but generally regarded negatively as a translation, in part also because of their (presumed) limited use for text-critical purposes. Recently, Septuagint scholarship has slowly been moving beyond the “literal” versus “free” divide, acknowledging that these categories are inherently flawed for various reasons. Modern translation scholars have emerged from this distinction fifty years ago, recognizing that it constituted an “imprecise and circular debate” (Munday 2008: 16). Their work (see particularly Toury 1995; rev. ed. 2012) is now being used by Septuagint scholars in order to conceptualize the translation process not as a singular choice between “literal” or “free” but, instead, as a complex, target-oriented phenomenon that is governed by socioculturally determined factors. The fact that translation is an intricate sociocultural activity implies that it cannot be described along a single binary axis, and that a translator’s decisions during the translation process are governed by a multidimensional interplay of various factors that are determined by the translator’s context.5 By way of example we may look at the first words of the Septuagint, ἐν ἀρχῇ, which render the Hebrew ‫בראׁשית‬. The latter is traditionally translated as an independent clause (“in the beginning”), even though the syntactic ambiguity of these opening words leaves room for other interpretations, such as a dependent temporal clause (“when God began”). The Greek equivalent, ἐν ἀρχῇ, is a so-called literal translation of ‫בראׁשית‬: ἐν (“in”) represents the prepositional bet, while ἀρχή (“beginning”) corresponds to ‫ראׁשית‬ (“beginning”). Both are standard equivalents. We can say more about this rendering, however. The phrase ἐν ἀρχῇ appears frequently in Greek sources. It can express “in the beginning” of a concrete time period or event, as in Thucydides (Hist. 1.35.5) and Plato (Phaedr. 67d12), or the notion “first” (as opposed to “later”), as in Isocrates (Archid. 103) and particularly in sources contemporary to the translation of the Septuagint, such as P.Cair.Zen. 1.59029 line 1 (258 BCE) and OGI 219 line 3 (197 or 279–274 BCE). With ἐν ἀρχῇ, the Septuagint translator has made a statement about the temporal nature of creation. Thus, the use of ἐν ἀρχῇ in Gen. 1:1 was not merely considered appropriate as a mimetic rendering of the Hebrew; it was also a common phrase in Greek literature that

We unfortunately know relatively little about the historical context of the Septuagint translators, and we are dependent on the only evidence we have: the Greek text itself.

5

TRANSLATION TECHNIQUE AND JEWISH EXEGESIS 25

adequately represented the Jewish translator’s interpretation of the source text. It retained some of the ambiguity inherent to the phrase in Hebrew, and is, as such, a theological statement. In other words, the notion of “literal” so commonly used in standard discussions of translation technique is not a sufficient explanation for such a rendering: translation involves a multidimensional decision-making process—partly conscious, partly unconscious—regarding how to interpret the source text and how to render it. This approach to describing translation phenomena is known as multicausality, a notion only recently introduced into Septuagint studies (Aitken 2015a). Recognizing the possibility of a wide range of motivations for any given rendering, whether traditionally described as “free” or “literal,” acknowledges the complexities of the translation process.

Exegesis in the Septuagint Some Septuagint renderings have been described as “exegetical.” In Septuagint studies, the term “exegesis” refers to the relation between the Greek text and the MT, specifically to cases where the Greek deviates from the Hebrew as a result of the translator’s intervention. Traditionally, three levels of exegesis have been distinguished, namely, lexical, contextual, and theological exegesis (Tov 2015a: 49–54). Lexical and contextual exegeses are interrelated and pertain to the linguistic identification of the source text (Tov 2015a: 50).6 Among his list of examples, Tov includes the following renderings: Exod. 32:34

‫אל אשר דברתי לך‬ To that about which I have spoken to you εἰς τὸν τόπον, ὃν εἶπά σοι To the place about which I have told you

The plus of τὸν τόπον in Greek, which explicates the antecedent of the relative clause, is seen as an addition that “served to improve [the Greek text’s] readability from a linguistic and contextual point of view” (Tov 2015a: 50). Josh. 4:14

‫ויראו אתו כאשר יראו את משה‬ And they revered him, as they had revered Moses καὶ ἐφοβοῦντο αὐτὸν ὥσπερ Μωυσῆν And they feared him like Moses

Tov (2015a: 49–50) is not entirely clear in his description of these categories. He writes that “every translation reflects linguistic exegesis … [which] involves the grammatical identification of all words in the source language as well as their semantic interpretation. On the other hand, not every translation contains additional forms of exegesis … Leaving aside the linguistic exegesis found in all translation units, we now turn to other forms of exegesis. In a way, all forms of exegesis might be called ‘contextual exegesis,’ because the translators’ concept of ‘context’ was wider than ours.” On the one hand, only linguistic exegesis is said to be common to all translation, as opposed to contextual and theological exegesis, while on the other hand, all exegesis is said to be contextual. In my view, words can only be interpreted semantically within a context, particularly in the case of homonyms, for example. Furthermore, Tov applies “exegesis” only to deviations in the Greek text. Yet, as an example of a translation that attests mainly to linguistic exegesis, he cites Aquila, whose translation is known to have been mimetic of the form of the Hebrew and contains only a limited number of deviations from the source text.

6

26

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

The translator is said to have “condensed … elements that they considered superfluous” (Tov 2015a: 52). Theological exegesis, in turn, pertains to deviations between the Hebrew and the Greek that are theologically motivated (Tov 1990b). Evidence of theological exegesis in the Septuagint is argued to be found in specific aspects of the translation technique (Tov 2015a: 49–54): (1) Lexical choices, either the consistent choice of a particular translation equivalent for one word or a distinctive choice that breaks the expected pattern; (2) The rewriting of a given verse in translation to reflect a contemporary understanding of it; (3) Pluses and minuses (of a few words or of extensive material). Many scholars use “exegesis” (without an adjective) in a strict sense, referring only to Tov’s category of theological exegesis.7 Sometimes, the label “exegesis” is reserved for those deviations that presumably are the result of a translator’s deliberate, systematic, and purposeful intervention (Pietersma 2006a: 45). Others have argued that it is complicated, if not impossible, to make a distinction between (unconscious) “theological influence” and “conscious theological exegesis” (Olofsson 2001: 143). In using the term “exegesis” in reference mainly to deviations between the Hebrew and the Greek, the Septuagint interpretation is measured against a modern, scholarly interpretation of the Hebrew, which is presumably the “correct” reading. Where the ancient and the modern interpretations of the Hebrew overlap, the Greek is seen as “proper translation”; where they differ, for reasons that can be ascribed to the translator, it has been regarded as “exegesis.” Research on interpretative methods in the Septuagint and the cultural background of the translators has thus taken deviations between the Hebrew and the Greek as its primary point of departure. By extension, interpretational tendencies and theological bias on the part of the translator are thought to be found mainly in “freely” translated books such as Isaiah and Job (Tov 1990b: 217; Jobes and Silva 2015: 330). A few considerations are in order at this point. First, what constitutes a deviation between the Hebrew and the Greek for modern scholarship may, for the Septuagint translator, have been an expression of what the source text was thought to mean, rather than an attempt to change its content. Second, even when according to our modern scholarly standards the Hebrew and the Greek texts correspond, the translation is the result of the translator’s interpretation and rendering of the source text. This process is governed by a translator’s sociocultural environment (Toury 1995). When it comes to interpreting the source text, for example, not just the translator’s education and working methods play a role, but also traditions of reading and interpreting the source text with which the translator may have been familiar. With regard to the question of how to render the source text, the varied translation profiles of different Septuagint books on the one hand and the revisionist activity of “the Three” on the other, for example, indicate that notions as to what acceptable translation looked like changed over time.8 These

Sometimes the term “interpretative translation” is then used to cover Tov’s categories of lexical and contextual exegesis. 8 Furthermore, the suggestion that the origins of the Kaige recension and the Old Greek translation of Job, for example, are both situated in the first century BCE indicates that different attitudes to translation coexisted. 7

TRANSLATION TECHNIQUE AND JEWISH EXEGESIS 27

observations imply that deviating and non-deviating renderings alike (in other words, all aspects of the translation technique) are the result of a complex process of interpretation and transmission into another language that has been governed by different socioculturally determined factors and that, in my opinion, can be meaningfully described only from the perspective of multicausality.

Septuagint Theology When the translation questions pertain specifically to issues related to God, such as the way in which God is addressed or represented, or related to religious beliefs, such as angelology, messianism, or eschatology, scholars have often spoken of the theology of the Septuagint translators (Tov 1987: 254; Seeligman 1990: 224). The degree to which the Septuagint reflects a Jewish theology distinct from that represented by the Hebrew Bible is a matter of debate. The scholarly field is traditionally divided into minimalists and maximalists (Pietersma 2006a: 35–6). Minimalists tend to argue that the theological differences between the Hebrew Bible and the Septuagint are limited (e.g., the work of the so-called Finnish school, represented by scholars such as Sollamo and Aejmelaeus). When trying to explain deviations between the Hebrew and the Greek texts, for example, they will ascribe a deviation to a theological development only when other explanations such as text-critical, translation-technical, linguistic, or contextual motivations cannot account for the difference. Maximalists, on the other hand, try to frame the transition from the Hebrew Bible to the Septuagint primarily within a framework of history of ideas (e.g., the works of Seeligman and Schaper). They generally favor theological development as the explanation for deviations between the Hebrew and the Greek. The question to what extent the Septuagint translation reflects theological bias on the part of the translators is thus inherently intertwined with matters of scholarly methodology (Palmer 2004; Glenny 2009).

RESEARCH METHODS AND THEORIES Studying the Septuagint translation as an expression of theological thought is a line of research located at the crossroads of the translator’s working methods and the linguistic, historical, and cultural environment in which the Septuagint originated. We must first say a few words on the linguistic environment. The Greek of the Septuagint needs to be framed within the context of post-Classical Greek. Through much of nineteenth- and early twentieth-century scholarship, the Septuagint was thought to be written in a unique Jewish dialect or sociolect of Greek, one that required the creation of neologisms or the semantic stretching of existing lexemes for theological purposes. Comparative linguistic analysis has shed much light on this issue, however. It has been convincingly demonstrated that the Pentateuch reflects standard contemporary postClassical Greek, and exceptions to this are much less common than what has previously been assumed (Lee 1983, 2018; Evans 2001; Aitken 2014b). Vocabulary choices can tell us about the target setting of the Septuagint or the social position of Jews in the Hellenistic world. A good example in this regard is the use of the term εὐείλατος (“merciful”) to characterize God, an epithet typical of the Greek Psalter. While long thought to be a distinctly biblical word, papyrological finds have shown that the word was, in fact, used in non-Jewish circles, in reference to both royal favor (e.g., P.Petr. 2.13 frg. 19, l. 3) and divine favor (e.g., P.Cair.Zen. 1.59034 l. 19). In

28

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the context of deities, εὐείλατος is generally used in reference to small cults rather than traditional Greek religion. Thus, these documentary sources show royal petitionary language being used of a god, the same way it is used in the Septuagint. This observation suggests that the use of εὐείλατος in the Septuagint was a natural choice for the translators (Aitken 2014a: 62–5). On the basis of this and similar examples from the Psalter, Aitken (2014a: 69) argues that vocabulary among emerging new ‘Greek’ cults, often eastern cults being expressed in Greek form, developed from classical terms, but the words chosen enabled the users to distinguish themselves from the Greek traditional religions. The Jewish adoption of these terms in the same manner as the other cults represents an element of self-awareness, of a shared experience with other religious groups in the face of the Olympian deities. After concisely considering the Greek used in the Septuagint, we may turn to aspects of translation technique. A first issue that has often been discussed is the potential theological background of Hebrew-Greek equivalences. It is commonly accepted that later translators were influenced by the word choices of the Pentateuch translators, but the question has often been posed as to what determined the lexical decisions of the first translators.9 The nature of divine names in Greek is a pertinent example (Tov, 2020a). As mentioned above, however, the main focus of research in theological trends in the Septuagint is on deviations between the Hebrew and the Greek. The question scholars seek to answer is what may have caused the deviation in question. The answers provided may depend on insights into other factors related to the translation. Merely focusing on deviations can lead to misleading conclusions if matters of language convention and social context are ignored or downplayed. Let us consider Job 11:4 by way of example. Job 11:4

‫ותאמר זך לקחי ובר הייתי בעיניך‬ For you say, “My conduct is pure and I am clean in your sight” μὴ γὰρ λέγε ὅτι Καθαρός εἰμι τοῖς ἔργοις, καὶ ἄμεμπτος ἐναντίον αὐτοῦ Do not say, “I am pure in my actions and blameless before him”

The Hebrew expression ‫“( בעין‬in the eyes of”) has been rendered using a simple preposition in Greek, ἐναντίον (“before”). Such a rendering represents the common prepositional function of ‫ בעין‬in Job as well as in other books, and reflects the grammaticalization of the expression in Hebrew (Sollamo 1979: 123–46). In the past, scholars have argued that this example in Job is indicative of the Greek translator’s tendency to eliminate Hebrew anthropomorphisms in reference to God (Gerleman 1946: 59). However, this exact equivalent is also found in other passages in Job, regardless of whether ‫ בעין‬in the Hebrew text is associated with God or man, such as 15:15; 18:3; 19:15; 25:5; and 32:1. As a standard equivalent regardless of the context, therefore, we cannot argue that the rendering of ‫ בעין‬as ἐναντίον should be taken to represent a theological tendency on the

Some scholars suggest the possibility that oral traditions preceded the written translation and that translators adopted already existing equivalences. The question, however, remains the same: what determined the lexical equivalence?

9

TRANSLATION TECHNIQUE AND JEWISH EXEGESIS 29

part of the translator. Rather, the lexical choice in question shows a concern for idiomatic usage of the Greek language as part of the translation technique (Orlinsky 1959: 164–5). A thorough and nuanced understanding of the working methods of a translator is thus crucial when trying to analyze and evaluate a rendering as interpretative or theological. A translator’s working habits may, however, appear varied and unsystematic. The book of Numbers provides us with an example. Num. 12:8

‫ותמנת יהוה יביט‬ And he beholds the form of the Lord καὶ τὴν δόξαν κυρίου εἶδεν And he has seen the glory of the Lord.

Where the Hebrew text speaks of seeing God’s ‫“( תמנה‬form”), the Greek uses δόξα (“glory”). Dorival (1994: 67) suggests that the expression in the Greek rendering of Num. 12:8 is the result of an intertextual reference to Exod. 24:17, where τὸ δὲ εἶδος τῆς δόξης κυρίου (“the appearance of the glory of the Lord”) renders MT ‫“( מראה כבוד יהוה‬the appearance of the glory of the Lord”). Intertextual references to other Septuagint translations occur regularly in Greek Numbers (Dorival 1994: 66–72).10 Often called “anaphoric translations,” they are an important theological aspect of Septuagint translations, as they tell us about the textual and interpretative framework of the translators. The deviation between the Hebrew and Greek texts of Num. 12:8 may well be the result of the translator’s wanting to avoid the anthropomorphism in the Hebrew, as has been argued by some (Fritsch 1943: 9; Tov 2015a: 53). After all, there appears to be a tendency to avoid the idea of “seeing God” in other passages in the Septuagint Pentateuch as well. This tendency is furthermore paralleled in other ancient scriptural translations, particularly the Peshitta and the Targums (Frankel 1851). Yet, the translator of Numbers also introduces the theme of seeing God in instances where it was not explicit in the Hebrew, such as in Num. 6:25. Num. 6:25

‫יאר יהוה פניו אליך‬ The Lord make his face to shine upon you ἐπιφάναι κύριος τὸ πρόσωπον αὐτοῦ ἐπὶ σέ May the Lord show his face to you

This verse shows that the translator did not oppose the idea of seeing God altogether (see Ruark 2020: 245–60). Inconsistency in translation method may at times seem to hamper our understanding of the translators’ motivations. Precisely at this point, the notion of multicausality can be of use to move the discussion forward and deepen our understanding of the translation process.

They are, in fact, a common aspect of many Septuagint books (e.g., Heater 1982 [Job]; Koenig 1982; Cook 2001 [Proverbs]; Van der Vorm-Croughs 2014: 299–454 [Isaiah]). This phenomenon may be significant from a literary and theological viewpoint and requires further investigation.

10

30

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

When it comes to weighing translation technique and ideology, we also need to take into account Jewish beliefs in the Second Temple period, and the likelihood that such beliefs were part of the cultural background of the Septuagint translators (Collins 2006: 129). The tendency exists to retroject the process of interpretation that was involved in the reception of the Septuagint onto the Septuagint as a witness to the interpretation of the Hebrew. This issue is particularly pertinent when it comes to theologically important concepts in the New Testament. New Testament usage of certain terms and concepts is often informed by their use in the Septuagint. As a result, the question arises whether or not the underlying theological idea would have been introduced into the Greek text already by the Septuagint translator. A good example of this is the use of ἐλπίς (“hope”) in the Greek Psalter and the concept of eschatological hope in the Septuagint and the New Testament. In the New Testament, ἐλπίς has taken on a specific theological meaning. Peter, in his Pentecost sermon (Acts 2:14–41), cites Greek Ps. 15:9 (MT 16:9) in the context of hope for resurrection. This has led to the hypothesis that the Septuagint version reflects the intentional assertion of an early, pre-Christian notion of eschatological hope (Schaper 1995). When we analyze the equivalent within the context of the translator’s working methods, however, we observe that in Ps. 15:9 (MT 16:9), ἐλπίς renders ‫“( בטח‬security”)—a common equivalent in the Greek Psalter. The choice of ἐλπίς thus does not imply the development of a specific eschatological view on the part of the translator (Cox 2001: 292; Boyd-Taylor 2006a: 17–22). When talking about the beliefs of Jews in the Hellenistic era, the question also arises whether the Septuagint reflects influence from Greek culture (Harl, Dorival, and Munnich 1988: 254–9; Jobes and Silva 2015: 343–6). A Septuagint translator may use words or motifs that carry specific connotations associated with Greek philosophical or pagan traditions. Does that mean that these connotations were intended by the translator when they used the word, and does its use thus reflect a theological development or a desire to recast the theology of the passage in question? For example, ‫“( נפׁש‬soul”) is generally rendered as ψυχή (“soul”) in the Septuagint. The latter may bring to the reader’s mind connotations related to a Platonic, dualist understanding of a mortal body and an immortal soul. Some have therefore argued that the Septuagint translator’s use of this equivalent is a theologically motivated decision that reflects a new understanding of the concept “soul” in Judaism (Rösel 1994: 61). However, the semantic field of ψυχή is not limited to its use in Platonist writings: ψυχή appears in pre-platonic Greek writings with a range of physical and biological connotations similar to those implied by ‫נפׁש‬ (Bratsiotis 1966; compare also Lys 1966). In closer proximity in terms of date and provenance to the composition of the Septuagint stand non-literary sources such as the letter P.Tebt. 1.56 (late second century BCE), which also attests to a non-metaphysical use of ψυχή (van der Meer 2016: 55–6). How, then, do we understand words or concepts in the Septuagint that have their roots in Greek culture? A frequently cited example in this regard is the use of Ἀμαλθείας κέρας for ‫ קרן הפוך‬as the name of one of Job’s daughters in Job 42:14. The incorporation of a reference with origins in Greek mythology is seen as evidence of the translator’s cultural “Hellenization” (Gerleman 1946: 38; Fernández Marcos 1994b: 257–8). One can, however, understand this rendering as linguistic data rather than as indicator of cultural change on the part of the Jews: it may have been the most suitable equivalent, perhaps even the only available expression in Greek, to render the translator’s

TRANSLATION TECHNIQUE AND JEWISH EXEGESIS 31

understanding of ‫ קרן הפוך‬as “the horn of plenty” (Dhont 2018: 128–9). The question one may ask when encountering such a rendering is whether the translator’s goal was to develop the theology of the Jewish scriptures in a new direction or to explicate his Jewish understanding of the sense of the source text in a new language. If, after due consideration of text-critical and translation-technical matters, certain renderings in the Septuagint appear to indicate evolving beliefs on the part of the translator, scholarship should consider the broader context of such beliefs and trace ideological development in other contemporary Jewish sources (Jobes and Silva 2015: 328). An exemplary article in this regard is John Collins’s treatment of the topic of messianism in the Septuagint Pentateuch (Collins 2006). Yet, the Septuagint as a whole, and any rendering in it, reflects the way in which Jews in the Hellenistic era understood their scriptures and how they thought that interpretation was best expressed in Greek, regardless of the translation approach. The rendering of the opening words of Gen. 1:1 discussed above illustrates this point. From this follows that even when there is no clear evidence of theologically relevant differences between the Hebrew and the Greek, the Septuagint is a significant source for our understanding of Judaism in the Hellenistic era. Every rendering is an expression of Jewish theology that can tell us more about the way in which the Jews interpreted their source text and about the cultural and theological background of the Septuagint translators. Continuity can be as telling as change, particularly when looking at translation as a multicausal phenomenon.

ONGOING RESEARCH QUESTIONS The Septuagint offers many opportunities to enhance our knowledge of the translators’ backgrounds and theological viewpoints. In this regard, scholarship can derive useful insights from a variety of areas in the humanities, while offering significant contributions to these fields in turn as well. First, our understanding of the Septuagint begins with an understanding of the Greek language used. The linguistic context of the Greek of the Septuagint is post-Classical Greek. Research in this area has focused on the Pentateuch. Less work has been done on the language of other books in the Septuagint corpus or the revisions; this needs to be explored further. Considering “the question of the appropriate context from which parallels are drawn that are credited with evidentiary value for determining the meaning of a Greek translation” (van der Meer 2016: 51), it will prove beneficial to bring in more evidence from documentary and sub-literary papyri as well as inscriptions contemporary to the translation of the Septuagint. A broad and diachronic understanding of the Greek language, with a focus on post-Classical Greek, is indispensable for this line of research. Besides vocabulary, other aspects of the language of the Septuagint that are awaiting further investigation include the balance between post-Classical Greek idiom and source language interference, as well as the interaction between style and translation technique, between vocabulary choice and register, and between register and genre. Second, further research is needed on the translation technique beyond the traditional “literal”/“free” divide. The incorporation of Gideon Toury’s framework of Descriptive Translation Studies (Toury [1995] 2012) into Septuagint studies has significantly advanced the field by conceptualizing the translation process as a target-oriented phenomenon

32

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

governed by sociocultural dynamics. From this derives the insight that so-called literal translation reflects sociocultural dynamics as much as so-called free translation does. The majority of studies into the translators’ background thus far, however, have focused on deviations between the Hebrew and the Greek, and by consequence on books whose translation has traditionally been described as “free.” As such, there is a vast amount of research yet to be done in this area, in considering every aspect of the translation and by studying different books regardless of their translation character. In addition, Septuagint studies is invited to engage modern translation studies beyond Toury’s Descriptive Translation Studies.11 The notion of multicausality may help to move discussions in this area forward: while Septuagint renderings have traditionally been explained on singular grounds, a translation choice tends to be the result of the interplay of numerous factors. Appreciating the complexity of the translation process will lead to a more nuanced understanding of the sociocultural and theological context in which the translators worked. After all, the Septuagint translators were not working in a vacuum. Their context pertains both to the way in which they interpreted the Hebrew and to the way in which they rendered it into Greek. More work needs be done in order to understand the position of the Septuagint in Jewish traditions of interpretation, translation, and literary activity. Particularly open for further investigation are the theological aspects of the translation technique, such as harmonization, anaphoric translations, and the Kaige approach. Finally, the position of the Septuagint within the history of Judaism also needs further consideration. Recently, scholarship has moved toward an understanding of the Septuagint translation as a cultural phenomenon. As a result, more attention is being paid to the translators and their background, but that background has yet to be understood in all its complexities. The historical context of the Septuagint is often presented as obvious or well understood, but we have yet to answer many questions, and we need to allow the historical evidence at our disposal to speak for itself. Contextualizing the Septuagint requires a combination of the study of the translation technique and historical analysis. Scholarly understanding has been heavily influenced by the presumption that Hellenism and Judaism were cultural forces in opposition, with Hellenism as the dominant force. The fact that Jews both adopted the Greek language and subsequently translated their scriptures into it are often seen as “Hellenization,” but to what extent did this part of the intercultural process influence the Jewish worldview and cultural understanding? Contemporary scholarship is moving toward a more nuanced understanding of cultural dynamics in the Hellenistic era and the developments of Second Temple Judaism. The Septuagint is the main locus at our disposal for understanding these dynamics and developments; yet, historical studies on Hellenistic Judaism hardly consider the Septuagint in this regard. With Septuagint scholars increasingly studying the Septuagint as a Hellenistic Jewish document, it is part of our scholarly task to demonstrate its place

The use of DTS in the field of Septuagint studies has shown that we stand to gain from deriving insights from modern translation theories. Septuagint scholarship has yet to engage views and theories from modern translation studies beyond DTS (for an example, see Dhont 2018). In doing so, scholars will have to take into account that the field of translation studies is occupied mainly with modern translation. Scholars of modern translations generally know about variables that Septuagint scholars can only hypothesize, such as the identity of the translator, the purpose of the translation, and often even the translation approach. This means that when Septuagint scholars adopt frameworks from modern translation studies, we must be willing to adapt them to our evidence and our purposes when and where necessary. In that process, it can be useful not to focus solely on the work of one main proponent of an approach but also consider how their work and views have been received by fellow translation theorists.

11

TRANSLATION TECHNIQUE AND JEWISH EXEGESIS 33

within the history of Judaism. In this way, the broader fields of biblical and Jewish studies will be able to appreciate the significance of the Septuagint and incorporate this body of literature into the evidence used to understand Hellenistic Judaism.

RECOMMENDED RESOURCES 1. Jobes and Silva (2015). A good introductory volume to the Septuagint. The authors discuss the question of theological development in Judaism during the Hellenistic age (326–51). 2. Barr (1961). A seminal work in the field of biblical studies that questions the presumed existence of a link between language and culture or mindset. The author emphasizes the importance of linguistic understanding of any biblical text and translation. 3. Lee (2018). The most recent and most comprehensive work on the language of the Pentateuch. The author shows that many features previously thought to be indicative of interference or supposedly “poor” Greek are, in fact, natural and standard in post-Classical Greek. 4. Aejmelaeus (2007a). A collection of essays that provide a comprehensive introduction to the study of the Septuagint translation technique. 5. Aitken (2015a). The author introduces the notion of multicausality to Septuagint studies to explain the use of what has traditionally been seen as one of the most “literalistic” translation features in the Septuagint, namely καί γε. 6. Tov (1990b). An often-cited article that has informed much of traditional Septuagint scholarship. It needs to be read critically within the broader framework of Tov’s approach to the Septuagint. 7. Ausloos and Lemmelijn (2020). This volume represents the most recent and systematic attempt at presenting the state of the art of research into the theology of the Septuagint with contributions by various Septuagint scholars.

34

PART T WO

Language

36

CHAPTER 3

Septuagint Transcriptions and Phonology PETE MYERS

INTRODUCTION Transcriptions are found throughout the Greek scriptures and have implications for numerous fields. These include the phonology of Greek, Hebrew, and Aramaic, the reconstruction of the history and original forms of the Greek and Semitic texts, the translators’ attitudes to their Vorlagen, the existence and nature of any oral tradition that may have been received by the translators, the dating of Septuagint books, and exegesis. This chapter will focus on the use of transcriptions for reconstructing historical phonology.1 Terminology regarding transcriptions is inconsistent in the secondary literature, and there is no agreed set of definitions. It is important to strive for precise and neutral terminology, but with a pragmatic realism that a history of usage already exists for most terms.

TRANSCRIPTION Transcription is the act of writing a source language entity in a script associated with a target language. I use the term entity, rather than lexeme, as transcriptions sometimes comprise more than simply a lexeme; that is, they may include inflectional morphemes and/or more than one lexeme. For example:2 (1) 1 Kgs 4:9 (2) Deut. 2:21   𝔐 ‫ ּוב֣ית ָ ׁ֑שמֶ ׁש‬ ֵ 𝔐 ‫ָּכ ֲענ ִ ָ֑קים‬   βαιθϲαμυϲ OG ενακειμ OG

This chapter makes use of numerous sigla (e.g., ‹ ›, [], ~, etc.). See the Abbreviations and Glossary sections in the present volume for further details. In places of ambiguity, the representation of a feature is preceded by an abbreviation for the language from which it is cited (e.g., BH = Biblical Hebrew, PS = Proto-Semitic), which again can be found in the Abbreviations Table. 2 Given that the shape of lunate sigma and lack of distinction between medial and final form sigma are the source of a considerable number of textual variations in transcriptions, and that sigma is written as lunate in the majority of significant manuscripts for transcriptions, I use lunate sigma throughout this chapter. OG throughout this chapter designates my own assessment of the most likely original text. 1

38

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Transcription as defined here is thus a broad category that encompasses any word that is (a) not a target language word (in our case not Greek) and (b) written down. Tov (1999b) makes many helpful, astute critiques of previous terminology to discuss loanwords, homophonous translations, transcriptions, and the general category that includes these phenomena. However, he uses the terms transliteration and transcription as synonyms, which should be distinguished. He also distinguishes the terms transcription and loanword, which should be treated as effectively synonymous.

TRANSLITERATION Transliteration is the act of transcribing the written form of a source language entity. The term has frequently been misused as a synonym of transcription, but Knobloch (1995: 98) rightly distinguishes transliteration as a specifically graphic process. A wellknown example of transliteration is the use of the graphs «πιπι» in some Hexaplaric manuscripts to approximate the shape of the divine name ‹‫ ›יהוה‬in Aramaic script (Metzger 1991: 35). Tov’s (1999b: 166) treatment of transcription and transliteration as synonyms renders his definition of transliteration a tautology: “Transliteration, that is, the transcription into Greek characters of Hebrew and Aramaic words.”

LOANWORDS A loanword is a word that is copied from one language to another.3 This definition means that transcriptions are simply written loanwords. In contrast, it has been usual practice in Septuagint studies to distinguish loanwords from transcriptions. Thackeray (1909: 32) distinguished loanwords from transcriptions on the basis that the former are inflected. This distinction cannot be consistently maintained; for example, indeclinable toponyms such as ιερουϲαλημ are no less loanwords than declinable toponyms such as βαβυλωνοϲ. Tov (1999b: 165–6) distinguishes these categories by defining loanwords as words that were “accepted into the Greek language … in the time preceding the translation.” This definition is vague and impractical: What does “accepted” mean? How does one determine whether a word was “accepted” prior to the translation of any given book? The only information available to modern scholars is whether a word is attested in a written work that is still extant today, but the majority of ancient works are no longer extant. Words may have existed as loanwords in spoken Greek for a long time prior to ever being written down. For example, ϲαββατοϲ is not (currently) attested in written form prior to the translation of the Pentateuch, but it surely must have been in use among Greek-speaking Jews. Conversely, we know of other words that are previously attested in written form and were probably known to a translator, but nevertheless are transcribed afresh. In the following example, whether 1 or 2 Esdras was created first, it is likely the earlier text was known to the creator of the later text:

See Haspelmath (2009) for an explanation of why the verb copied is preferable to the widespread term borrowed or loaned. Despite this, the term loanword is so well established that coining a term such as copiedword would be unnecessarily confusing.

3

SEPTUAGINT TRANSCRIPTIONS AND PHONOLOGY 39

(3) Ezra 8:7 𝔐 ‫עֲתַ לְ יָ ֑ה‬ γοθολιου 1 Esdras OG *αθελεια 2 Esdras OG If distinguishing loanwords from transcriptions is to have any practical use for the task of historical phonology, it is because one would assume that the Greek spelling of a loanword was supposedly “fixed” and therefore did not necessarily reflect the sound of the word in Hebrew. In the above example, however, the spellings did reflect the sound of the word in Hebrew. The anthroponym μω(υ)ϲηϲ “Moses” is another more prevalent example that loanwords did not necessarily have a fixed spelling, and that distinguishing loanwords from transcriptions lacks practical utility for historical phonology. The name Moses with the spelling μωυϲηϲ was well established in Jewish Greek literature prior to the translation of 2 Esdras, and often declined. According to the definitions of both Thackeray and Tov, it is therefore a loanword, not a transcription. In Hebrew it is consistently spelled the same, whereas in Greek its spelling changed over time. While in the Old Greek the root is spelled μωυϲη-, in the Lucianic tradition it is spelled μωϲη-. This change in Greek spelling may reflect a change in the Hebrew pronunciation of ‫משה‬. Barr (1989: 34–6) observed that ‫ מֹ ׁשֶ ה‬was one of a number of words that are consistently spelled defectively in the Hebrew Bible. His explanation for this spelling consistency was that “it had simply been agreed by convention, in most or all circles, that a particular spelling should be followed” (Barr 1989: 197). Alternatively, I contend that this consistent defective spelling probably reflected a feature of the name’s pronunciation. Gignac (1976: 186–7) lists several Egyptian personal names that are attested in Greek with the (nonGreek) diphthong ωυ, the spelling used in the Septuagint for Moses. A simpler explanation therefore, not considered by Barr, is that ‫ משה‬was always written defectively because the first vowel was a non-Hebrew (i.e., Egyptian) sound, which was different from the sounds normally represented by the grapheme ‹‫›ו‬. Sometime after the orthography of the Hebrew Bible had become fixed, this non-Hebrew vowel sound was lost in the reading tradition and replaced with Hebrew /o/. So, when the Tiberian Masoretes later developed their written notation to record the oral reading tradition they received, they pointed the initial vowel in the name Moses with holem: ‫מֹ ׁשֶ ה‬. This change in the Hebrew pronunciation of Moses is reflected in the change of the Greek spelling μωυϲη → μωϲη. There are other possible explanations for this change in the Greek spelling of Moses, but the key fact is that its spelling does change over time, despite being a loanword according to both Thackeray’s and Tov’s definitions. The above demonstrates that there is at least a case to consider that this spelling change results from a change in the Hebrew pronunciation of Moses. Classifying μω(υ)ϲη- as a loanword as opposed to a transcription has no practical bearing on the value of μω(υ)ϲη- for historical phonology.

HOMOPHONOUS TRANSLATIONS Homophonous translations are translations that sound similar to the source language entity. Technically speaking, because such translations rarely, if ever, sound identical, a more accurate term would be homoeophonous. But the term homophonous is well established in Septuagint studies and so is retained here.

40

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Tov (1999b: 166) defines homophonous translation as “the choice of Greek equivalents which resemble the sound of their Hebrew-Aramaic counterparts but differ in meaning” (emphasis added). This extra qualification is because homophony within a language occurs only when two lexical entities with different semantics sound identical, as in British English /naɪt/ that could be night or knight. This requirement within a language exists because if two entities sound identical, then in order to be distinct entities they must differ in meaning. This requirement does not exist for entities that sound similar or identical between two different languages, as the mere fact that the entities exist within different language systems means they are already distinct. For example, English much and Spanish mucho sound similar and have similar semantics (Campbell and Mixco 2007: 29), yet as lexical entities in different language systems they can be distinguished and are therefore not the same word (and, incidentally, are neither cognates nor related by areal diffusion). If English much were used to translate Spanish mucho, this would be a valid case of homophonous translation, pace Tov. A further reason not to restrict the definition of homophonous translation only to entities that differ in meaning is that to do so excludes many translations that have been considered homophonous within Septuagint studies, such as: (2) Exod. 30:28 𝔐 ‫הָ עֹ לָ ֖ה‬ ὁλοκαυτωμάτων OG “burnt offering(s)” The practical value of distinguishing homophonous translations from transcriptions for historical phonology is the different ways their spellings are restricted in their ability to reflect the translators’ pronunciation of the source languages. Unlike loanwords/ transcriptions, the spelling of homophonous translations is fixed, as they are entities that exist in the target language. The spelling of homophonous translations is therefore less able to reflect the sound of the source language entity to which they correspond. Distinguishing transcriptions from homophonous translations is therefore useful for historical phonology.

VIEWS AND DEBATES Transcription studies have seen a relative lack of interaction between scholars working in the field. Some contributions have apparently gone completely unnoticed (e.g., Krašovec 2010). The following non-comprehensive selection of views and debates are chosen to illustrate the sorts of issues that need to be considered, in order to encourage new scholars entering the field to interact with previous work.

JAMES BARR Barr (1990: 31) claimed that the spelling of names in Greek is “notoriously wild and inaccurate.” This statement was made in defense of his view that the Septuagint translators were essentially unfamiliar with a reading tradition for their consonantal Hebrew Vorlagen. His explanation of those places where the Septuagint apparently corresponds with the later Tiberian vocalization is nuanced, but in essence is that the Greek translators knew

SEPTUAGINT TRANSCRIPTIONS AND PHONOLOGY 41

the Hebrew language and so made sensible guesses as to the vocalization of the text. Barr goes on to reference Swete (1900: 322) as a precursor to his position. If he so wished, Barr could have drawn on a number of previous scholars who also judged the vocalization of names in the Septuagint as chaotic to different degrees. For example, Frankel (1841: 123–4) found in Septuagint transcriptions evidence of variation of vocalization between Septuagint books, while Staples (1927: 6) claimed they evidenced living diversity in the pronunciation of Hebrew vowels, which contrasts with what he considered to be the Tiberian Masoretes’ imposition of their supposedly “artificial” system to preserve the sound of spoken Hebrew.

EINAR BRØNNO Barr did not engage with Brønno’s (1940) important contribution to this problem. Brønno observed that much confusion arises in the analysis of phonology if the possibilities for the morphology of transcriptions are not taken into account. Relying upon Codex Vaticanus (B) as his source of data, he filtered forms that he judged had likely become corrupted in Greek textual transmission.4 On this basis, Brønno found a direct correlation between B and the Tiberian Masoretic text for the following noun patterns: ‫ּכָתָ ב‬-χαθαβ, ‫ּכָתֵ ב‬-χαθηβ, ‫ּכָתֹוב‬-χαθωβ, ‫ּכ ִָתיב‬-χαθειβ & -χαθιβ, ‫ּכָתּוב‬-χαθουβ Crucially, Brønno observed that noun patterns were not necessarily interpreted with the same distribution in the Tiberian tradition as they were by the Greek translators. If a text ‫ כתב‬was pointed by the Tiberians as ‫ּכָתָ ב‬, but rendered in the Septuagint as χαθηβ, then it is not the case that a vowel equivalent to Tiberian qameṣ has been transcribed with both ‹α› and ‹η›; rather, this consonantal text has been interpreted as the reflex of a *qatal pattern noun by the Tiberians, but as the reflex of a *qatil pattern noun by a Greek translator. According to Brønno, therefore, noun patterns within a set in the Tiberian tradition (e.g., ‫ ּכֵתָ ב‬,‫ ּכֹותֵ ב‬,‫ ּכֹ תֶ ב‬,‫ ּכֵתֶ ב‬,‫ )ּכֶתֶ ב‬always correlate with a noun pattern from within a corresponding set in B (e.g., χαθεβ, χεθεβ, χοθοβ/χοθεβ, χωθηβ, χηθαβ), even if they are distributed differently within the text. He developed this argument further in Brønno (1943). Brønno claimed that exceptions to these spelling patterns were due to corruption in either the Tiberian tradition or, more usually, B (Brønno 1940: 211). Brønno’s work contributed several very important observations to the field, which have often gone unnoticed by subsequent scholars. First, morphology matters. Septuagint transcriptions of names do indeed appear notoriously wild and inaccurate if one immediately attempts a morphologically blind phoneme-by-phoneme comparison with the Tiberian Masoretic vocalization. Second, diachrony matters. Brønno hints that nouns in B equivalent to segolates in the Tiberian tradition are probably variable because these words were transitioning from their earlier, and varied, monosyllabic structures to the stable segolate patterns in Tiberian. Third, just as all written text traditions exhibit variety and plurality, so do oral text traditions. Diverse reading traditions exist today among Jewish communities. If a Greek translation often agrees with the Tiberian tradition, but sometimes differs, this is not evidence that the translator was not following a reading tradition, pace Barr (1990: 23–4), but merely that, if they

All Greek MSS in this chapter are referred to by their siglum in Rahlfs (2012).

4

42

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

were following a reading tradition, it was not identical to that recorded by the Tiberian Masoretes. One critique that can be levelled at Brønno’s method is its potential for circular reasoning. Even Brønno himself acknowledged the existence of some spelling variety within noun patterns (e.g., χαθειβ/χαθιβ and χοθοβ/χοθεβ), which raises the possibility that genuine diversity in the representation of Hebrew vowels might be overinterpreted as morphological difference. All working hypotheses must be logically falsifiable, and there must be some controls to avoid explaining away any potentially difficult data as speculative or rare morphological types. A simple control is to ensure that all morphological interpretations fit a plausible account of linguistic development. For example, in the case of segolates, Brønno is not unreasonable to suggest that multiple spellings may have existed in the Greek tradition, for example, (BH) /*katb/ → (SepH.) /*katɛb/, /*ketab/, etc. → (Tib.) /*kɛtɛb/.

JOSHUA BLAU It is difficult to overestimate the importance of Blau’s (1982) study of transcriptions in the Septuagint equivalent to ‹‫ ›ח‬and ‹‫›ע‬. This article may be the single most significant work concerning Septuagint transcriptions ever published. Prior to Blau, transcriptions equivalent to ‹‫ ›ח‬and ‹‫ ›ע‬were hotly debated, as these graphemes sometimes correspond to ‹χ› or ‹γ› in the Septuagint and sometimes not. Part of the solution is that these Hebrew graphemes both corresponded to two phonemes (‹‫›ח‬ ~ /ḥ, *ḫ/; ‹‫ ~ ›ע‬/ʿ, *ǵ/), but even knowing this does not explain the distribution of spellings with ‹χ› or ‹γ›. Wevers (1970) was the first to propose that Septuagint transcriptions reflect the gradual loss of /*ǵ/ from the Hebrew language over time, but Blau supplied the robust dataset and convincing narrative of their development that conclusively demonstrated this explanation to be the case for both /*ḫ/ and /*ǵ/. Blau’s attention to detail even extended so far as to compare the distribution of ‹χ, γ› spellings corresponding to ‹‫ח‬, ‫›ע‬, with the statistical distribution of the reflexes of Proto-Semitic /*ḫ, *ǵ/ against ProtoSemitic /*ḥ, *ʿ/ in other Semitic languages. Blau’s observations provide a phonological basis for identifying the order in which books of the Septuagint were translated (Genesis, then the rest of the Pentateuch, then most of the Bible, and finally Ezra-Nehemiah). His work thus demonstrates the practical impact that historical phonology can have on other fields.

‹ει› AND ‹ι› For a long time, there was consensus that the distribution of spellings with ‹ει› and ‹ι› in Greek manuscripts from the fourth century CE and later was meaningless. Many believed these orthographies were interchangeable for much of the history of Greek transmission, and that Byzantine scribes had a stylistic penchant for replacing ‹ι› with ‹ει› (Wevers 1974b: 209). In contrast, Williams (2018) has recently challenged this consensus, arguing that the distribution of ‹ει› and ‹ι› in the NT text of B correlates with the expected distribution of stress. In 2 Esdras, Myers (2019) identified a statistically significant correlation between features in the Tiberian vocalization, such as single /y/ : geminate /yy/ and ‹ι› : ‹ει›. There is a need for analysis of further datasets, and for the methods and conclusions of research that is challenging the consensus to be tested.

SEPTUAGINT TRANSCRIPTIONS AND PHONOLOGY 43

FRICATIVE ALLOPHONES OF /b, p, d, t, g, k/ (THE “BEGADKEPHAT” CONSONANTS) 5 In the Tiberian tradition, the stops /b, p, d, t, g, k/ were fricativized, that is, pronounced as [v, f, ð, θ, ʁ, χ] after vowels. Kahle (1959: 102–10) claimed these fricative realizations were an innovation by the Tiberian Masoretes, and cited in support of this claim a Tannaitic instruction for reciting the Shema, transcriptions in the Hexapla, and Jerome’s discussion of the ‫ פ‬in ‫ אַ ּפ ְַדנּו‬in Dan. 11:45. Garbini (1960: 24) followed Kahle’s lead on the basis of Hebrew transcriptions in Akkadian, Greek, and Latin. Kutscher (1965: 24, 27) definitively refuted this view by arguing, among other points, that Aramaic /b, p, d, t, g, k/ are always transcribed the same way in Greek script, just as in Hebrew. He notes the contradiction that while Garbini and Kahle cited this fact as evidence for the lack of fricative allophones in Hebrew, they acknowledged the existence of fricative allophones in Aramaic. The Greek data have been interpreted in various ways (Gray 1936; Torczyner 1937; Lisowsky 1940: 121; Murtonen 1981–2: 69; Waldman 1989: 85), and some have argued that Greek transcriptions cannot provide reliable data to resolve this issue (Speiser 1926: 380–1; Barr 1967a: 9–11; Moscati 1980: 27, §8.10).

DOES THE DISTRIBUTION OF ‹α› AND ‹ε› PROVIDE EVIDENCE FOR OR AGAINST PHILIPPI’S LAW? The supposed “law” of Philippi pertains to sound change where the Proto-Semitic vowel /*i/ becomes /*a/ in closed, stressed syllables, but is “notorious for having as many exceptions as examples” (Huehnergard 2013: 71a). Suggestions for when the law was operative have ranged from the Proto-Semitic period to the time of the Tiberian Masoretes. In contrast, it is often the case in Tiberian that /a/ has attenuated to /i/ in closed, unstressed syllables (Koller 2013). The usual reflexes of Proto-Semitic /*i/ are ṣere in Tiberian and ‹ε, η› in Greek transcriptions, and of Proto-Semitic /*a/ are pataḥ in Tiberian and ‹α› in Greek transcriptions. Qimron (1986b) has proposed that Philippi’s law was not operative in the Hebrew reflected by transcriptions in Origen’s Hexapla on the basis of the distribution of ‹α, ε, η›. This claim was opposed by Ben-Ḥayyim (1988), partly by arguing that Greek ‹ε› could at times represent an open [a] vowel sound. His article prompted a later rejoinder by Qimron (1991). Qimron (2006) has since published an English language article within which he lays out evidence for his position. At present, few scholars outside of Israel have engaged in this discussion.

RESEARCH METHODS Transcriptions have received little theoretical attention, such that the only “methods” that exist are the approaches taken by disparate studies. This discussion will concentrate on the use of Septuagint transcriptions for reconstructing the historical phonology of Hebrew and Aramaic, a task that involves four steps: collation of a dataset, restoration of the best

/b, p, d, t, g, k/reflects the order of these consonants at their point of articulation. In contrast, /b, g, d, k, p, t/ reflects the order of these consonants in the traditional sequence of the Hebrew abjad.

5

44

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Greek text, morphological and etymological analysis, and finally phonological analysis. Many of the premises that lie behind these procedures emerge from the views and debates previously discussed. Each step will outline the task that needs to be performed, key theoretical ideas and practical points, and finally an essential knowledge base required to perform the task successfully.

Collate the Dataset The Task  Identify all the transcriptions within a given corpus, and all the spelling variations attested in every manuscript. Issues of Theory and Practice  Any gathered data must be consistent. It is not the case that the spellings of transcriptions need be consistent, but rather that there should be consistency of sources and text-critical method. For example, a study covering the scope of the entire Septuagint should not use the main text of the Göttingen editions, as the fascicles are each produced by different editors with varying approaches to the text of transcriptions. In contrast, the text of Rahlfs (1935), or the updated edition by Rahlfs and Hanhart (2006), is appropriate for a study with such a broad scope. Most transcription spellings in Rahlfs’s text are based on B, with sensible editorial corrections in light of how a spelling in B was the result of plausible development in Greek textual transmission. While the main text of the Göttingen volumes is often assumed to be superior to that of Rahlfs’s single volume text, this is often not the case for transcription spellings. Example: (5)

Neh. 7:7 𝔐 ‫ַ ֽרע ְַמ ָ֨יה‬ δαεμια S (Rahlfs and Hanhart 2006) ναεμια B ρεελμα A (Hanhart 1993)

The above example compares the Rahlfs and Hanhart (2006) single volume text against the main text of Hanhart’s Göttingen edition (1993). In this case, the RahlfsHanhart text is superior. S can be explained by a Vorlage read as ‫*דעמיה‬, whereas A is probably a harmonization to ρεελια in the parallel list at Ezra 2:2, partially conflated with the beginning of the following word μαρδοχαιοϲ. When performing broad surveys of transcription spellings across the Septuagint, Rahlfs’s text is therefore probably adequate. While reliance upon Rahlfs’s text is practical for drawing broad and generalized conclusions, a study that explores transcriptions in any more depth will require a more laborious data gathering process. At the time of writing, transcriptions are not tagged in any Bible software or electronic dataset. It is possible to identify most of the transcriptions in a given corpus by searching for noun classes that are almost always transcribed (i.e., proper nouns). The only way to gather all transcriptions in any corpus exhaustively is by reading the Hebrew and Greek in parallel. The procedure I undertook to gather the dataset for my doctoral thesis (2019) on 2 Esdras was a word-by-word comparison of a photographic facsimile of 𝔐 against the main text and apparatus of the Göttingen volumes. I then subsequently checked the readings in the majuscules using photographic and online facsimiles, and spot checked for

SEPTUAGINT TRANSCRIPTIONS AND PHONOLOGY 45

the minuscules where online facsimiles were available. Aside from physically travelling to every relevant holding library, this method allowed the greatest depth in the gathering of data. A factor specific to my dataset was that Hanhart (1993) had exhaustively collated all spelling variants in his apparatus. Not all Göttingen editors have done this, but nevertheless the apparatus in those volumes is designed for this sort of task. When gathering data in this way, be aware that there may be transcriptions in the apparatus that have not been included in the main text. An exhaustive collation requires reading the Hebrew/Aramaic in parallel with the main text and apparatus simultaneously. The basic rule of thumb with the Göttingen apparatus is that, if a manuscript is not mentioned in the apparatus, then it agrees with the main text. Minuses are only marked at the beginning of the point where the text is not attested. The only way to ensure one knows what every manuscript attests at any given point is to read the entire apparatus from the very beginning of the book. Not all Septuagint books have Göttingen volumes, and in these cases the best alternative is the apparatus in Brooke and McLean (1906–40). Information on how to obtain images of manuscripts is given below. Essential Knowledge  Aside from searches in bespoke Bible software described above, more advanced electronic searches can be performed by obtaining datasets in spreadsheet or database formats, enabling analysis using more generic software. These searches can allow one very quickly to gather the data needed to answer some specific research questions. One source of datasets with liberal copyright requirements are the Tyndale House STEP Project datasets made available on github.6 With these datasets, many more complex searches that can be difficult or impossible to perform with Bible software are easy to perform using a spreadsheet program. More complex searches can be undertaken by importing the datasets into a database and using a query language.7 Bibliographic data for all manuscripts consulted for each Göttingen volume are given in the introduction for each volume. An exhaustive list of manuscript witnesses to the Septuagint is available for free online from the Göttingen Academy of Sciences (Rahlfs 2012). The publication includes a column listing the contents of each manuscript. Images of many, but not all, Septuagint manuscripts are available online, and this list is growing. Physical facsimiles have also been published for the major early majuscule codices. Codex Vaticanus can be viewed on the Vatican website (digi.vatlib. it), but there is also an excellent (and extremely costly) reconstructed facsimile edition available (Canart, Bogaert, and Pisano 1999). Codex Sinaiticus has a dedicated website (codexsinaiticus.org) and is also available in high-quality large photographic facsimile (2010). In a recent publication, Myers (2018) exhaustively collated all the corrections to 2 Esdras that include a proper noun—which in the case of these corrections means all the transcriptions—and made the readings available in tables. Codex Alexandrinus is the least accessible majuscule codex, as at the time of writing the Septuagint portion is not

See github.com/tyndale/STEPBible-Data. The most common database query language is SQL, which is sometimes non-intuitive and can be difficult to learn and debug. For advanced work on linguistic datasets, I have found the scripting language Python to be extremely powerful and straightforward. It can be installed freely on any computer system, and aside from the system’s native features there are many extra freely accessible modules written specially for the purpose of linguistic analysis. Python 2 will require some workarounds to ensure that it always handles data formatted as Unicode. Python 3 works natively in Unicode, but at the point of writing there are still many non-native modules not ported to Python 3.

6 7

46

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

available online, and the facsimiles are old and not widely available (Thompson 1883; Kenyon 1957). Other online MS collections not already mentioned that may be of help to the reader include: 1. mss.bmlonline.it (Florence) 2. bl.uk/manuscripts (London) 3. ambrosiana.comperio.it/manoscritti (Milan) 4. digitalcollections.universiteitleiden.nl (Leiden) 5. gallica.bnf.fr (Paris) 6. nlr.ru/eng/opac (St Petersburg) 7. onb.ac.at/en/digital-library-catalogues (Vienna)

Restore the Text The Task  Choose a spelling for every transcription, whether attested in an extant manuscript or a conjectural reconstruction, that is most likely to be closest to the spelling as written by the translator. Issues of Theory and Practice  Three premises help determine how to restore the best text of transcriptions from the extant data. The first premise is that the best text will explain all variants. Since the original texts of Septuagint books are no longer extant, any study that relies upon transcription spellings must work from hypothetically restored texts. What is meant by a “restored” text is a spelling that is identified as having the best chance of reflecting what the original translator actually wrote. A restored text might be selected from an extant reading or be a conjectural reconstruction (i.e., unattested in any extant manuscript). There are two perspectives one can take toward this task: What is the translator most likely to have written? and How are the extant readings most likely to have come about? Both are valid and necessary, but the second question should carry more weight. We have far more evidence about the history of copying texts than we do about the history of creating them, so we can be far less certain about the motives and mechanisms of the original translators. In contrast to this assertion, when discussing the spelling of transcriptions, some text critics have placed more emphasis on the first perspective than the second. For spelling variation of proper nouns in Greek Exodus, Wevers (1992: 207) maintained that a “translator would hardly mix up his spellings in such an arbitrary fashion; only scribes would introduce variation.” Knobloch (1995: 5) claimed that not just Wevers, but all the Göttingen editors (at his time of writing), relied on the Tiberian text as a guide to identify the best text for transcriptions, and that “[i]mplicit in [this] focus on the Hebrew form and on paleography is some idea of what the expected manner of transcription into Greek is.” He even states that “[a]rguments appealing to regular transcription equivalents seem to have been, in some cases, the deciding factor in the choice of the critical text” (Knobloch 1995: 8, emphasis added). Knobloch (1995: 5) therefore attempted to establish a translator’s “transcription practice” and use this as the basis for identifying the best text for transcriptions in the Pentateuch.

SEPTUAGINT TRANSCRIPTIONS AND PHONOLOGY 47

There are a number of problems with placing this much emphasis on how one would expect a translator to spell transcriptions and on assuming their spellings would be consistent. First, the methods by which translators created transcriptions are unknown to us. A transcription can be produced without the translator making a conscious decision about how to map from the source to target language. (For this reason, when referring to this process, it is also preferable to speak of the mechanism of transcription, which does not imply a conscious reflection on the task.) Second, translators may have been inconsistent in how they transcribed words. What may appear to be an inconsistency at first glance to a modern scholar may sometimes reflect a deeper feature of the source language. For example, when ‫ בית‬is in construct, the vowel is usually transcribed with ‹η› by most Septuagint translators. Sometimes, however, the vowel is represented by ‹αι›. Could this reflect secondary stress, intoned reading of the text, and/or slowed speech? Third, scribes frequently harmonize spellings within texts. This fact is a well-known textcritical reality (Colwell 1969: 112). If so, then imposing consistency is more likely to favor younger spellings over older ones. This last point leads on to the second question posed earlier: How are the extant readings most likely to have come about? Spelling changes in the transmission of texts occur in mostly predictable ways. Jongkind (2013) is a good example of a study of scribal habits for one manuscript in the text tradition of the Greek Bible. The best proposed original text will require the simplest path of textual development to explain all extant readings, given known or demonstrable tendencies of textual change. Some practical points are worth mentioning: (a) Scribes are more likely to spell words in ways that are easier to write and/or say. Therefore, more complex spellings are more likely to be original than simple spellings, longer spellings than shorter spellings, and spellings with vowels in hiatus than spellings with consonants breaking vowel hiatus, and so on. (b) Scribes are more likely to misspell a word in a way that makes it look or sound more familiar. Spellings are therefore likely to become harmonized in transmission, spellings different from common Greek features are more likely to be original than spellings similar to common Greek features, and if some manuscripts attest a transcription with Greek inflection and others not, this is more likely to have been added in transmission than lost, and so on. (c) Since scribes are likely to make similar sorts of errors to one another, and younger MSS are more likely to have survived to the modern day than older manuscripts, then spellings with majority attestation will often not be the best readings. The second text-critical premise for transcriptions is that the best text will imply a plausible Vorlage. Every possible spelling for a Septuagint transcription implies a consonantal Vorlage. The best implied Vorlage will require the simplest explanation in light of extant Hebrew/Aramaic texts (mostly L). Example: (6) Ezra 10:24   𝔐 ‫ ו ֶ ָ֖טלֶם‬ ‫ **וטלמן‬VOG   τελλημ S A L OG → τελμην a b10 119 (Hanhart 1993) Gen. 43:26; Lev. 23:17; Ezr. 8:18; Job 33:21.

10

48

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

τελλημ implies a Vorlage with the same consonantal text as 𝔐, but a different vocalization. τελμην implies a Vorlage with a different consonantal text than 𝔐. One of the Greek variants developed from the other by metathesis and graphical confusion of ‹ν› : ‹λ›, but there is no reason from Greek grammar to prefer one direction of this change over the other. Since τελλημ requires a simpler explanation for the text of its Vorlage, it is therefore more likely to be the original spelling (pace Hanhart in the above example). The third premise is that the best text will have an etymological and morphological explanation. Etymology and morphology are the most important factors when assessing spellings from the perspective of what a translator is most likely to have written. Two assumptions are appropriate here: (a) The Septuagint translators knew at least some Hebrew/Aramaic. (b) They at least tried to render words in ways that sounded Hebrew/Aramaic. On these grounds, transcription spellings that have absolutely no basis whatsoever in etymology or Semitic grammar are very likely not to be original. For example: (7) Neh. 7:58 ‫𝔐 יַעְ לָ ֥א‬ * ιεηλα OG →

ιεαλη a 119 (Hanhart 1993)

All other manuscripts (aside from those witnessing to L) attest a reading that has likely developed from ιεαλη. The text suggested for OG at this point is not attested here in any manuscript, but is attested at Ezra 2:56 (Ezra 2 and Nehemiah 7 are parallel lists). This situation means there are two attested spellings for this lexeme: ιεηλα and ιεαλη (and variants from the latter form that need not concern us here). These spellings could reflect two different vocalizations. Alternatively, OG originally attested the same spelling in both places, but one changed by metathesis in Greek transmission. To adjudicate between these possibilities, it is important to note that the text adopted here by Hanhart, ιεαλη, makes no sense as a Hebrew or Aramaic form. There is no explanation for the final vowel. In contrast, final ‹α› consistently corresponds to final ‹‫ ›א‬or ‹‫ ›ה‬in other words where the vowel /*ā/ would be expected. In Tiberian, a common reflex of *qatil nouns with a suffix is -‫( קְ טֵ ל‬Sagarin 1987: 18–19). The spelling ιεηλ- is consistent with such a vocalization. It is therefore likely that ιεαλη is the result of metathesis in Greek transmission. Essential Knowledge  Text criticism requires a working knowledge of the Greek manuscript tradition for the corpus being analyzed. An ability to weigh non-Greek evidence from translations can also be beneficial, but is not essential, and exciting areas of research such as the Ethiopic and Syriac versions should not be ventured upon until a good familiarity with the Greek tradition is obtained. While geared toward the NT, a good starting point to understand the writing and transmission of the Greek Bible as a whole is Metzger (1991). This can then be supplemented by the essays in McKendrick and O’Sullivan (2003), which are more oriented toward the Septuagint. Prior to the invention of the codex around the second century CE, the “Septuagint” or “Greek Bible” did not exist as a cohesive entity. Rather, individual books were written and transmitted on scrolls, and so each Septuagint book has a unique textual history. The quickest way to become familiar with scholarly discussion on the textual history

SEPTUAGINT TRANSCRIPTIONS AND PHONOLOGY 49

of independent books is by consulting Aitken (2015b) and Kreuzer (2019). Discussion of more depth may be found in the relevant volumes of La Bible d’Alexandrie series, and more advanced discussion again in that of the Mitteilungen des SeptuagintaUnternehmens series, but at the time of writing volumes in these series do not yet exist for every Septuagint book.8 A final area with which it is necessary to familiarize oneself is paleography. Tov (2012) is a very fine introduction to text criticism of the Hebrew Bible. He discusses graphemes and ligatures of Aramaic square script that are confused in the Qumran material (see 228–9, 231–2). Tov’s volume on Septuagint text criticism (2015a) is generally geared toward issues inherent to translation, and so is less well suited to our purposes. For Greek paleography, Thompson (1912), while dated, is still the most helpful introduction available. His tables of scripts for majuscule (144–7) and minuscule (191–4) across different centuries are extremely helpful.

Etymological and Morphological Analysis The Task  Develop a judgment on the etymological and morphological structure of every transcription, as the translator likely understood it, and the relevant diachronic and synchronic contexts of the non-Greek word. Issues of Theory and Practice  When considering the spelling of transcriptions alone, there are a number of variable factors that can make their interpretation unclear. For example, ‹α› could transcribe a vowel equivalent to Tiberian pataḥ or qameṣ. Alternatively, it could represent an original /*a/ vowel that has attenuated to /i/ in Tiberian, or be the result of an original /*i/ vowel that developed to ṣere in Tiberian, but shifted to /*a/ in the Hebrew of the translator by Philippi’s law. As discussed above, Brønno (1940) showed that many phonemes in the Septuagint had been misidentified by previous researchers who assumed that vowels in Septuagint transcriptions corresponded to their Tiberian equivalents, when in fact these words had been vocalized according to a different noun pattern. One example of such misidentification is the name of the sect ϲαδδουκαιοϲ (“Sadducee”). There has been much debate over the source of this Greek spelling (e.g., Derenbourg 1867; Manson 1938; Meyer 1971; Schürer 1979; BDAG, ad loc. Σαδδουκαῖος). A popular explanation is that it derives from ‫( זָדֹוְך‬zādôk; “Zadok”), which would associate the group with the priestly class and idea of religious reform (cf. 1 Kings 1–2). This interpretation fits the cultural situation in the Hasmonean dynasty, during which the Sadducees emerged. It does not, however, make sense of the double ‹δδ› or second vowel ‹ου›. Morphology provides a more straightforward explanation. In the postexilic period, an increasingly popular method of hypocorism was to use a qattūl noun pattern in place of a sentence name. For our example, ‫( צִ ְדקִ ּיָה‬ṣidqiyyâ; “Zedekiah”) would be shortened to ‫( צַ ּדּוְך‬ṣaddûk; “Zadduk”), a form attested in pointed manuscripts of the Mishnah. According to this explanation, the reflex of Hebrew geminate /dd/ is represented by ‹δδ› and of Hebrew long /uː/ by ‹ου›, which are entirely as would be expected.

On these series see the chapter by Jones in the present volume.

8

50

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Etymological and morphological analysis helps filter out incorrect associations between Hebrew/Aramaic phonemes and Greek graphemes. It is therefore a necessary task before drawing phonological conclusions. There are three steps: (a) Develop an initial set of working assumptions for the representation of the reflexes of historic Hebrew/Aramaic vowels in the Greek corpus being considered. The “reflex” of a vowel is a “descendant” of a vowel at a later period, which could potentially be very different in quality to the “ancestor.” The section below on Greek vowels will demonstrate why the list of possibilities may need to be adjusted depending on likely date of composition and other factors. One generic set of possibilities for vowels that can serve as an initial guide is given in Table 3.1. (b) Identify all possible etymological and morphological explanations for every form. Most of this task can be described as “diachronic” and is focused on identifying the historic non-Greek language forms that may lie behind the transcriptions. For Hebrew/Aramaic words, the most significant factor for this step is identifying the noun pattern. Sentence names often follow a verb pattern, but frequently with changes that make the researcher’s life harder. For example, D stirps (for this term see below) sentence names (Piel in Hebrew, paʿel in Aramaic) not infrequently lose the doubling of the second radical consonant, and sometimes the prefix of the participle form (e.g., in Tiberian ‫ ׁשֶ ל ְֶמיָהּו‬is attested, but never ‫)** ְמׁשַ ּל ְֵמיָהּו‬. 9 This process does therefore require detective work. The next section discusses various sources of information to help make the task less daunting. (c) Once enough data have been gathered, where possible, narrow down the most likely etymological and morphological interpretations for the forms in the dataset. Essential Knowledge  Hebrew and Aramaic words can be analyzed as comprising a root, upon which is applied a pattern, and then to which affixes may be added. In Semitic and Greek linguistics, the same terms, such as root or stem, are used with different meanings. Fox’s (2003) grammar of Semitic noun patterns uses terminology that is quite technical, but which therefore avoids these potential confusions. His definitions are used here. The Semitic root is “the sequence of consonants that stay constant in a set of verbs and nouns with meanings in some semantic field,” and the pattern is “the element of a Semitic lexeme that is neither a root nor an analyzable sufformative” (Fox 2003: 37). The stem is that part of a word (Hebrew or Greek) onto which affixes can be added. The Semitic Table 3.1 Reflexes of Biblical Hebrew vowels in Tiberian and Greek transcriptions Biblical Period

Possible representation in Greek

Tiberian Period

/*a/

‹α›

/a, ɔ/

/*i/

‹η, ε›

/e/

/*u/

‹ο, ω, ου, α, ε›

/o, u, ɔ/ or šewa

/*aː/

‹ω, ο, α›

/o, ɔ/

/*iː/

‹ι, ει›

/i/

/*uː/

‹ο, ου›

/u/

∅

‹α, ε› (or vowel harmony)

/ɛ/ or šewa

On the term “stirps” see below.

9

SEPTUAGINT TRANSCRIPTIONS AND PHONOLOGY 51

pattern comprises a template, which is “the arrangement of consonants and vowels” and melody, which is “a sequence of vowels” (Fox 2003: 41). For example: (8) For the name ‫“( נָתָ ן‬Nathan”):  

root: √ntn pattern: CaCaC template: CVCVC melody: /a…a/   

For verbs, sets of patterns are used to form full paradigms within stirpes (singular stirps; e.g., Qal, Piel, Niphal). The term stirps is preferred to “stem” to prevent confusion with the meaning of stem as defined above, and preferred to binyan as this latter term only denotes the Hebrew stirpes (Fox 2003: 34). The stirps is defined by Diakonoff (1988: 104) as a “derivative lexico-grammatical formation” that “evolves a full paradigm of verbal forms” by making use of a root. A simple language-neutral way to refer to the different Proto-Semitic patterns is to use qtl as the root letters (e.g., *qatal, *qatil, *qal, etc.). A Proto-Semitic noun pattern is a historic morpheme, of which there are different reflexes in the different Semitic languages. For example, Proto-Semitic *qatil becomes qɔ̄tēl in Tiberian Hebrew, but qtēl or qtīl in Tiberian Aramaic. A translator may still transcribe a vowel sound between the first two consonants of words with this pattern even if the word was vocalized as Aramaic (which would therefore be a phonetic transcription of a non-phonemic sound), but variation between ‹(ε)ι› : ‹η› in transcriptions of the vowel between the second two consonants may be a telltale sign that a translator knew Aramaic and this was interfering in their pronunciation of the Hebrew text. The etymological task involves first making use of dictionaries, then studies on specific word categories, and finally by consulting grammars. At present, the best dictionary for etymological analysis is the single volume edition of Donner (2013). For words in the Hebrew Bible that are not Hebrew or Aramaic, see Donner’s bibliography. Koehler and Baumgartner (HALOT) is the best English language dictionary for this purpose, but it is not hard to learn enough German just to consult Donner’s dictionary. Koehler and Baumgartner frequently do not provide etymological information on their entries, so learning to read Donner will save time in the long run. For studies specific to anthroponyms, see Noth (1966) and Zadok (1988). The most recent work is Rechenmacher (2012). For toponyms see Borée (1930). On Proto-Semitic noun patterns, see Fox (2003). For other features not covered in his work, consult Moscati (1980). Bauer, Leander, and Kahle (1922) is very helpful, as they attempt to describe the historical background to every Hebrew morpheme. Geoffrey Khan is currently leading a team at the University of Cambridge to produce a fully updated grammar of Hebrew, with a similar historical perspective, but will also incorporate the breadth of Hebrew traditions. A quick reference for Tiberian noun patterns is Sagarin (1987). The section on noun patterns in Joüon and Muraoka (2006) is very accessible. For Aramaic morphology, see Bauer and Leander (1927), Beyer (1986), Nöldeke (1904), and Rosenthal (2006).

Phonological Analysis The Task  Distinguish every phone and phoneme that is represented in every transcription, and through comparison with relevant Hebrew/Aramaic traditions draw

52

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

conclusions about the phonological system of the translator’s Hebrew/Aramaic and the sounds this system possessed. Issues of Theory and Practice  At this point, the reader has a consistently sourced dataset of transcriptions with spellings they trust that include logical guesses as to where the Vorlage might differ and plausible etymological and morphological explanations for most of them. Despite this preparation, phonological analysis still requires adopting some further working assumptions. The main practical issue at this stage is to minimize these assumptions and make them explicit. Essential Knowledge  Phonetic terminology can appear obtuse, but is a precise and efficient way to talk about the production of sounds in the human mouth. For an introduction see the Handbook of the International Phonetic Association (1999). The consonant systems of Aramaic and Hebrew from the Pre-Septuagint to Medieval periods can be described using the inventory of thirty-five phones given in Table 3.2. Pre-Septuagint Hebrew possessed twenty-five consonant phonemes /ʾ, b, g, d, h, w, z, ḥ, *ḫ, ṭ, y, k, l, m, n, s, ʿ, *ǵ, p, ṣ, q, r, ś, š, t/ and twenty-two graphemes to represent them ‹‫ ת‬,‫ ש‬,‫ ר‬,‫ ק‬,‫ צ‬,‫ פ‬,‫ ע‬,‫ ס‬,‫ נ‬,‫ מ‬,‫ ל‬,‫ כ‬,‫ י‬,‫ ט‬,‫ ח‬,‫ ז‬,‫ ו‬,‫ ה‬,‫ ד‬,‫ ג‬,‫ ב‬,‫( ›א‬Yuditsky 2013). Tiberian Hebrew possessed twenty-two consonant phonemes as a result of the mergers /*ḫ → ḥ/, /*ǵ → ʿ/, /ś → s/ (Blau 1982; Khan 2012: 93). The following describes the likely pronunciation of these phonemes when the Pentateuch was translated into Greek with a summary of the key developments into the medieval period. The consonant systems of Aramaic spoken in Egypt and Palestine are presented where they may contrast with or influence the development of Hebrew: ʾAleph /ʾ/ ‹‫›א‬, glottal plosive [ʔ]. In many words the phoneme has acquiesced. In 𝔐, ‹‫ ›א‬is marked with daghesh in four words to prevent readers failing to pronounce the guttural.10 Spelling variation in the DSS evidences a weakening of the gutturals (Qimron 1986a: 25–6, §200.11), and Samaritan Hebrew manuscripts reveal that the gutturals in that dialect had merged no later than the fourth century CE, when Aramaic had become the Samaritan vernacular (Ben-Ḥayyim 2000: 38, §1.1.8). Some cases of the replacement of ‹‫ ›א‬by ‹‫ ›י‬in Egyptian Aramaic could be interpreted as assimilation of /ʾ/ to the following vowel (Muraoka and Porten 1998: 24, §3o). Bet /b/ ‹‫›ב‬, voiced bilabial plosive [b]. In Tiberian, when written without a daghesh then a voiced labiodental fricative [v]. Thanks to transcriptions in Akkadian, the development of the fricative allophones of Aram. /b, p, d, t, g, k/ can be ascribed to the sixth century BCE (Kutscher, Ben-Ḥayyim, Dotan, and Sarfatti 1977: 116). The spread of this feature to Hebrew cannot be identified with the same certainty, but many scholars believe it occurred under Aramaic influence in the fifth century BCE (Rendsburg 2013: 104b). Once the phonetic rule was operative, /b, p, d, t, g, k/ fricativized when post-vocalic, but remained plosive in other environments. Sometime before the Tiberian period, this process stopped being productive in all but word-initial position, such that fricatives did not revert to plosives in cases where the preceding vowel is lost (Blau 2010: 79–81, §3.3.2.2). Evidence of fricative Mishnaic Hebrew /b/ can be found in the use of ‹‫ ›ב‬to transcribe the second element of Greek ‹ευ› (Segal 1927: 31, §48). Conversely, possible correspondences between ‹‫ ›ב‬and diphthongal ‹υ› in Greek transcriptions require careful assessment on text critical grounds, and may have potential for future research. Interchanges of /b/ : /p/ occur between some words that are attested in the Bible and the Mishnah (Segal 1927: 31, §49). In Samaritan Hebrew “the difference between voiced and unvoiced ‫ ב‬and ‫ פ‬vanished entirely” (Ben-Ḥayyim 2000: 34, §1.1.5). Additionally,

Lateral Approximant

Approximant

Lateral Fricative

Fricative

Trill

Nasal

Plosive

p pʰ

m

b

Bilabial

f

v

Labiodental

Table 3.2 Hebrew and Aramaic Consonants

θ ð

r

d

Dental

ɬ

s s̙

t̙ t, tʰ

l

z z̙

r̙̺

n

Alveolar

ʃ

Postalveolar

j

ɟ

Palatal

x

k kʰ

w

ɣ

g

Velar

χ

q

ʁ̞

ʁ̖

ʀ

Uvular

ħ

ʕ

Pharyngeal

h

ʔ

Glottal

54

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

/b/ → /p/ before voiceless consonants is attested in Egyptian Aramaic (Muraoka and Porten 1998: 17, §3g). Gimel /g/ ‹‫›ג‬, voiced velar plosive [g]. In Tiberian without daghesh, a voiced uvular fricative [ʁ]. /g/ and /k/ never developed fricative pronunciations in Samaritan Hebrew (Ben-Ḥayyim 2000: 34, §1.1.5). If, as suggested above, the fricativization of /b, p, d, t, g, k/ was a result of Aramaic influence during the exile, then it is not surprising that the process only partially spread to Samaritan. Some words containing /k/ in the Bible are attested with /g/ in the Mishnah (Segal 1927: 29, §45). Dalet /d/ ‹‫›ד‬, voiced post-dental plosive [d]. In Tiberian without daghesh, a voiced post-dental fricative [ð]. Spelling interchanges of ‹‫ ‹ט‬: ›‫ ›ד‬and ‹‫ ‹ז‬: ›‫ ›ד‬in manuscripts of the Mishnah demonstrate the existence of both allophones in Mishnaic Hebrew (Segal 1927: 30, §47). The distinction between [voiced] and [voiceless] was maintained most strongly in Samaritan Hebrew between /d/ and /t/ (Ben-Ḥayyim 2000: 34, §1.1.5). Contiguous dental consonants assimilate (e.g., /dt/ → /t/) in Egyptian Aramaic (Muraoka and Porten 1998: 13, §3b). He /h/ ‹‫›ה‬, glottal fricative [h]. Vav /w/ ‹‫›ו‬, labio-velar approximant [w]. In Tiberian, the default pronunciation was a voiced labio-dental fricative [v], and [w] was an allophone restricted to the immediate environment of /u/. The [v] pronunciation is evidenced in Mishnaic Hebrew, where ‹‫ ›ו‬is more common word-initially than in Biblical Hebrew, is sometimes confused with ‹‫›ב‬, and is sometimes used to transcribe Latin /v/ (Segal 1927: 34, §55). In Samaritan Hebrew, /w/ was a labio-velar approximant [w], a voiced labio-dental fricative [v], and a voiced bilabial plosive [b]. All three pronunciations are said to have been present in the word ‫ווויהם‬, but this is not reflected in modern pronunciation (Ben-Ḥayyim 2000: 33, §1.1.4.1). In the Dead Sea Scrolls, /w/ and /y/ sometimes appear as glides between two vowels in crasis after the elision of a guttural (Qimron 1986a: 26, §200.133). Zayin /z/ ‹‫›ז‬, voiced alveolar fricative [z]. Ḥet /ḥ/ ‹‫›ח‬, voiceless pharyngeal fricative [ħ]. In Samaritan Hebrew and Samaritan Aramaic, /ḥ/ → /ʿ/ preceded the merger /ʿ/ → /ʾ/ (Ben-Ḥayyim 2000: 41–2, §1.1.8.2). *Ḫet /*ḫ/ ‹‫›ח‬, voiceless uvular fricative [χ]. Blau (1982) established the fact and significance of the mergers /*ḫ/ → /ḥ/ and /*ǵ/ → /ʿ/, which he claimed were completed by the time of the translation of 2 Esdras (second century CE). Some words in the Mishnah are spelled with ‹‫›כ‬, which in the Bible are spelled with ‹‫›ח‬, e.g., ַ‫לִ כְ ּלְך = לִ חְ לּוח‬. According to Segal (1927: 28, §43) this fact “shows that there was also a hard pronunciation of the ‫ח‬.” They could, however, simply reflect an alternative spelling or pronunciation that emerged earlier than the time of the Mishnah (i.e., due to a hardening of some instances of /*ḫ/ → /k/). Ṭet /ṭ/ ‹‫›ט‬, emphatic voiceless alveolar plosive [t̙]. The emphatics were distinguished by retraction of the tongue root (Blau 2010: 68, §2.7). Yod /y/ ‹‫›י‬, palatal unrounded approximant [j]. In Tiberian with daghesh, a voiced palatal stop [ɟ]. The writing of ‹‫ ›א‬for ‹‫ ›י‬in the Mishnah evidences a weakening of /y/ (Segal 1927: 35, §56). Kaf /k/ ‹‫כ‬, ‫›ך‬, voiceless aspirated velar plosive [kʰ]. In Tiberian without daghesh, a voiceless uvular fricative [χ]. Most evidence seems to suggest that /k, p, t/ had come to be always aspirated in Hebrew and Aramaic when not fricativized at some time before the translation of the Pentateuch. In contrast to this general picture, transcriptions in the Zenon papyri from around this time do attest ‹κ, π, τ› for Aramaic /k, p, t/, suggesting the existence of an unaspirated allophone.

SEPTUAGINT TRANSCRIPTIONS AND PHONOLOGY 55

Lamed /l/ ‹‫›ל‬, voiced alveolar lateral approximant [l]. Mem /m/ ‹‫ ם‬,‫›מ‬, voiced bilabial nasal [m]. Nun /n/ ‹‫ ן‬,‫›נ‬, voiced alveolar nasal [n]. Samek /s/ ‹‫›ס‬, voiceless alveolar fricative [s]. ʿAyin /ʿ/ ‹‫›ע‬, voiced pharyngeal fricative [ʕ]. *Ǵayin /*ǵ/ ‹‫›ע‬, voiced velar fricative [ɣ]. Pe /p/ ‹‫ ף‬,‫›פ‬, voiceless aspirated bilabial plosive [pʰ]. In Tiberian without daghesh, a voiceless labiodental fricative [χ]. Ṣade /ṣ/ ‹‫ ץ‬,‫›צ‬, voiceless emphatic alveolar fricative [s̙]. Qof /q/ ‹‫›ק‬, voiceless uvular or postvelar unaspirated plosive [q]. Resh /r/ ‹‫›ר‬, in Proto-Semitic likely a voiced dental trill [r] (Moscati 1980: 32, §8.25). In the DSS it is frequently omitted, often near gutturals, suggesting a weaker pronunciation (Qimron 1986a: 26, §200.14). In Tiberian, the default pronunciation was either a voiced uvular roll [ʀ] or uvular frictionless continuant [ʁ̖]. It was articulated as an emphatic apico-alveolar roll [ṛ] “when preceded by ‫ דזצתטסלן‬or followed by ‫ לן‬and when either resh or one of these consonants has shewa” (Khan 2012: 92). Sin /ś/ ‹‫›ׂש‬, voiceless alveolar lateral fricative [ɬ], but later merged with /s/. The use of dots to distinguish Sin and Shin is part of the Tiberian pointing system. Shin /š/ ‹‫›ׁש‬, voiceless postalveolar fricative [ʃ]. Tav /t/ ‹‫›ת‬, voiceless aspirated alveolar plosive [tʰ]. In Tiberian without daghesh, a voiceless alveolar fricative [θ]. As for Hebrew/Aramaic vowels, Proto-Semitic and the pre-Septuagint Hebrew that followed it possessed six vowel phonemes (Joüon and Muraoka 2006: 33, §6b; Blau 2010: 111, §3.5.5.1; Rendsburg 2013: 105b), as shown in Figure 3.1. Before the translation of the Greek Pentateuch, Hebrew vowel length was phonemic, but this was mostly lost by the medieval period. With the loss of phonemic vowel length, more distinctions in quality emerged. The Tiberian tradition possessed seven distinct vowel phonemes (Khan 2012: 94–5), as shown in Figure 3.2. The other medieval pronunciation systems known to us are the Babylonian and Palestinian. The Babylonian pronunciation system was similar to Tiberian, but did not distinguish /a/ and /ɛ/, so only possessed six vowel phonemes: /i, e, æ, ɔ, o, u/ (Khan 2013c). The Palestinian pronunciation system distinguished neither /ɔ/ and /a/, nor /e/ and /ɛ/, and in the manuscripts marked with Palestinian signs there is some confusion between /o/ and /u/, indicating that the sounds [o] and [u] were collapsing into a single phoneme. Palestinian therefore possessed five vowel phonemes: /i, e, a, o, u/ (Heijmans 2013). Samaritan Hebrew is primarily known to us from the oral liturgical tradition still in use in Samaritan synagogues today. This tradition may preserve features of the vernacular

FIGURE 3.1  Proto-Semitic Vowel Sounds.

56

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

FIGURE 3.2  Tiberian Vowel Sounds.

Hebrew spoken by Samaritans and Jews in the Second Temple period (Ben-Ḥayyim 2000: 4). In this tradition, the collapsing of /o/ and /u/ that is apparently under way in Palestinian has hardened such that they have combined to become complementary allophones of a single phoneme /u/ realized as [u] in open syllables and [o] in closed syllables. It does, however, possess /ɔ/. Samaritan therefore possesses five vowel phonemes: /i, e, æ, ɔ, u/. Turning now to Greek, the consonant systems from the Koine to Modern periods can be described using the inventory of twenty-eight phones in Table 3.3.

Table 3.3 Greek Consonants Bilabial Plosive

p pʰ

Nasal

Labiodental

Dental t tʰ

b

*β

f

v

θ

ð

r̥

r

s

z

Glottal

g ŋ

n

Approximant Lateral Approximant

Velar k kʰ

m *ɸ

Palatal

d

Trill Fricative

Alveolar

x j

ɣ

h

w

l

Classical Attic possessed fifteen consonant phonemes, /p, t, k, pʰ, tʰ, kʰ, b, d, g, m, n, l, r, s, h/ and seventeen graphemes to represent them ‹β, γ, δ, ζ, θ, κ, λ, μ, ν, ξ, π, ρ, σ, τ, φ, χ, ψ› (Allen 1987: 12–61). The phonetic realization of several of these phonemes had shifted by Modern Greek, but the phonemic structure of the consonant system remained mostly stable. In other words the sounds of some of the consonants changed, but this did not cause any to merge or be lost. The only exception is that the voiced environmental allophone of the sibilant gained phonemic status in the Byzantine period [z] → /z/, put simply: Greek /z/ ‹ζ› became a “real” letter. The following describes the likely pronunciation of these phonemes when the Pentateuch was translated into Greek with a summary of their key developments into the medieval period:

SEPTUAGINT TRANSCRIPTIONS AND PHONOLOGY 57

Beta /b/ ‹Β, β›, voiced bilabial plosive [b], later a voiced labiodental fricative [v]. Occasional interchange of ‹β› : ‹π› indicates that some speakers only possessed /p/ (Gignac 1976: 77). Gamma /g/ ‹Γ, γ›, voiced velar plosive [g], later a voiced velar fricative [ɣ]. Prior to another /g/ a velar nasal [ŋ]. Many speakers only possessed /k/ and /t/, evidenced by the fact that the interchanges ‹γ› : ‹κ› and ‹δ› : ‹τ› occurred “very frequently throughout the Roman and Byzantine periods,” (Gignac 1976: 76–7). Delta /d/ ‹Δ, δ›, voiced dental plosive [d], voiced dental fricative [ð]. Zeta /sd/, [z] ‹Ζ, ζ›, consonant cluster /sd/, or the voiced allophone of the sibilant [z]; later an independent phoneme /z/. Theta /tʰ/ ‹Θ, θ›, voiceless aspirated dental plosive [tʰ], later a voiceless dental fricative [θ]. Kappa /k/ ‹Κ, κ›, voiceless velar plosive [k]. Lambda /l/ ‹Λ, λ›, voiced alveolar lateral approximant [l]. Mu /m/ ‹Μ, μ›, voiced bilabial nasal [m]. Nu /n/ ‹Ν, ν›, voiced alveolar nasal [n]. Word-finally often assimilates to a following consonant. Pi /p/ ‹Π, π›, voiceless bilabial plosive [p]. Rho /r/ ‹Ρ, ρ›, voiced alveolar trill [r], but word-initially and after another /r/ voiceless [r̥]. Sigma /s/ ‹Σ, σ, ϲ, ς›, voiceless alveolar fricative [s], but before a voiced consonant voiced [z]. Tau /t/ ‹Τ, τ›, voiceless dental plosive [t]. Phi /pʰ/ ‹Φ, φ›, voiceless aspirated bilabial plosive [pʰ], later a voiceless labio-dental fricative [f ]. Chi /kʰ/ ‹Χ, χ›, voiceless aspirated velar plosive [kʰ], later a voiceless velar fricative [x]. The aspirate /h/ ‹῾›, voiceless glottal fricative [h], is only phonemic word-initially. As for Greek vowels, Allen’s (1987: 62–79) reconstruction of the Classical Attic system is the standard description. Classical Attic is very close to the Greek that became the widespread Koine Greek across the Mediterranean and Middle East after 323 BCE, before the translation of the Septuagint (Figure 3.3). Modern Greek possesses only five vowels (Figure 3.4), with no phonemic length distinctions (Arvaniti 1999: 3, 2007: 120). Most of the changes from Classic Attic to Modern Greek took place during the period when the books of the Septuagint were being translated; furthermore, Greek was not pronounced the same by all speakers. Teodorsson (1978: 94–8) identified at least two systems of pronunciation: a conservative pronunciation by the elite and an innovative pronunciation by the non-elite educated majority. This observation is commended by Horrocks (2010: 164), while also assessing Teodorsson to have “overinterpreted his data” when he claims the innovative system of spoken Attic in Attica was so radically advanced that vowel length distinctions had been lost even as early as the mid-fourth century. The most significant data currently available for reconstructing these developments are spelling variations in Attic inscriptions (Threatte 1980) and Egyptian papyri (Gignac 1976). The traditional chronologies of developments in Greek pronunciation (Schwyzer 1938: 233; Threatte 1982; Allen 1987: 78, 86) describe the conservative pronunciation (cf. Teodorsson 1977: 256), as does Vessella (2018: 37). Figure 3.5, inspired by Allen (1987: 78, 86), is an attempt to present the chronology of these developments graphically, with some modifications in light of Modern Greek (Arvaniti 2007). One problem with a

58

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

FIGURE 3.3  Classical Attic Vowel Sounds.

FIGURE 3.4  Modern Greek Vowel Sounds.

FIGURE 3.5  Diachronic Vowel Change in Conservative Greek Pronunciation.

SEPTUAGINT TRANSCRIPTIONS AND PHONOLOGY 59

FIGURE 3.6  Diachronic Vowel Change in Innovative Greek Pronunciation.

visual presentation of this kind is it fails to convey that individual sound changes develop and disseminate over periods of time. Changes are marked at the point when they had likely been adopted by the majority of conservative speakers. Dashed lines indicate unreconstructable chronology. The innovative system was spoken by the educated majority, which would have included scholars such as translators. Figure 3.6 is a summary of this system based on Horrocks’s (2010: 166) judicious assessment of Teodorsson (1977: 251–2), and is consistent with Gignac’s (1976) data.

ONGOING RESEARCH QUESTIONS A broadly accepted set of terms and definitions for working on transcriptions is a desideratum. There is still much scope for refining research methods. The methods and data sources employed by different studies lack consistency, making a comparative survey of previous conclusions very difficult. Given the disparate and unconnected nature of discussions of transcriptions, there is real value to be found in detailed literature surveys that are willing to sift through journal indices and lists of unpublished dissertations. Some particular topics that may prove fruitful are: a survey of literature prior to 1900, discussions of proper nouns and transcriptions in the Mitteilungen des SeptuagintaUnternehmens, La Bible d’Alexandrie, IOSCS Congress volumes, and comparisons of Greek spellings within studies that are organized by nominal types (i.e., qal, qatl, qatal, etc.) such as Brønno (1943) and Yuditksy (2017). Translations of relevant Israeli works into English would also be a boon for the field.

60

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Gathering lists of transcriptions that already exist and assessing how consistent the sources and methods are between them would be a valuable contribution. There is also much work still to be done in creating exhaustive lists of transcriptions in the various corpora. Exhaustive lists of transcriptions can be found in Knobloch (1995) for the Pentateuch, Myers (2019) for 2 Esdras, and Yuditsky (2017) for the Hexapla. Even though the sources of data used for these lists mean the spellings of these transcriptions cannot be compared consistently, the indexing of all transcriptions in these corpora is an important tool to facilitate further research. Books for which it would be most useful to have transcriptions exhaustively indexed are Paralipomenon (Chronicles), because of the high frequency of proper nouns and since it was translated so late, and perhaps also a book that was likely translated halfway between the Pentateuch and Chronicles. Benjamin Kantor’s thesis (2017) discusses the grammar of transcriptions in the Hexapla. He is currently working with the Hexapla project to produce an online database of the Secunda. The study of transcriptions overlaps with a number of other fields that have not yet as significant an impact as they deserve upon Septuagint studies. Onomastics and loanwords are now sprawling disciplines in their own right, with major publications addressing methodological issues and attempting to gather broad datasets to enable comparative research. Krašovec’s (2010) monograph can be seen as one example of an interdisciplinary attempt to combine OT exegesis, transcription studies, onomastics, and lexicography (despite some significant shortcomings). One obvious point of comparison for transcriptions in the Septuagint are transcriptions in the Hexapla, but other, more disparate, datasets also contain Hebrew/Aramaic transcriptions and await an enterprising researcher to bring them to bear upon Septuagint studies. These datasets include transcriptions in Greek magical papyri, on ossuaries and on other objects, such as, for example, the Halbturn Amulet (Bar-Asher 2010). An article by Steiner (2005) makes use of transcriptions in inscriptions to good effect. The Tiberian text itself contains several text critical layers, most notably Ketib/Qere variations. Little work has been done comparing these texts to the Vorlagen implied by transcriptions. Research questions on Semitic and Greek phonology can be easily lifted from the section above on debates in the field. Beyond that discussion, some other avenues worthy of exploration include: What evidence can be gathered to determine when the fricativizations of Greek /au/ and /eu/, Latin /w/, and Hebrew/Aramaic /w/ occurred, and how are these developments related to one another? What contribution can the development of transcription spellings in textual transmission make to our understanding of Byzantine and Medieval Greek phonology? and To what extent can Septuagint transcriptions help identify more precisely the development of the sounds of Hebrew and Aramaic vowels in the Second Temple period? The study of transcriptions and their contributions to the phonology of Hebrew, Aramaic, and Greek has huge potential. The field remains wide open for new scholars interested in entering the space.

RECOMMENDED RESOURCES Survey of Previous Research 1. Knobloch (1995). Chapter 1 of this unpublished dissertation is the best summary of previous research up to the mid-1990s (available on ProQuest). For more recent work see Yuditsky (2013) who was apparently unaware of Krašovec (2010).

SEPTUAGINT TRANSCRIPTIONS AND PHONOLOGY 61

2. Kantor (2017) and Myers (2019). Recent doctoral dissertations with relevant literature surveys. 3. Giannakis (2014) and Khan (2013a). Essential reference works.

Historical Phonology 4. Campbell (2013). The principles of diachronic phonology in Chapter 2 are essential for understanding both the phonology of transcriptions and factors of unintentional change in their text-critical history. Phonological and phonetic description can seem incomprehensible without an understanding of the physical processes that the notation is describing. 5. The Handbook of the International Phonetic Association (1999). The fastest tool for selfstudy. 6. Ladefoged and Johnson (2011). Further explanation and exercises than (5). 7. More advanced reading should include Lass (1984), Ewen and Hulst (2001), and for phonetics Laver (1994).

Hebrew Phonology 8. Khan (2012). This accessible book is essential reading for Tiberian Hebrew phonology. Chapters 1 and 2 provide a helpful summary of the development of the Hebrew text, and can be read before tackling the topic further in Tov (2012). 9. Khan (2013b). An overview of the spectrum of Hebrew traditions. 10. Reference works for the individual traditions are: Blau (2010) for Tiberian, Yeivin (1985) for Babylonian, and Fassberg (1990) for Palestinian.

Hebrew Derivational Morphology 11. Sagarin (1987). Gives lists of contextual forms and lexemes under each sound pattern, and a short descriptive summary. 12. See also the etymological notes in Joüon and Muraoka (2006: §88).

Hebrew/Aramaic Lexicography and Etymology 13. Donner (2013). The most up-to-date etymological dictionary of Biblical Hebrew and Aramaic. Can be used effectively with minimal German.

Greek phonology 14. Allen (1987). An accessible and essential introduction to Greek phonology. 15. Horrocks (2010: 160–72). A lucid summary of the conservative and innovative systems of Koine Greek phonology. 16. Vessella (2018). A detailed study of conservative speech. 17. For other helpful diachronic summaries of Koine Greek vowels see Schwyzer (1938: 232–4) and Gignac (1989). Significant studies of the Egyptian data are Gignac (1976) and Teodorsson (1977). For the development of Greek within Attica, see Threatte (1980). For the relationship between Egyptian and Attic Greek, see Threatte (1982).

Greek Derivational Morphology 18. Palmer (1945). An analysis of word formation in Greek papyri. Further work in this area would be a boon to the analysis of Septuagint transcriptions.

62

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

19. For analysis based on the wider Greek corpus, Schwyzer’s (1953) indices helpfully list references to derivational morphemes in his grammar.

Text Criticism 20. Tov (2012). One of the best introductions to text criticism in any corpus. See especially Chapter 4 on the transmission of the Hebrew text and Chapter 3 §II.B.1-2 on the text of the Septuagint. These discussions are of more immediate application to transcriptions than Tov (2015a). When attempting to reconstruct the development history of spellings, the following works are helpful to compare one’s own proposals to other analyses of spelling changes during the same periods: Mayser (1970) for Ptolemaic, Gignac (1976) for Roman and Byzantine. For Medieval Greek we await the publication of Holton et al. (2019) with great interest. For paleography see Thompson (1912).

Onomastics 21. Eichler, Hilty, Löffler, Steger, and Zgusta (1995). A comprehensive work covering both onomastic theory and the study of onomastica from across the globe, including sections on names in language contact and research methods. There are individual chapters on the Hebrew onomasticon, as well as other related ancient onomastica.

Loanwords 22. Zuckermann (2003). The theoretical discussion in the introduction and Chapter 1 are an excellent treatment of the problems around “loanwords.” Zuckermann’s terminology solves many long-standing problems. See also Haspelmath and Tadmor (2009), and the associated online database wold.clld.org.

CHAPTER 4

Septuagint Lexicography PATRICK POUCHELLE

INTRODUCTION Lexicography is the science of collecting and analyzing words of a given language in order to produce a lexicon or a dictionary. As for the Septuagint, the key issue is not how to collect words—as the corpus is somewhat limited—but rather how to determine the meaning of words that occur in the Greek version. In the past, some scholars have argued that many Septuagint words express the meaning of their Hebrew equivalents. But other scholars prefer to think the Greek words mean the same thing whether they are used in translations of Scripture or in any other Greek text. The truth lies somewhere in the middle. But how can we know where? The answer to this question is complex. Important factors include the style of translation underlying the Greek version, the intertextual relationship between books within the Septuagint corpus, the manifold interpretative traditions known to the translators, and the extent to which Hellenized Jews spoke to one another in a special jargon or dialect (if at all). A brief survey of the history of modern lexicography can offer insights upon the current tools for this kind of research, including concordances, dictionaries, and encyclopedias. The subsequent section then presents the main questions raised in Septuagint lexicography: How do we define the Septuagint? What base text should be used? Should the research be oriented toward the source language (Hebrew or Aramaic) or should the Greek text be studied for its own sake? How should we present the meaning of the Greek words? Finally, the last section in this chapter will suggest some ways to delve deeper into the topic.

HISTORY AND TOOLS OF MODERN SEPTUAGINT LEXICOGRAPHY The Text Modern lexicography of the Septuagint actually began when the text of the Septuagint was first printed. It is not the aim of this chapter to describe the manuscripts with which it was established or how they were collected and used to produce a critical edition. However, one should note that it is impossible to produce a bilingual dictionary until a text is established and words are collected in a concordance. This first became possible after the printing revolution. The first printed edition of the Greek Bible was part of the Complutensian Polyglot Bible published in 1522 in Spain (see also Fernández Marcos, 2017). This work contains

64

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the OT in Latin, Hebrew, and Greek as well as the Targums with Latin translations. Notably, the Greek text was republished in the Polyglot of Antwerp (1572) and of Paris (1645) and alone in the Aldine edition in Venice in 1518. This Aldine edition was republished in the editions of Strasburg (1526), Basel (1545), and Frankfort (1597). Around this time, the Sixtine edition (1587), which was based mainly on Codex Vaticanus, considerably improved the quality of the text compared to other printed editions. The Sixtine was reprinted in Paris (1628) and was part of the London (or Waltonian) Polyglot (1657). Another text was then produced by Grabe (1707–20), mainly according to Codex Alexandrinus. Nevertheless, the Sixtine edition was the preferred text for about two centuries and was used as the basis for the first critical edition of the Septuagint published by Holmes and Parsons between 1798 and 1827. As a consequence of his extensive work on the New Testament, Tischendorf also published his own critical edition of the Septuagint in 1850. Nevertheless, scholars at the University of Cambridge felt the need for a better critical edition. This initiative resulted in a complete but preliminary text published by Swete between 1887 and 1894, and this edition then became the preferred Septuagint text. A similar but larger project in Cambridge set out to edit the entire Septuagint with a comprehensive critical apparatus, which made significant headway but was eventually left unfinished. Then in 1908, the so-called Septuaginta-Unternehmen was formed in Göttingen. Its director, Alfred Rahlfs, who was a student of Paul de Lagarde, produced a smaller, minor critical edition of the Septuagint, which Hanhart revised in 2006. This edition is now the main reference text. The larger Göttingen project of editing the text of the Septuagint with an up-to-date and comprehensive critical apparatus has so far managed to complete about two-thirds of the Septuagint corpus and should be the preferred text wherever available. For the present purposes, it is worth pointing out that even establishing the text of the Septuagint itself is still an ongoing project. Many important books, like that of Proverbs, still need an updated critical edition in the Göttingen series. In this connection, the overall impact of the ancient Christian recensions, such as those of Origen and Lucian (also called Antiochene text), is still poorly explored, but is now edited for 1–4 Kingdoms as well as 1–2 Chronicles (Fernández Marcos and Busto Sáiz 1898–1996). The same could be said about the Jewish revisions undertaken by Aquila, Theodotion, Symmachus, and others, for which there are only a few remaining fragments, and for the contribution of the secondary versions (see Den Hertog 2002).1

Concordances A medieval monk, Euthalius of Rhodes, is thought to have written the very first Greek concordance of the Bible around 1300, although it was unfortunately lost. The first modern one was that of Conrad Kircher, a Lutheran theologian from Augsburg in 1607. His concordance also served as a lexicon, followed by a treatise in 1622. His concordance was based on the Aldine text in its Basel edition. This work was criticized for several reasons, notably because he organized the entries according to the Hebrew roots to which a given Greek word was thought to correspond. More than a century later in 1718, Abraham

On the major recensions of the Septuagint and Origen’s Hexapla, see the respective chapters by Mäkipelto and Gentry in the present volume. On the secondary versions, see Cox.

1

SEPTUAGINT LEXICOGRAPHY 65

Trommius, a Dutch scholar, aimed at producing a better concordance, still based on the Aldine text, in its Frankfort edition. He organized words according to the Greek alphabet. Within a given word entry, the occurrences are organized according to the Hebrew root that the Greek headword is thought to correspond to (with the deuterocanonical books listed at the end of each entry), the instances of which are then listed in their canonical order. The concordance volume also includes a Hebrew/Aramaic-Greek lexicon, as well as a work by Montfaucon concerning the major Septuagint revisions. Trommius’s concordance was followed but not superseded by the simpler one of Morrish, published in 1887, and can still be consulted with some profit against the now-classic concordance of Hatch and Redpath (HR), as Trommius gathered some variant Greek words that are absent from Hatch and Redpath. The concordance of Hatch and Redpath is still a primary reference tool today. Initially published between 1897 and 1906, this concordance was later supplemented with a concordance of Greek proper names, Ecclesiasticus, and the Hexaplaric fragments. It was republished in 1998 by Baker Academic in one volume, with an introductory essay by Kraft and Tov, as well as a Hebrew/Aramaic Index to the Septuagint by Muraoka that replaced the old one produced by Hatch and Redpath. This index was revised some years later (Muraoka 2010a). In HR, each Greek word is followed by a numbered list of the Hebrew words to which it corresponds. Each form of a given Hebrew word (e.g., Qal, Piel) is distinguished by a small Latin letter. Then each occurrence of the Greek word is given, following the canonical order of the Sixtine edition of the Septuagint, adding 3–4 Maccabees and providing both texts of Daniel, but excluding the Psalms of Solomon or the Prayer of Manasseh.2 Each word occurrence, including any variants in Alexandrinus, Vaticanus, or Sinaiticus, is provided in its near context and followed by a number and, if necessary, a letter corresponding to the Hebrew word to which HR thought it corresponded. When correspondence is doubtful or otherwise problematic a question mark (?) or dagger (†) is added, respectively. When there is no corresponding Hebrew word, the number is replaced by a dash (–). In HR, correspondence is interpreted as formal, word-for-word alignment with the MT, even if another Hebrew word from a different Vorlage would have been more coherent. Hatch and Redpath acknowledged the fact that the Septuagint Vorlage was not identical to the MT, but they did not try to reconstruct it. After all of the occurrences of a word in the Septuagint, a list of occurrences in the fragments of the major revisions appears, although with no context or corresponding Hebrew. Apart from the choices made for determining Hebrew equivalents, which are sometimes debatable, the main drawback of the HR concordance is the fact that it obviously does not take into account the critical editions published later. But other limitations are worth mentioning as well. For example, the most frequent Greek words, mainly particles or prepositions, are mentioned only with the indication passim or, like παρά, with a bare list of references with no corresponding Hebrew at all. Also, very few grammatical and lexical hints are provided. For instance, there is no mention of the cases or the stems of the Greek words or of their antonyms, synonyms, or derived forms. The absence of the latter was addressed by Jacques (1972). Other indices and concordances help supplement shortcomings or gaps in HR. To begin with, the Hebrew reverse index in HR was improved by Dos Santos (1973), and

The Hebrew corresponding to Sirach is provided in Appendix 2 of HR.

2

66

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

later replaced by Muraoka (2010a).3 In the first part, Muraoka presents each Greek word followed by its Hebrew/Aramaic equivalent, including statistics and corrections from HR, and the second part contains each Hebrew/Aramaic word with a list of Greek equivalents. Although the index includes texts absent from HR, it nevertheless excludes the major revisions as well as Greek proper names. The index of the Antiochene text may also be used as a supplementary concordance (Fernández Marcos, Spottorno Díaz Caro, and Cañas Reíllo 2005). It provides Greek words with a simple list of occurrences organized by Hebrew equivalents in the MT. Owing to their poor attestation and their fragmentary nature, the major revisions are of difficult access. In 1875, Field published their attestation in Origen’s Hexapla. Since then, some further material has been discovered so that the second apparatus of the Göttingen edition is more up-to-date. A concordance has also been produced for Aquila (see Reider 1916, 1966; Tov 1973). The rise of the so-called digital humanities opened up broader possibilities than printed concordances. The Thesaurus Linguae Graecae digitalized the text of Rahlfs (1935), as well as of the main OT Pseudepigrapha. Soon after the Stuttgart edition of the MT (BHS) was digitalized, Tov directed the CATSS (Computer Assisted Tools for Septuagint/ Scriptural Studies) project, which vertically aligned the Rahlfs edition with the BHS. The CATSS database is also the basis of similar tools in the main Bible software programs.

Lexicons and Dictionaries The first Septuagint lexicon was that of Biel, edited in 1779–80 and based on the concordance of Trommius. It was followed by Schleusner in 1820–1, which remained the primary lexicon until the end of the twentieth century. In it, each Greek word is followed by a Latin gloss, its Hebrew equivalent, and the verses in which it occurs. At points a philological discussion follows, which is sometimes still useful for current research. However, since the important work of Deissmann (1908), the Septuagint lexicography must take the contemporary papyrological and epigraphical material into account. As for the papyrological evidence, the dictionary of Kiessling and Preisigke (1925–) could be used, although much newer evidence has not yet been added. Unfortunately, the epigraphical evidence is still poorly explored in Septuagint lexicography, a need that Aitken (2014b) has discussed. These sources of evidence are important for understanding how the language of the Septuagint is informed by contemporary Greek culture. This recognition led to the work of Moulton and Milligan (1930), which was limited to the vocabulary of the NT, and later, to the New Documents Illustrating Early Christianity series initiated by Horsley (1981–2012).4 While including some such nonliterary materials, along with the classical literary corpus, the lexicon by Liddell, Scott, and Jones (LSJ; 1925–40) sometimes discusses word meanings apparently unique to the Septuagint, as does the LSJ supplement (1968). Unfortunately, LSJ does not always account for the Hebrew source text, a problem noted by Caird (1968, 1969) and Lee (1969).5 In 1989, Rehkopf edited a short As a matter of fact, Muraoka (1984) began by editing an index of 1 Esdras, then edited the index that was added to the new edition of HR in 1998, until finally he revised his work again (2010a). 4 For etymological studies, one could also use Chantraine (2009) or Beekes (2010). 5 Even the revised supplement (Liddell, Scott, and Jones 1996) should be used with caution. More recent dictionaries could be of value, such as the Brill Dictionary of Ancient Greek (Montanari 2015) translated from the Italian dictionary of Montanari (1995) or the Diccionario Griego-Español, which is still incomplete (Adrados 1980–2007). 3

SEPTUAGINT LEXICOGRAPHY 67

German lexicon of the Septuagint in which glosses were drawn from LSJ.6 Some hints on Septuagintal words also appear in BDAG (Bauer, Danker, Arndt, and Gingrich 2000) for the NT, in the patristic lexicon of Lampe (1961), as well as in the lexicon by Sophocles (1914) that is dedicated to the Roman and Byzantine period. Trapp (2001–17) can also be interesting for reception history. Despite the existence of some helpful resources, the need for a full-fledged lexicon of the Septuagint helped motivate the formation of the International Organization for Septuagint and Cognate Studies (IOSCS) in 1968 (see Fritsch 1970b; Kraft 1972a; Gehman 1974, and Boyd-Taylor 2001: 48). Although it would take decades, eventually Lust, Eynikel, and Hauspie published the first exhaustive lexicon of the Septuagint between 1992 and 1996 (hereafter LEH, third edition in 2015). In this work, up to five references are given per word, with a gloss for each specific meaning given in English with some additional information (e.g., “neol.” if the word is thought to be a Septuagint neologism). Bibliography also appears for each word.7 This lexicon is translation-oriented, that is to say, the suggested word meanings are often clearly influenced by the meaning of the underlying Hebrew/Aramaic word translated. Not long after, in contrast to this approach in LEH, Muraoka produced another lexicon of the Septuagint that is oriented toward a Greek approach, initially published in three successive volumes (1993, 2002, and 2010b). Generally speaking, recourse to the Hebrew/Aramaic is scarce. Each Greek word is presented with (a) select morphological information; (b) semantic analysis organized by distinct word meanings; (c) a list of Greek words semantically related to the headword; and (d) bibliography. Finally, Chamberlain (2011), who had initially participated in the LEH project, also published his own lexicon that contains only Septuagint words that are unattested in the New Testament, as a kind of supplement to BDAG.

Encyclopedias The Theological Dictionary of the New Testament (TDNT; Kittel and Friedrich 1964) frequently offers a section on the Septuagint use of a word occurring both in the New Testament and in the Septuagint. These contributions are mainly due to the work of Bertram, a German scholar who specialized in the Septuagint. These sections should be used with prudence, as in general TDNT suffers many drawbacks, on which see Barr (1961). First, Septuagint words absent from the NT are, of course, omitted. Second, this dictionary conflates words and concepts. For example, in the entry for the word ἀνάστασις, the concept of resurrection is discussed even when denoted by other words. Moreover, when it is discussed, the Septuagint is considered as a whole without taking into account the particularities of each book. Finally, the most problematic issue in TDNT is the bias of many of the contributors. Produced within the context of the German Nazi government, some of them, like Kittel and Bertram, considered the Septuagint to be the first step of the de-Judaization of the religious mind. They viewed history as a continuous process in this regard, in which the Septuagint represented a preparatory step for the New Testament, which itself contributed to the removal of the Jewish character of the Christian religion. For instance, in his entry on παιδεία (“education”) and παιδεύω (“to

Although Rehkopf claims that he used the first edition of Rahlfs, he rather follows HR and therefore likewise omits word occurrences in the books that are absent from Swete (e.g., Psalms of Solomon). 7 Another useful, though now dated, resource in this connection is RBLG, edited by Boned Colera and Rodríguez Somolinos (1998). See also l’Année Philologique. 6

68

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

educate”), Bertram considered the use of these words to translate ‫“( יסר‬to educate” or “to discipline”) and ‫“( מוסר‬discipline”) as the introduction of the concept of education into Jewish thought, implying that this concept was absent from the Hebrew Bible, an idea that he himself then has to contradict in the same article (notably with Proverbs). This impact is insidious as these scholars expressed themselves as openly anti-Semitic in some pamphlets but never stated it explicitly in an academic work like the TDNT (on this see Heschel 2008 Nicklas 2013). Some other resources deserve brief mention. The different encyclopedias of Old Testament, mainly the Theological Dictionary of the Old Testament (Botterweck, Ringgren, and Fabry 1970–2019), but also the Theological Lexicon of the Old Testament (Jenni and Westermann 1997) and the New International Dictionary of Old Testament Theology and Exegesis (VanGemeren 2012), include some thoughts on the Septuagint for many Hebrew and Aramaic word entries, but not systematically. Similarly, the NT encyclopedias produced after the TDNT also deals with the Septuagint. For instance, the Theological Lexicon of the New Testament (Spicq 1994) offers good insights for words that are badly dealt with by the TDNT or even absent from it, and also takes into account some papyrological and epigraphical material. The Exegetical Dictionary of the New Testament (Balz and Schneider 1990) and the New International Dictionary of New Testament Theology also (Brown 1975–86) can also be consulted.8 The first installment of an important new resource is entitled the Historical and Theological Lexicon of the Septuagint (Bons 2020, hereafter HTLS). This lexicon will eventually be published in five volumes and will include a select range of Greek words of particular importance. The semantic analysis in each word entry is divided into six parts: classical literature, papyri and inscriptions, the Septuagint, Jewish Greek literature, New Testament, and early Christian literature, which is then followed by a bibliography. The HTLS project aims at avoiding the problems of the TDNT. In this connection, the six-part division is mainly pragmatic, and is not meant to promote a simplistic, evolutionary view of Greek lexemes from classical literature to early Christian literature. On the contrary, the HTLS acknowledges and gives considerable attention to the nuanced and often nonlinear development of Greek words over time. To date, this lexicon is the only tool that exhaustively discusses the history of a Greek word, from etymology through the second century CE, including discussion of important papyrological and epigraphical material, interference from Hebrew/Aramaic, and the development of technical and religious word meaning.

VIEW AND DEBATES The Scope of Septuagint Lexicography The concept “Septuagint” is not that easy to define. Which books are part of it? Which text should be used? On one hand, the Septuagint initially referred to the legendary seventy translators who in the Epistle of Aristeas are said to have translated only the Pentateuch. On the other hand, the word “Septuagint” also commonly refers to the canon of the Old Testament of the Greek-speaking Christians.9 However, even this canon varied over This work translated and enlarged the German Theologisches Begriffslexikon zum Neuen Testament edited between 1967 and 1971, but a newer edition has been released. See Coenen and Hacker et al. (2014). 9 On the Septuagint and the biblical canon, see Meade in the present volume. 8

SEPTUAGINT LEXICOGRAPHY 69

time. For instance, today 1 Enoch is included only in the canon of the Ethiopian church. This church reported to the patriarch of Alexandria and its scriptures ultimately derive from the Septuagint (possibly by way of the Coptic). Of course, 1 Enoch was included by some Greek early Christian communities. Is 1 Enoch then part of the Septuagint or not? Conversely, the book of the Psalms of Solomon is only sporadically included in few Septuagint codices, and so many scholars therefore categorize it as part of the OT Pseudepigrapha. Yet this book was included in the critical edition of Rahlfs, and retained in that of Rahlfs-Hanhart, notably because Codex Alexandrinus mentions the Psalms of Solomon in its table contents (however not in the OT sections) but the folios that should contain them are missing. These marginal cases demonstrate how the current delimitation of the Septuagint is a choice made by modern scholars. Therefore, a pragmatic approach tends to limit itself to the contents of the critical edition of Rahlfs, thus avoiding a direct affiliation to any known church, but with a clear Christian delimitation (see Brock and Lee 1972). But by default, this approach isolates Septuagint lexicography from other important sources, like 1 Enoch, that could be of evidentiary value. With this in view, a historical and maximalist approach to Septuagint lexicography will include any and all Jewish literature in Greek, including the revisers, but also the Pseudepigrapha, provided they are not of Christian origin, as well as Philo or Josephus (see Kraft 1972a).10 Yet this maximalist approach would result not so much in a lexicon of the Septuagint per se as it would a lexicon of Jewish literature in Greek during Antiquity, which introduces its own methodological questions. Of particular importance are the lexicographical differences that can be expected between translation and works composed directly in Greek, such as 4 Maccabees or even Philo. But this kind of question can be pressed further. Even among translated books, differences might be expected between very literal books, like Greek Ecclesiastes, and those with a higher register of language, such as Greek Job. The same question applies even among books originally composed in Greek, as between technical works like 4 Maccabees, which is a philosophical treaty, and Judith, which was composed in a way that imitates the style of translated books. Carried to its logical conclusion, this line of thinking might ultimately require a lexicon for each individual book, an approach that clearly would be neither practical nor desirable.

The Base Text Another important and related consideration for Septuagint lexicography is the base text used for lexical examination. The text of the Septuagint as it has been transmitted over the centuries is composite, contaminated by the texts of the Jewish revisers as well as the Christian recensions. The task of textual criticism is, of course, to recover the oldest possible text based on the available evidence. But the Göttingen edition is not yet complete for every book of the Septuagint. Thus, for Septuagint lexicography the choice of a base text is always a pragmatic decision that is aware of the inevitable inconsistencies. LEH was based on the critical edition of Rahlfs (1935), a decision that was made for practical reasons, such as Rahlfs’s edition being the most recent of all the critical editions of the entire Septuagint. Some variants found in the Göttingen and Cambridge editions were

Although this could be debated, since some OT Pseudepigrapha may originate from early Judeo-Christian communities. On what basis may these be excluded from Septuagint lexicography?

10

70

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

accounted for in LEH, but not systematically. For his lexicon, Muraoka (2009) used the Göttingen edition wherever possible and Rahlfs’s edition otherwise. He sometimes used the Cambridge edition (ibid.: ix) and also added the Lucianic material for 1–4 Kingdoms, 1–2 Chronicles, and Judges (ibid.: vii–viii). The question of the daughter translations should also be dealt with (see for instance Den Hertog 2002).

Source versus Target Language Orientation The debate over the possible existence of a specific Jewish Greek dialect is now over and answered in the negative. The number of apparent lexical, semantic, and syntactic anomalies of the Septuagint, as well as the New Testament, has been drastically reduced owing to a better knowledge of the popular Koine language, notably found in the Egyptian papyri. Although Hebrew and Aramaic interference is certainly present in the Septuagint, it does not amount to a specific Greek dialect but rather creates a specific linguistic register. That register would also be used to produce some texts composed directly in Greek, like Judith. In such cases, the text does not contain Hebraisms or Aramaisms, as interference from Hebrew or Aramaic, but rather “Septuagintalisms”: interference—or perhaps simply “influence”—from the register of other translated text of the Septuagint. Research into this question has been renewed today with the rise of modern translations studies and their application to the Septuagint (see van der Louw 2007). The question of approach remains for the translated texts: Should the meaning of a given Septuagintal word be deduced from the Hebrew word to which it corresponds, or should it be deduced from its place within the Koine Greek of its time? The approach of LEH, which is supported by the so-called interlinear paradigm underlying the recent English translation (Pietersma and Wright 2009: xiv–xv), focuses on the work of the translator and conceives of Septuagint lexicography as somewhat subordinated to Hebrew or Aramaic lexicography.11 This approach assumes that the feature of translated texts is radically different from non-translated text (see Boyd-Taylor 2001, 2004, following Toury 1995) and presumes that the translators aimed to render the source text as faithfully as possible. Therefore, from this perspective, Greek words are mainly “pointers” to underlying Hebrew/Aramaic words. This approach fails to take into account the fact that the Greek translated text was understandable by itself (see Muraoka 2013). Of course, the reception of the Septuagint text by those unable to read Hebrew offers a rationale for studying how it was understood by the Greek Fathers or the community reading one specific codex, approaches exemplified by the Bible d’Alexandrie or by the Brill Septuagint Commentary Series, respectively.12 Obviously, such an approach may lead to confusion of at least two different dimensions of a Septuagint word (Tov 1976): the meaning it had for the translators and the meaning(s) it has for people who later read it. Broadly speaking, the source-oriented approach to Septuagint lexicography pursues the understanding of the translator, while the target-oriented approach pursues the understanding of the reader. While the former is singular, the latter would be diverse. Yet it is crucial to remember that, whatever their skill level, the Septuagint translators had mastered the Greek language in one way or another. So it is incorrect to assert that the target text was not autonomous in any sense. Even for proponents of the interlinear Although LEH did not apply this approach slavishly (xvi). Edited by Stanley E. Porter, Richard Hess, and John Jarick. In 2019, fourteen volumes had been published, beginning with Joshua (Auld 2005) up to Leviticus (Awabdy 2019). As for issues related to commenting on the text of one specific codex, see Joosten (2008b). On the reasons for doing so, see Porter in the present volume.

11 12

SEPTUAGINT LEXICOGRAPHY 71

paradigm, the target text is not viewed as so distorted by interference from the source text that it was rendered completely unintelligible for ancient readers. Hence—moving away from the perspective of Boyd-Taylor (2001, 2004)—an approach oriented to the target culture could also be used to understand the work of the translators, viewing the Septuagint as a literary product of the Greek Jewish diaspora (Dhont 2018), as suggested by polysystem theory (e.g., Even-Zohar 1978 and 1990). This alternative advocates for pure Greek lexicography, as represented by Muraoka’s lexicon. From this perspective, a translation is the product of the target culture and is therefore primarily executed according to conventions from the target culture, not the source culture. As a consequence, the boundary between translated texts and compositional texts becomes blurry. Texts like Judith, for instance, appear to have been produced with a “Septuagintal style”—that is to say, it conformed to certain linguistic conventions expected by its readers (see Joosten 2007b). Still, the target-oriented approach to Septuagint lexicography has difficulty dealing with transliterations, stereotyped translation of Hebraic/Aramaic idioms, and semantic borrowing, so that in the end the source-oriented approach is, in fact, also needed.

Translation Equivalents versus Definitions The debate over the superiority of translation equivalents versus definitions is not limited to Septuagint lexicography. Louw and Nida’s (1989) New Testament lexicon was a milestone in this regard, in that it not only offers a definition rather than a translation equivalent for each word but also organizes words according to semantic field rather than alphabetically. For Septuagint lexicography, LEH and Muraoka’s lexicon took two different approaches to this matter. On the one hand, LEH prefers giving as many translation equivalents as necessary, just as in LSJ. This decision is justified as a help for the user of the lexicon who can quickly gain an idea of the different translation choices available (2003: xv). On the other hand, Muraoka prefers providing one, or a few, definitions that are as precise as possible, with translation equivalents for some key occurrences. Muraoka (2009: xii) maintains that this approach is the best way to avoid misinterpretation that is possible when providing only word-for-word translation equivalents. With an English lexicon of the Septuagint that is widely used by non-Englishspeaking scholars to translate a Greek translation from Hebrew or Aramaic, as many as three levels of translation could occur: from Hebrew/Aramaic to Greek, from Greek to English, and then from English to the language of that specific scholar. Moreover, although Muraoka does not organize lexical entries according to semantic fields, he does provide words that are semantically related to each entry.

ONGOING QUESTIONS The main lexicographical issue in Septuagint studies is that of “interference,” that is to say, the degree and manner in which the thought, language, and writings of one specific cultural sphere, broadly defined, influenced the thought, language, and writings of another cultural sphere. Traditionally, the only interference accounted for in Septuagint scholarship is that from Hebrew and Aramaic, which has been tied to the question of Hebraism that has marked Septuagint lexicography from its inception. However, joined to the question of Hebraism and Aramaism is the question of Septuagintalism, as mentioned above, namely, the influence of certain books within the Septuagint (especially the Greek Pentateuch) on other books within the Septuagint, whether translated or not.

72

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

At the same time, the question of influence from non-Jewish Greek is no less important. If present, did such influence come more from contemporary spoken Koine Greek or from the written, literary corpora produced by more educated Greeks? Each of these possibilities is discussed below in their various forms.

Hebrew or Aramaic Interference The types of interference from the source text (Hebrew or Aramaic) or from the likely mother tongue of the translator (Aramaic) can be ranked according to degree: (a) transliteration, loanwords, and neologisms; (b) stereotyped translations and semantic borrowing; (c) other lexicographic phenomena (e.g., phonetic similarity, variations of syntax, and multiple translations); and (d) interference from the Aramaic language and late biblical Hebrew. Transliteration, Loanwords, and Neologisms  The translators sometimes transliterated a Hebrew or Aramaic lexeme—mostly indeclinable and sometimes through Aramaic (e.g., σεραφιν for ‫ׂשרפים‬, “seraphim”)—or borrow it to coin a new Greek loanword (e.g., σαββατίζω, “to keep Sabbath” or even “to Sabbathize”). The motivation for transliterating varies from book to book, and sometimes a single word is both translated and transliterated within one book. The reason for this choice is not always clear. Sometimes the translators might have interpreted the Hebrew or Aramaic word as a proper noun (see Deut. 10:6, Βηρωθ for ‫בארת‬, the plural of ‫באר‬, “watering place”). Other times, they might have ignored the meaning of the Hebrew or Aramaic term (perhaps this would explain the choice of σαβεκ to render ‫סבך‬, “thicket,” in Gen. 22:13), or considered it too technical for translation (as with some units of measurement, e.g., γόμορ for ‫חמר‬, a “homer,” or dry measure). It cannot be excluded that the translators have added transliteration as a stylistic feature to emphasize the translational characteristics of their text. Translators could also create neologisms (see Aitken 2013). In this case, instead of transliterating or using a loanword, a new word is coined within the semantic system of the target language. For instance, θυσιαστήριον was apparently coined to designate a legitimate altar, whereas the altars of other gods are usually a βωμός. As such, the word θυσιαστήριον does not exist in any non-Jewish or non-Christian texts, but it would nevertheless have been comprehensible to any Greek speaker as “place for sacrifice.” Notably, it is impossible to identify a neologism with total certainty, as many words that were once believed to be neologisms have been proven not to be so after being found independent of the Septuagint in contemporary papyri (e.g., προσήλυτος; see e.g., Butera and Moffit 2011, 2013; Moffit 2013). Moreover, coining a neologism is easy in Greek, and Greek authors are fond of it entirely apart from the work of translation, as it is something that embellished style. Hence, an apparent neologism could have arisen owing to a Hebrew/Aramaic lexeme that was difficult to render into Greek, but it could also have been motivated by the translators’ desire to demonstrate their facility with and style in the Greek language. Accordingly, the reasons why Septuagint translators opted to transliterate their text source, employ a loanword, or coin a neologism are not always clear and need further investigation.13 On these issues and a discussion of transcriptions and phonology in the Septuagint, see Myers in the present volume.

13

SEPTUAGINT LEXICOGRAPHY 73

Stereotyped Translations and Neosemy  A stereotyped translation (or simply “stereotype”) is a phenomenon where translators tend to render one Semitic lexical root consistently using the same Greek word or words from the same Greek lexical root. However, since the semantic range of Hebrew/Aramaic words do not correspond fully to that of Greek words, the semantic range of the latter tends to be expanded through stereotyping to include the nuances of the former. For instance, παιδεύω (“to educate”) systematically renders ‫“( יסר‬to educate, to discipline”), the latter of which could convey the nuance of corporal punishment. Thus, in the Septuagint, παιδεύω develops this nuance as well and will be used that way all the way through even Modern Greek. Similarly, χείρ (“hand”) gained the metaphorical nuance of “power” that is conveyed by the Hebrew ‫יד‬, since the former stereotypically rendered the latter in the Septuagint. The process evident in these phenomena is sometimes referred to as “semantic neologism,” although a more accurate term is neosemy, the development of a new meaning for an existing word. However, wherever this process appears to occur in the Septuagint, it cannot be totally proven to be the direct result of stereotyping translation, as new papyrological or epigraphical evidence could attest a word meaning in contemporary compositional Greek that was thought to be unique to the Septuagint. A second effect of stereotyped translation is to create grammatical collocations that are not attested in non-Jewish literature, such as the idiomatic expression υἱὸς τριάκοντα ἐτῶν, a literal translation of ‫“( בן־ׁשלׁשים ׁשנה‬son of thirty years”) to express an individual’s age (2 Kgdms 5:4; compare the more classical rendering of the same Hebrew idiom Ιωσηφ δὲ ἦν ἐτῶν τριάκοντα, “Joseph was thirty years old,” in Gen. 41:46). Other Lexicographical Phenomena  Hebrew/Aramaic interference sometimes led to various other phenomena. One of these is homophony, where Greek words were chosen because they share the same phonetic features as the source text in addition to suitable semantics. A good example is ἀλαλάζω (“to wail loudly”; e.g., Jer. 4:8) or ὀλυλύζω (“to cry out”; e.g., Ezek. 21:17) rendering ‫“( ילל‬to shout”). Consistent differences between Hebrew/Aramaic usage and the Greek translation could imply some kind of motivated reluctance. A good example of this phenomenon is grammatical number. Dorival (2016a: 292–3) has noticed that σωτήριον is usually used in plural (τὰ σωτήρια) to denote sacrifice in Greek religions. Although it would correspond naturally to the plural ‫ׁשלמים‬, also denoting a sacrifice, the translators use the Greek word only in the singular. Dorival suggests that this distinction in number reflects the translators’ unwillingness to identify a Jewish cultic term (‫ )ׁשלמים‬with terminology for what they considered to be invalid sacrifices (τὰ σωτήρια). Dorival (2016a: 294–5) has also suggested another type of Hebrew/Aramaic interference that he calls multiple translation, which occurs when multiple instances of a single Hebrew/Aramaic word are translated with different Greek words within only a few verses. This variatio could be due to stylistic considerations, but it could also represent an attempt to render all the nuances of the Hebrew/Aramaic word by several different translation choices.14 Interference from Aramaic and Late Biblical Hebrew  This subsection deals with interference in interpretation of the source text itself, rather than in the language of the target text. Indeed, the Septuagint translators lived many years after the composition of

On the question of style in the Septuagint, see Bons in the present volume.

14

74

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the text they translated. Obviously, the Hebrew language evolved over time. Additionally, their mother tongue may well have been Aramaic. Hence, sometimes the translators understood a Hebrew word in their source text according to its meaning in Aramaic or Late Biblical Hebrew (LBH).15 Aramaisms include loanwords, such as γ(ε)ιωρας (“proselyte”) and forced collocation, such as ἁμαρτάνω + ἐναντίον (“to sin before”; e.g., Josh. 7:20). Some examples of LBH interference include the use of ἀναβάλλω (“to delay”) corresponding to the Hitpael of ‫ עבר‬i i , which initially meant “to be furious” but later acquired the meaning “to neglect, to delay” (Joosten 2005: 152–64).

Septuagint Interference This question of Septuagintalisms may be discussed on at least three levels: (a) the debate about the influence of the Pentateuch; (b) anaphoric translations; and (c) imitating Septuagint style in translated texts as well as in compositional Greek texts. The Use of the Pentateuch as a Dictionary  The Pentateuch was probably the first translated text of the Septuagint and equally likely the first piece of Jewish literature in Greek. It is therefore reasonable to assume that it exerted influence on later Jewish writings. The more Jewish literature grew, the more each work interacted with and influenced others. Hence, some associations between Greek and Hebrew words that had been established by the translators of the Pentateuch were later maintained. But this idea of the Greek Pentateuch as a kind of dictionary for later translators must be nuanced (Tov 2016: 316–28). For no later translator slavishly adhered to the lexical choices of the translators of the Pentateuch. Each book is specific in this regard, even if some Greek words corresponding to technical Hebrew terms or central Biblical notions tend to be maintained (e.g., συναγωγή for ‫)עדה‬. This idea too could be nuanced further, since these relationships may have been established even before the translation of the Pentateuch. On this model, the Pentateuchal translators are thought to have used some preexisting correspondences, perhaps in oral translation, although this theory cannot be proven (see Aejmalaeus 2013), or through short Hebrew-Greek glossaries, although not found until now (see Yardney 2018). The Greek Pentateuch may well also have served as a lexical resource for translating rare or difficult Hebrew words. This again could be nuanced, as a word like ὀρτυγομήτρα (denoting perhaps “corn crake,” which renders ‫ׂשלו‬, “quail,” in Exod. 16:13; Num. 11:31, 32; Ps. 104 [105]:40) could have been introduced into Jewish Greek culture by the Greek Pentateuch, so that the later translators of Psalms and the writer of Wisdom of Solomon (Wis. 16:2) could have used it instinctively without needing to refer to the text of the Greek Pentateuch specifically. In this case, the Greek Pentateuch served as a center of the Jewish Hellenistic corpora, but it is hard to prove that later translators really used it as a dictionary. Anaphoric Translations  When a translation in one book clearly borrows from another book, or from other parts of the same book one could speak of anaphoric translation. This phenomenon is not specific to the Septuagint, but in the Septuagint it is notably, although not always, attested when a given book in Hebrew/Aramaic alludes to another This approach has been promoted by Joosten (2003: 587–600, 2016a). See also Loiseau (2016).

15

SEPTUAGINT LEXICOGRAPHY 75

Hebrew/Aramaic book—that is, allusion within the Hebrew Bible itself—especially to the Pentateuch. Yet, anaphoric translations may also be created in Greek when there is no explicit allusion in the source text. The book of Job is specifically translated according to this technique (Heater 1982). The earlier text that is borrowed might be the Greek Pentateuch or other books (e.g., Isaiah or the Psalms), so this phenomena—also referred to as “intertextuality”—could be used to estimate the date of the later, alluding translation.16 Imitating Septuagint Style in Translation and Composition  Although the debate over the existence of a unique Greek dialect of the Jewish Greek-speaking community is now closed (and answered in the negative), the corpus of Greek Jewish texts produced in that community nevertheless established a specific way to write in Greek. In other words, some texts that were written in Greek—whether translations or original compositions—were written by authors who sought to imitate the style of the former Septuagint translations. This phenomenon has biased the apparent statistical differences between translated texts and non-translated (compositional) texts (noted by Martin 1974), which therefore can hardly be used to prove that a given text must be a translation (see Sollamo 2001: 38–9). For instance, most scholars think that the Psalms of Solomon was translated from Hebrew owing to this kind of argumentation. Yet Pss. Sol. 2:12 uses the rare verb παραδειγματίζω (“to make an example”) with the phrase ἀπέναντι τοῦ ἡλίου (“before the sun”), just like in Num. 25:4. However, while Num. 25:4 uses the verb to refer to corporal punishment, Pss. Sol. 2:12 rather speaks of unveiling sins. This latter sense would not work in Num. 25:4, where the corresponding Hebrew rather speaks of physical impalement. So if there is an intertextual link between Pss. Sol. 2:14 and Num. 25:4, it only appears in Greek and not in the theoretical Hebrew Vorlagen. Hence, the question of the original language of the Psalms of Solomon may need to be reexamined (Joosten 2015: 35–6, 46–7).

Interference from Contemporary Jewish or Greek-Egyptian Culture The Septuagint is a product of a specific Greek culture. Even if both Hebrew/Aramaic and Septuagint interference are relevant for lexicography, as discussed above, “interference” from Greek is in fact more important and is still a poorly explored field of research. The translated texts of the Septuagint were accepted and used by people reading Greek, whatever their level of understanding was for Hebrew or Aramaic. Therefore, although it is important to notice when interference from Hebrew or Aramaic appears, it is also important to notice when such interference is absent and, conversely, when a correct Greek idiom is found where a Hebraism might have been expected (see also Lee 2020). For example, why are some words transliterated and others not? Why are some Hebrew and Aramaic idioms translated in a stereotyped way and others not? This field of research has still hardly been touched. Other forms of interference or influence may be detected in (a) conjectural or contextual translations; (b) the replacement of a Hebrew/Aramaic word, realia, or concept by a Greek one pertaining to the Hellenistic culture, the Egyptian Ptolemaic context, or the Greek-Jewish culture; and (c) influence of Egyptian language. Conjectural or Contextual Translations  If translators were unable to understand a Hebrew/Aramaic word, they may well have translated it contextually or by guessing. For instance, a ‫( קיקיון‬qîqayôn) is a specific and now unknown plant (Jon. 4:6–11). The 16

On the phenomenon of intertextuality in the Septuagint, see Theocharous in the present volume.

76

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

meaning of the word is lost. But the Septuagint offers κολόκυνθα (“gourd”), probably representing a guess at the meaning of the Hebrew. Aquila transliterates ‫( קיקיון‬κικεωνα) and Symmachus suggests κισσός (“ivy”), followed by Jerome, who likewise used hedera in the Vulgate. Contextual translations, on the other hand, could have been made according to etymological considerations. That is, when a Hebrew term is translated according to the basic meaning of its lexical root, rather than its specific lexical meaning. Other reasons for contextual translation could be stylistic. For instance, parallelism might be used to discern the meaning of difficult Hebrew words. Such phenomena are, however, hard to prove. Replacement of a Hebrew/Aramaic Word, Realia, or Concept by a Greek One  Some of these influences are obvious, for instance, rendering ‫“( צפון‬north”) with ἀπηλιώτης (“east”), as in Exod. 27:11. The rationale here is that the word θάλασσα (“sea”), corresponding to ‫“( ים‬west”) in this verse, is understood by the translator as “north,” just as it should be in Egypt, geographically speaking. Another case is τοπάρχης (“governor”) in Gen. 41:34, which is a Ptolemaic administrative term used to translate ‫“( פקדים‬officers”), demonstrating that the translator had knowledge of such administration. However, the extent of the impact of the cultural milieu is sometimes more difficult to prove. Dorival (2016a: 294) has noticed the systematic correspondence of δῆμος with ‫“( מׁשפחה‬clan”). He suggests that this choice shows that the Jewish Greek translators considered the Israelite clans in the desert a moving Greek polis. Joosten (2006: 349–61) has also suggested that some lexical choices in the Greek Pentateuch reveal a military culture that was familiar to the translators (e.g., the use of ἀποσκευή in Exod. 10:24). Some of these replacements are syntactic in nature. For instance, in Greek, βιβλίον could mean “letter” or “book,” but for the latter meaning, the word is used in the plural. The typical corresponding Hebrew, ‫ספר‬, is used in the opposite way: the plural rather means “letters.” Hence, this explains why βιβλίον would be used for “letter” in the singular when the corresponding Hebrew has the plural ‫ספרים‬, as in 2 Kgs 10:1. The translators seem to avoid the use of βιβλίον in plural for designating a “letter” (Pouchelle 2019). The Influence of Egyptian Language  Another way in which the influence of Egyptian culture affects translation is the presence of transliterations of Egyptian words. The case of θῖβις (“basket”) should probably be understood as interference from the contemporary Greek language, as this term appears in several papyri. The Egyptian word was likely adopted by Greek before the translation of the Septuagint. On the contrary, the name Μωϋσῆς (“Moses”) may well have been coined specifically for rendering the Hebrew name ‫“( ׁמׁשה‬Moses”) from two Egyptian words, mw designating “water,” and hsy meaning “to be glorified through immersion” in the context of the cult of Osiris (Pfeiffer 2016).

CONCLUSION Returning to our original question, what is the best way to produce a lexicon of an ancient translation? Although some issues are similar to producing a lexicon of any ancient language, the Septuagint offers some challenging particularities. Septuagint lexicography tends to oscillate between source-oriented study (i.e., focusing on the Hebrew or Aramaic behind the Greek text) and target-oriented study (i.e., reading the Septuagint as a work for its own, not only as a work received by Greek reader but also as a work produced by a Greek-speaking Jewish community). The interaction and tension between these two

SEPTUAGINT LEXICOGRAPHY 77

orientations are prevalent throughout Septuagint scholarship, but they should be held together. Study of Septuagint lexicography should certainly focus on the work of the ancient translators, never losing sight of the Hebrew/Aramaic Vorlage or ignoring possible phenomena produced by the various forms of interference and influence discussed above. Yet the translators did not work in a vacuum. Their work was produced at the complex intersection of scripture, language, and culture. For this reason also, the question of reception should not be ignored in Septuagint lexicographical research, whether examining influence upon later Septuagint translations, other Jewish Greek writings, or upon the New Testament and early Christian era. Studies in this field of research do not always lead to clear-cut results, but they often illuminate that most intractable of hermeneutical problems: the relation between language and thought.

RECOMMENDED RESOURCES 1. Hatch and Redpath (1998). The standard reference tool for a concordance of the Septuagint. This new edition contains a valuable introduction by Tov, as well as a Hebrew/Aramaic index to the Septuagint by Muraoka (but see below). 2. Muraoka (2010a). An improved version of the index provided in the new edition of HR (1998) discussed above. 3. Lust, Eynikel, and Hauspie (2015). A valuable lexicon devoted entirely to the Septuagint corpus—but without inclusion of external evidence—that is source-text-oriented, giving priority to the underlying Hebrew/Aramaic in cases of semantic uncertainty. It is based on the Rahlfs-Hanhart edition and provides translation equivalents 4. Muraoka (2010b). Another valuable lexicon devoted entirely to the Septuagint corpus—but without inclusion of external evidence—that is target language–oriented, with priority to the Greek linguistic context. It provides definitions rather than translation equivalents. 5. Bons (2020). This resource is the very first theological lexicon of the Septuagint. It is part of a broader project that offers comprehensive analysis of important words in the Septuagint, beginning with classical literature, including papyrological and epigraphical material, and also surveying its reception in Jewish and Christian work. 6. Jobes and Silva (2015). The second edition of this introductory book to the Septuagint offers a valuable and updated presentation of matters related to Septuagint lexicography, which is a good starting place for beginners (see 289–307). 7. Dorival (2016a). This essay is another valuable and recent overview of Septuagint lexicography. Although written in French, it is exhaustive and discusses the central tools, some case studies, as well as a valuable list of important lexicographical essays on the Septuagint. The present chapter is indebted to this essay. 8. Kraft (1972b) and Muraoka (1990). These books contain edited collections of essays and can be considered the starting points of contemporary lexicography on the Septuagint. 9. Joosten and Tomson (2007), Bons, Brucker, and Joosten (2014), Bons, Joosten, and Hunziker-Rodewald (2015), Bons, Pouchelle, and Scialabba (2019). These edited volumes contain a broad array of more technical and theoretical essays related to Septuagint lexicography, including numerous case studies.

78

CHAPTER 5

The Septuagint and Discourse Grammar CHRISTOPHER J. FRESCH

INTRODUCTION In the study of Septuagint translation technique, semantic and syntactic analyses have been the primary methods for comparing Hebrew Bible and Septuagint.1 Given the obvious import and effect of lexical and syntactic decisions in translation, it is no wonder that these two pillars of linguistic analysis serve as the foci of translation-technical work. Septuagint scholarship has benefited greatly from such studies, as they provide more insight into the work of the Septuagint translators and their theology, as well as have a bearing on text-critical decisions.2 However, language and communication cannot be reduced to semantics and syntax. There are other components of language that are necessary in order to achieve meaning and facilitate its successful communication. Since little has been done in Septuagint translation-technical studies outside of semantic and syntactic analyses, then, attention to those other components could undoubtedly complement the work that has been done by providing further insights into the translators, their work, and how they read and understood their Hebrew Vorlagen. One method of investigation that would serve this purpose is discourse-grammatical analysis.3 As an example, consider Gen. 11:26-27. Here an issue arises that cannot be sufficiently analyzed by the more traditional methods but is ripe for discourse-grammatical investigation.4 The Old Greek (OG) of Genesis mirrors the MT almost exactly. However, rather than rendering the conjunctive wāw in ‫“( ואלה תולדת‬and these are the generations”) with καί (thus: καὶ αὗται αἱ γενέσεις), the translator used δέ.5 This is not a negligible difference. As attested throughout the Greek translations of HB, καί is the stereotypical equivalent of conjunctive wāw (Aejmelaeus 1982: 29). The translator’s decision, then, not to render I wish to thank Steven Runge, Jacob Cerone, and Andrew Keenan for their comments on an earlier version of this paper. 2 On the translators see, for example, the numerous excellent studies of Anneli Aejmelaeus. Her work has been seminal in the field. On theology, see, e.g., Glenny (2009) and in the present volume. On textual criticism see, e.g., Tov (2015a). 3 See the next section for a description of discourse grammar. 4 All translations are the author’s unless otherwise noted. 5 There is also the difference in word order in ἔτη ἑβδομήκοντα (vs. ‫)שבעים שנה‬. This too is no small difference, and cases such as this merit further investigation. Consider Muraoka’s remark: “When a translator wishes, for whatever reason, to set himself free from the constraint of [the Hebrew text], he appears to opt for the sequence ” (2016: 470). However, there is significant manuscript evidence for the reading ἑβδομήκοντα ἔτη. 1

80

Gen. 11:25-27

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Καὶ ἔζησεν Θάρα ἔτη ἑβδομήκοντα καὶ ἐγέννησεν τὸν Ἀβρὰμ καὶ τὸν Ναχὼρ καὶ τὸν Ἁρράν. Αὗται δὲ αἱ γενέσεις Θάρα· Θάρα ἐγέννησεν τὸν Ἀβρὰμ καὶ τὸν Ναχὼρ καὶ τὸν Ἁρράν, καὶ Ἁρρὰν ἐγέννησεν τὸν Λώτ And Thara lived seventy years, and he fathered Abram, Nachor, and Harran. These δέ are the generations of Thara: Thara fathered Abram, Nachor, and Harran. And Harran fathered Lot ‫ויחי־תרח ׁשבעים ׁשנה ויולד את־אברם את־נחור ואת־הרן׃ ואלה תולדת תרח תרח הוליד‬ ‫את־אברם את־נחור ואת־הרן והרן הוליד את־לוט׃‬ And Terah lived seventy years, and he fathered Abram, Nachor, and Haran. Now these are the generations of Terah: Terah fathered Abram, Nachor, and Haran. And Haran fathered Lot

wāw with καί is itself significant. This is even before considering the choice to use δέ, which does not share the same functional overlap with conjunctive wāw as καί does. Moreover, δέ does not have a lexical equivalent in the Hebrew language (Fresch 2015: 70), so its use here cannot be tied to any lexical motivation stemming from the translator’s Vorlage. Neither can it be tied to syntactic necessity in the Greek, as asyndeton is just as appropriate syntactically here as δέ. Furthermore, because δέ is a postpositive, it necessarily inverts the word order of the Hebrew original. Given that the Septuagint translators attempted, generally speaking, to maintain the word order of their Vorlagen, it is significant when a translator makes a conscious decision (especially one that is not lexically motivated or syntactically required) that results in an alteration of that word order. Thus, even before considering the general proclivities of the translator of Genesis or how he renders the Hebrew phrase elsewhere, one is faced with a small difference in the Greek text that is by no means a trivial difference. The situation is not made simpler by examining the Greek translator’s general tendencies. Mark Scarlata (2015: 13) describes the OG translation of Genesis as maintaining “a very close lexical and syntactical relationship to the Hebrew parent text.” Owing to this and in addition to the above, one would have every reason to expect the translator to render the wāw with καί, thereby mirroring his Hebrew Vorlage both lexically and in terms of word order. Furthermore, if the other occurrences of ‫“( ואלה תולדת‬and these are the generations”) and ‫“( אלה תולדת‬these are the generations”) are examined, an answer is still not reached. Rather, one is faced with a translator who prefers δέ for the formula but will also use asyndeton (Gen. 2:4) or καί (Gen. 11:10; 25:19). The Septuagint scholar is thus left with a problem without a clear solution, at least as far as traditional methods of analysis are concerned.6

This problem can be seen more generally in Aejmelaeus’s comments on δέ and καί in the Pentateuch. Aejmelaeus, who has produced some of the best work on Septuagint translation technique, has provided useful statistics and commentary on how the Greek translators of the Pentateuch rendered conjunctive wāw. However, without a framework for analyzing the pragmatic difference between the two discourse markers, she is left to conclude with regard to the translators’ techniques that for the translators of Genesis and Exodus, “δέ was regarded as a normal equivalent of ‫ ו‬and was employed quite freely in various connections in keeping with natural Greek usage,” and for the translators of Leviticus–Deuteronomy, “δέ was considered to be neither a normal equivalent of ‫ ו‬nor an alternative for καί to be used in continuative contexts, but occurred to the translator in such cases only where the connection was felt to require some conjunction other than καί” (1982: 42–3). There is a meaningful choice to be

6

THE SEPTUAGINT AND DISCOURSE GRAMMAR 81

However, a discourse-grammatical analysis is able to make sense of the translator’s decision. The translator used δέ because he conceived of a discourse boundary between vv. 26 and 27, separating two discourse units. Discourse units are thematic groupings and boundaries between them tend to occur naturally, as the discourse builds and develops, in places of thematic discontinuity (Dooley and Levinsohn 2001: 35–7; Hoey 2001: 54–5). Consider the context around Gen. 11:27. The chapter begins with the Babel narrative. This is then followed by Shem’s genealogy (which itself is picking back up on and continuing material from Gen. 10:22-32). Gen. 11:26 completes the list of Shem’s descendants, ending with Thara and his sons. Verse 27 and following then proceed to summarize narratively the story of Thara’s household, namely that Abram and Nachor took wives (v. 29), Abram’s wife Sara was barren (v. 30), and Thara took Abram, Sara, and Lot on a journey to Canaan, only to settle in Charran (v. 31), where Thara eventually died (v. 32). In Genesis 12:1, the narrative theme-line is continued: Καὶ εἶπεν κύριος τῷ Ἀβράμ Ἔξελθε ἐκ τῆς γῆς σου καὶ ἐκ τῆς συγγενείας σου καὶ ἐκ τοῦ οἴκου τοῦ πατρός σου εἰς τὴν γῆν, ἣν ἄν σοι δείξω (“And the Lord said to Abram, ‘Go out from your country, your relatives, and your father’s house to the land that I will show you’ ”). Thus, between Gen. 11:26 and 27, one can detect a seam, a pivot within the discourse. There is a distinct shift in the content preceding and following v. 27. The material preceding v. 27 was a narrative event concerning not one person but humankind, and this was followed by a genealogy that resumed the prior genealogies of Noah’s sons, thereby tying off a major narrative unit and introducing the reader to new characters. The material contained in vv. 27–31 is the beginning of a new narrative block, providing crucial story elements that serve as the introduction to the Abram account. There is a distinct shift here at a macro-level, as the narrative moves from primeval history to patriarchal history. There is also a distinct shift at a more localized level. One can observe in v. 27 markers of discontinuity, which tend to cluster at discourse boundaries. First, there is the invocation of the tôlǝdōt formula, a formulaic boundary marker that is also a topicswitch device. Second, one can observe tail-head linkage (the repetition of a preceding sentence), a back-reference device that can serve as a point of departure at the beginning of a new segment of discourse (Levinsohn 2000: 280–1). These two features together make the transition from v. 26 to v. 27 a natural place to begin a new discourse unit. Furthermore, the function of δέ must be considered. As I have written elsewhere (Fresch 2015: 58): “[δέ] signals a new segment, a distinct information unit. By using δέ to segment, the communicator both clearly indicates the structure of the discourse and divides it into smaller meaningful units. This aids the reader in the building of their mental representation of the text, explicitly informing them of where to chunk information units.”7 Simply put, δέ is a discourse marker that signals distinct information units at macro- and micro-levels (Fresch 2015: 58–9).8 Therefore, given the movement of

made between using καί and δέ, but a statistical analysis without further linguistic investigation that asks questions beyond quantitative lexical representation is unable to comment on that choice, that is, why a translator uses καί or δέ in a given instance. Thus, Aejmelaeus has no recourse but to state that δέ is viewed as an equivalent to ‫ ו‬for some translators and not for others, because the statistical analysis looks only at lexical representation and not at the linguistic context and other non-lexical factors that affect the translation. As discussed in Fresch (2015: 28–73), the motivation to use δέ is based less on lexical representation and more on the translator’s conception of the movement and structure of the discourse. 7 Similar descriptions of δέ can be found elsewhere, such as in the historical grammarians, among some Classicists, and among some Koine Greek scholars. See the discussion in Fresch (2015: 59–67). 8 A discourse marker is an inherently functional device that instructs the recipient on how to process and put together the discourse (Fresch 2015: 3–5).

82

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the narrative and the markers of discontinuity discussed above, this occurrence of δέ fits nicely. It explicitly structures the discourse by marking the movement to a new discourse unit, which, in this case, is a major development. When a reader processes this δέ and its context, they are alerted that a break has occurred within the discourse, that v. 27 and the material following are a distinct unit from what preceded, and, therefore, that they need to build their mental representation of the discourse accordingly.9 This insight is particularly informative when it comes to the study of the translator’s technique. The particle καί could have been used. Not only would it have lexically represented the conjunctive wāw, but it would have also been the “easy technique” (Aejmelaeus 2007b: 61, 69). However, καί would not indicate a new, distinct segment of discourse. It would not encourage the reader to build their mental representation by cognitively chunking the information.10 Therefore, given this and given that δέ is employed rather than καί, it would seem to be the case that the translator of LXX Genesis considered lexical representation and strict adherence to the word order of his Vorlage to be less important, at least in this instance, than representing his conception of the discourse structure. His conception of the discourse could have been informed by the tôlədôt formula, the tail-head linkage, the topic shift, the change from genealogy to narrative, the movement of the broader discourse into the Abram narrative, the fact that there is a break in the wayyiqtōl chain that leads into v. 27, or, most likely, a combination of some or all of these features.11 The clustering of these features in HB, particularly with the break in the wayyiqtōl chain, marks a transition in the discourse. What is crucial to realize, then, is that the translator of LXX Genesis based his work on more than just semantic and syntactic representation. Whether conscious of it or not, he allowed his understanding of the wider context, the pragmatic features of his Hebrew text, and his conception of the discourse structure to inform his translational decisions—δέ could not have been motivated otherwise.12 This example illustrates the need for linguistic analysis beyond semantics and syntax in studies of Septuagint translation technique. Semantic and syntactic analyses are crucial, but they are not always sufficient. Language and communication are more complex. The flow of a discourse, its structural makeup, how one utterance leads into another, and how utterances synergistically combine to create larger meaningful units are just as integral to language and how humans communicate as semantics and syntax. Scholarship stands to benefit from utilizing additional methods of linguistic analysis that examine such features of language. Discourse-grammatical investigations will complement existing work on

This is similar to the way that distinct information units are signaled in modern writing and typography by new paragraphs or line breaks separated by asterisks. 10 In this context in particular, the use of καί would result in requiring more cognitive effort on the part of the reader, as they attempt to hold the information together despite thematic discontinuity. 11 I say “most likely” given that the tôlədôt formula on its own was not enough to motivate the use of δέ. As mentioned above, the formula in Gen. 2:4 is introduced by asyndeton and its occurrences in Gen. 11:10 and 25:19 are introduced by καί. There is not space here to discuss these but suffice it to say for our present purposes that the asyndeton in Gen. 2:4 represents the asyndeton in Hebrew as well as fits the structure of the discourse. The καί in 11:10 and 25:19 both serve to connect their host utterances to previous, related content (the material following from 11:10 and 25:19 is not “new” in the same way that material following from the other instances of the tôlədôt formula is). 12 John William Wevers (1993b: 157) possessed keen insight into the discourse structure and the significance of δέ: “That an entirely new section begins here is clear both from the formula which introduces this section: ‘these are the generations of Thara,’ and from the repetition of the fact that Thara generated Abram and Nachor and Harran; [the translator] also makes this clear by the use of δέ.” 9

THE SEPTUAGINT AND DISCOURSE GRAMMAR 83

translation technique and result in fuller understanding and more robust descriptions of the Septuagint translators and their work.

Discourse Grammar and Septuagint Scholarship Discourse grammar is a method that applies a pragmatic, or functional, analysis to grammatical phenomena. As such, discourse grammar investigates how the functions and the uses of various language features convey meaning and contribute to the developing discourse. Such a method is a natural and necessary part of linguistic analysis. As Ronald Langacker (2008: 492) writes, There is no exaggeration in saying that all of grammar is shaped by discourse and only exists to make it possible. It is atypical for the structures examined in grammar—such as phrases, clauses, and even sentences—to be used in isolation. Normally they occur as integral parts of longer discourse sequences that provide the reason for their being assembled and assuming the form they do. Fundamental grammatical notions can be characterized in terms of their discourse function. Discourse considerations motivate grammatical choices.13 This reality is foundational to discourse grammar, and Steven Runge (2010: 5) encapsulates its significance well when he states, “One of the key presuppositions of discourse grammar is that choice implies meaning.” Or, in the words of Robert Longacre and Shin Hwang (2012: 15), “The discourse analyst rejects the assumption that variety simply occurs for variety’s sake. The counter assumption is that variation in form is a choice made by the speaker or writer with a certain end in view … Any variation in form can be studied with a view to what it implies.” It is these choices, made on account of their function within a discourse, and the meanings they convey that are the object of discourse-grammatical study. Rebecca Hughes and Michael McCarthy (1998: 271) write, “A discourse grammar … foregrounds the kinds of choices that speakers and writers routinely deal with in production—that is, how can one best formulate a message to make it clear, coherent, relevant, appropriately organized, and so on?” To put it simply, whereas semantics asks, “What does this mean?” and syntax asks, “How should this be structured?” discourse grammar asks, “How does this function and why is it here?” While it is not eminently clear why discourse grammar has not been utilized in Septuagint research, there are a few factors that have likely contributed to the lack of engagement. First, there are two issues that arise from linguistic scholarship. One is that, though there is a general consensus within linguistics over the method of such an analysis and its object of study, there is no standardized nomenclature for the method itself. One can find terminology that ranges from the pithy—“discourse grammar” (Hughes and McCarthy 1998; Runge 2010; Paltridge 2012: 113–43), “discourse pragmatics” (Lambrecht 1998: 2–5), or “functional grammar” (Buth 1995)—to the cumbersome—“grammar and discourse function” (Langacker 2008: 491), “discoursefunctional approach to grammatical phenomena” (Cumming, Ono, and Laury 2011: 8), “grammarians who do discourse analysis” (MacDonald 1992: 156). The other issue is that, whereas there is no lack of useful resources that introduce, describe, and provide bibliographies for methods of semantic or syntactic analyses, such resources are few and far between for discourse grammar. Thus, the inconsistent terminology and the dearth So also Hoey (2001: 61).

13

84

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

of the kinds of resources for which a specialist in a different field would look have likely created a barrier of entry for LXX scholars (or even for biblical scholars more generally). Second, some within Septuagint scholarship have claimed that the Septuagint translators were not aware of the linguistic context as they went about the task of translation. For example, Anneli Aejmelaeus (2007c: 84) states, The translator … had to concentrate on the few words he was translating. It has been discovered that the translators were often blind even to the most immediate context, so that they could leave in it structural inconsequences [sic], which they did not later return to correct … the range of vision of the translator at work was very limited. Such a perspective would prevent linguistic approaches that are predicated on the interaction between grammatical phenomena and contextual considerations, such as discourse grammar. This position, however, is untenable, as it disregards the necessity of contextual awareness to translation (Fresch 2015: 19)—a necessity that was noted by James Barr (1979: 22–3) when he wrote, Generally speaking, it is not possible in any text, in any language, to make even basic identifications of words without some attention to their context, which is the sole resource available to select between the multiple possible values of the signs … The [Septuagint] translator was commonly not able to make his basic diagnosis word for word. Even the literalist had to work by the context, as the freer translator did. There is ample evidence throughout the Greek OT that supports this argument. One example, as I have argued elsewhere (Fresch 2015), is the very use of discourse markers, that is, function words and phrases such as δέ. Discourse markers function to incorporate their host utterances (i.e., the utterances in which they appear) into the developing discourse. However, since discourse markers in the Greek OT are often not lexically motivated by an equivalent in the source text, they could not be used appropriately without the translator having some conception of the structure and flow of the discourse. Thus, while one can certainly find oddities in the Greek OT owing to interference from the source language or a translator who did not understand his source in a given instance, the nature of language and of translation should caution us against then making broad claims by extension about the translators’ competency or method. The default assumption when approaching the work of the Septuagint translators should be, based on the nature of language and translation, that they had at least some awareness of context (Fresch 2015: 20–1). Third, the assumptions of the Interlinear Paradigm, a school of thought within Septuagint scholarship, do not naturally motivate one to engage in discourse grammar. In the introduction to A New English Translation of the Septuagint, a translation based on the Interlinear Paradigm, Albert Pietersma and Benjamin Wright (2009: xiv) state concerning the Septuagint, The Greek had a dependent and subservient linguistic relationship to its Semitic parent. Or again, although the Septuagint was a translation of the Bible, it did not thereby automatically become a biblical translation. More particularly, for the vast majority of books the linguistic relationship of the Greek to its Semitic parent can best be conceptualized as a Greek interlinear translation of a Hebrew original within a Hebrew-Greek diglot. If one takes the metaphor of an interlinear seriously and approaches the work of the Septuagint translators with that preconceived notion, then there is little reason to consider context and discourse function. A translator intent on producing half of an

THE SEPTUAGINT AND DISCOURSE GRAMMAR 85

interlinear text would arguably default to verbal isomorphism in most cases. In other words, because of the nature of the task, discourse considerations would largely not drive grammatical choices.14 If one assumes the text they are investigating is a conceptual interlinear, then they are assuming a text that is often unintelligible (Pietersma 2010: 19) and that was produced with the goal of bringing the reader to the source text rather than the source text to the reader (Wagner 2013: 3), that is, rather than producing a translation qua translation. This understanding of the process of translation leaves little place for contextual awareness. Lastly, when describing the translation technique of the Septuagint translators, terms such as “literal” and “free,” or “formal” and “functional,” are often used, even if it is acknowledged that the terms themselves are problematic (Dines 2004: 119–21). However, discourse features in translation resist such categorization. For example, if there is a thematic break in a narrative and it moves to a new scene, this is often reflected in Hebrew by interrupting a wayyiqtōl chain and explicitly marking a topic shift with a nominal constituent placed before its verb. Greek, on the other hand, does not have wayyiqtōl chains that can be interrupted for pragmatic effect, but Greek does explicitly mark topic shifts by placing the new topic before its verb, like Hebrew. Greek also has a variety of means by which to indicate new units of discourse, some of which are mirrored in Hebrew and some of which, like δέ, are not.15 A Greek translation that represents a movement to a new scene in this way cannot be satisfactorily described as literal or as free simply construed. It is “free” in the sense that the translator did not feel restricted by his Vorlage. Yet it is “literal” in the sense that the translator is attempting to represent formal features of his Vorlage, but these features are pragmatic in nature, rather than semantic or syntactic. An incongruity exists between discourse-driven choices and the literal/free spectrum. Therefore, if one begins analysis with an a priori assumption that the data are most usefully classified in terms of a literal/free spectrum, then there will be no engagement with discourse grammar, as the answers it provides do not easily fall within that spectrum. In sum, whether because of the inconsistent terminology or lack of accessible material when it comes to discourse grammar, a conviction that the Septuagint translators were generally not aware of context as they translated, the belief that the Septuagint translations are best understood as interlinear translations, an a priori assumption that the data are best categorized on a literal/free spectrum, or some combination of these, Septuagint scholarship has largely not engaged in discourse-grammatical analysis. In what follows, then, I hope to demonstrate the usefulness of discourse grammar to the study of the Septuagint and provide a path forward for any who would take up the task.

APPLYING DISCOURSE GRAMMAR TO THE STUDY OF THE SEPTUAGINT There are many applications of discourse grammar. The examples given here are not intended to be exhaustive but rather to exemplify the method.

It bears noting that Pietersma seems to struggle balancing this dichotomy. On the one hand, he views a high degree of isomorphic transfer as constitutive of the Interlinear Paradigm (2010: 18–19). On the other hand, he (2002: 350) argues, “What is meant by subservience and dependence is not that every linguistic item in the Greek can only be understood by reference to the parent text, nor that the translation has an isomorphic relationship to its source, but that the Greek text qua text has a dimension of unintelligibility.” For more discussion of the interlinear paradigm and its application in the SBLCS, see the chapter by Hiebert in the present volume. 15 For an example, see Jon. 3:3. See also the discussion in Fresch (2015: 46–8). 14

86

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Discourse Markers As illustrated earlier, discourse markers are motivated by discourse considerations. A discourse marker provides “instructions to the hearer on how to integrate their host utterance into a developing mental model of the discourse in such a way as to make that utterance appear optimally coherent” (Mosegaard Hansen 2006: 25). They can be likened to linguistic road signs (Mosegaard Hansen 1998: 199), in that “they aid the reader in navigating the discourse, informing them of the structure of the text, alerting them to what is coming, and providing them with instructions on how to proceed” (Fresch 2015: 4). As such, they are a natural object of study for discourse grammar. Greek has no shortage of discourse markers. Whether conjunctions such as δέ, μέν, ἀλλά, πλήν, and οὖν (to name a few), particles like ἰδού, or phrases such as καὶ ἐγένετο, a Greek speaker had a plethora of choices for indicating the structure and flow of a discourse. Consider, in addition to the earlier investigation of δέ in Gen. 11:27, another example of δέ and one of οὖν that demonstrate how attention to discourse markers complements Septuagint study. In Job 1:20-22, the narrator tells of Job’s mourning over his losses, quotes Job’s faithful response to the Lord, and then sums up by commenting on how, despite the terrible events that transpired, Job did not sin. In 2:1, there is a distinct shift. Job is no longer the center of our attention. Instead, the scene changes to heaven and the main characters are the Lord and the slanderer. (1:22) In all these things that happened to him Iob did not sin at all before the Lord, and he did not charge God with folly. (NETS) Job 2:1

Ἐγένετο δὲ ὡς ἡ ἡμέρα αὕτη καὶ ἦλθον οἱ ἄγγελοι τοῦ θεοῦ παραστῆναι ἔναντι κυρίου, καὶ ὁ διάβολος ἦλθεν ἐν μέσῳ αὐτῶν. Now, it happened, when it was this day and the angels of God came to present themselves before the Lord, the slanderer also came among them. ‫ויהי היום ויבאו בני האלהים להתיצב על יהוה ויבוא גם הׂשטן בתכם להתיצב על יהוה׃‬ Now, the day came when the sons of God came to present themselves before YHWH, and the accuser also came among them to present himself before YHWH.

In narrative, thematic discontinuity arises when there is a change in participants, time, place, and/or action (i.e., event type) (Dooley and Levinsohn 2001: 37). It is natural when thematic discontinuity occurs, particularly in contexts of high discontinuity, to signal a new chunk of discourse, to explicitly segment it, given the break between thematic groupings (Dooley and Levinsohn 2001: 34–7). This has the benefit of easing the cognitive processing task for the reader as they attempt to comprehend the discourse and build their mental representation of it (Fresch 2015: 54–8). Note that between Job 1:22 and 2:1, there is a change of participants, a change in time, a change in place, and a change in event type. Given the high thematic discontinuity, there is good reason to regard 2:1 as the beginning of a new thematic grouping. Thus, in order to signal the break between discourse units, the translator of LXX Job uses δέ. Between the discourse marker and the explicit thematic discontinuities, the reader would know to regard 2:1 as

THE SEPTUAGINT AND DISCOURSE GRAMMAR 87

the beginning of a new scene in the narrative. The use of δέ here makes sense and reflects good, idiomatic Koine. However, δέ is not lexically motivated by the underlying Hebrew. One could argue that it represents conjunctive wāw quantitatively, but even then, the use of δέ has resulted in an inversion of the word order. This is not to say, though, that δέ is not motivated at all. If one looks to the features of the Hebrew as well as the context, a defense can be made for a translator who was attempting a faithful representation of his Vorlage. Contextually, in both the Hebrew and the Greek, the same thematic discontinuities arise between Job 1:22 and 2:1. Moreover, in Job 2:1, the MT begins with ‫ויהי‬. Although ‫ויהי‬ is formally a verb and can function as such, it often functions as a discourse marker that, as van der Merwe, Naudé, and Kroeze (2017: 429) state, “typically signals, on the one hand, that an event or scene follows on a preceding event or scene and, on the other hand, that the reference time of the event or scene is being updated (and by implication specified).” While ‫ ויהי‬and δέ are not co-extensive in function, they do share a significant amount of overlap. Thus, it would seem that an awareness of the discourse function of ‫ ויהי‬and an attention to the surrounding context motivated the translator to use δέ. There would be no reason to use δέ instead of καί otherwise. This observation provides us with insight into the translator of LXX Job. Without applying discourse grammar, this δέ looks like a free translation and mere “stylistic preference” (Cox 2015: 390). Such an analysis accords well with the recognition of this translator’s rather free translation technique (Cox 2015: 385), but it does not give adequate consideration to the function of δέ in Koine nor to what contextual factors may have motivated its use. This approach perhaps demonstrates the weakness of a literal vs. free assessment, in that it does not naturally consider pragmatic options available to the translator or the representation of discourse features in the process of translation. Uses of δέ such as this one, when analyzed according to their discourse function, indicate neither a “free” nor “literal” technique, but rather a translator who attempted to faithfully represent, in conventional Koine idiom, the discourse features of his Vorlage. In Exod. 1:9, the Pharaoh addresses the people of Egypt and raises concerns about the growing populace of the sons of Israel. There is then a shift that takes place in his discourse from v. 9 to v. 10: (v. 9) Now he said to his nation, “Look, the race of the sons of Israel is a great multitude and is becoming stronger than we.” (NETS) Exod. 1:10

δεῦτε οὖν κατασοφισώμεθα αὐτούς, μήποτε πληθυνθῇ καί, ἡνίκα ἂν συμβῇ ἡμῖν πόλεμος, προστεθήσονται καὶ οὗτοι πρὸς τοὺς ὑπεναντίους, καὶ ἐκπολεμήσαντες ἡμᾶς ἐξελεύσονται ἐκ τῆς γῆς So, come on! Let’s outwit them, lest they multiply and, whenever war befalls us, these ones are also added to our enemies and, after warring against us, they depart from the land ‫הבה נתחכמה לו פן ירבה והיה כי תקראנה מלחמה ונוסף גם הוא על ׂשנאינו ונלחם בנו‬ ‫ועלה מן הארץ׃‬ Come, let’s deal wisely with them, lest they multiply. If a war happens, they will also be added to our enemies and will fight against us, and they will leave the country

88

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Note that οὖν is used in v. 10. This discourse marker, like δέ, instructs its reader to build their mental representation of the discourse by regarding what follows as a new, distinct segment. However, unlike δέ, οὖν also explicitly indicates a continuative connection with relevant preceding material (Wakker 2009: 80; Fresch 2016: 459, 462– 3). The occurrence in Exod. 1:10 is an excellent example of its prototypical usage. In v. 9, Pharaoh comments on how great and strong the sons of Israel have become. Then, in v. 10, there is a topic switch from the sons of Israel to the people of Egypt (“us” in “Let us”) and a change in the kind of action (from indicative to cohortative). Such shifts create thematic discontinuity, which often leads to breaks between thematic groupings (Dooley and Levinsohn 2001: 37). However, breaks between thematic groupings are not always so large in scope as a scene change or the movement to a new story (as with the thematic groupings introduced by δέ in Job 2:1 and Gen. 11:27). They can occur at more localized levels of discourse, particularly in contexts where the discontinuities are not as strong, such as here in the Pharaoh’s speech in Exod. 1:9-10. By using οὖν, the translator recognizes the thematic discontinuity and marks what follows as a new, distinct unit but one that is closely connected to what preceded. This choice makes good sense. The topic switch and the shift to a cohortative represent a stark development in Pharaoh’s speech, but it is one that naturally flows from the content of his speech recorded in v. 9. As I have observed with regard to the use of οὖν in LXX Genesis and Exodus (Fresch 2016: 462–3), It is natural that half of the occurrences of οὖν (thirty-five out of seventy) co-occur with a shift to an imperative (change of kind of action), as such a shift accords well with a move to the next stage or step of the discourse while also often requiring the previous discourse to be kept in mind for the justification of the command. The situation in Exod. 1:9-10 is no different, save for the use of a cohortative rather than an imperative. Thus, given the function of οὖν and its use in Exod. 1:10, one can observe that it aids the reader in processing the discourse by signaling the very structure and flow of the discourse. The discourse marker indicates that its host utterance and what follows is a new thematic grouping that is connected in some respect to prior material. However, despite the presence of οὖν, it was not motivated by the underlying Hebrew. Verse 10 in the MT begins with ‫“( הבה נתחכמה‬Come, let’s deal wisely”). There is no lexeme in the Hebrew for οὖν to qualitatively or even quantitatively represent. Had the translator been primarily concerned with representing the individual words of his Vorlage, or if he had no awareness of context, he would not have used οὖν. Given the presence of οὖν, then, one would do well to ask what motivated its use here. It would seem that the translator, aware of the shift in Pharaoh’s speech from a declaration of the problem to calling on his people to join him in dealing with that problem, felt that v. 10 constituted a new development in Pharaoh’s discourse, albeit one that maintained some connection to the preceding material. By using οὖν, the translator makes this explicit, thereby guiding his readers along the same mental route. As such, we are given insight into the translator and how he went about his work. In this instance, he maintains a close connection to his Vorlage but feels the freedom to allow discourse considerations to influence his choices, even if it results in a particle that does not represent a lexeme in the Hebrew. The resulting translation, as it concerns οὖν, is certainly not “literal” but neither is it “free.” If one considers the context and the discontinuities between v. 9 and v. 10, a case can be made for a translator who was attempting to faithfully represent his reading of his source text.

THE SEPTUAGINT AND DISCOURSE GRAMMAR 89

Information Structure Another area of discourse grammar is information structure. Information structure is concerned with how propositions are structured in terms of what is presupposed and asserted (Lambrecht 1998: 6). It is a complex topic that deserves much more space than can be allotted here.16 However, to simplify it to its most practical outworking, information structure is the investigation of how variations in word order convey meaning. Koine has a default, or most neutral, word order of Verb-Subject-Object (abbreviated “VSO”; Levinsohn 2000: 16–7; Horrocks 2010: 108–9). However, as is typical of languages cross-linguistically (Dik 1997: 403, 408–10, 420–7) and as has been demonstrated in Koine (Levinsohn 2000: 1–67; Runge 2010: 181–313), certain constituents may be placed before the verb for the purpose of indicating prominence, whether topical or focal. In other words, to draw attention to the current sentence topic or to the most salient piece of information, a Koine speaker or writer could place those constituents before the verb. When it comes to the Greek OT, then, attention to information structure provides insight into how the translator regarded the nature of the information in a given proposition. Consider Ruth 3:2-3, wherein Naomi is speaking to Ruth about Boaz and then proceeds to give her instructions: (v. 2) Now, is not Boos our acquaintance, whose girls you were with? Look, he is winnowing the threshing floor of the barley this very night. (NETS) Ruth 3:3

σὺ δὲ λούσῃ καὶ ἀλείψῃ καὶ περιθήσεις τὸν ἱματισμόν σου ἐπὶ σεαυτῇ καὶ ἀναβήσῃ ἐπὶ τὸν ἅλω· μὴ γνωρισθῇς τῷ ἀνδρὶ ἕως τοῦ συντελέσαι αὐτὸν πιεῖν καὶ φαγεῖν Now, bathe, put perfume on, put your clothing on yourself, and go up onto the threshing floor. Do not make yourself known to the man until he finishes drinking and eating. ‫ורחצת וסכת וׂשמת ׂשמלתך עליך וירדתי הגרן אל תודעי לאיׁש עד כלתו לאכל ולׁשתות׃‬ So, bathe, put perfume on, put your outer garment on yourself, and go down to the threshing floor. Do not make yourself known to the man until he finishes eating and drinking.

At the beginning of v. 3, the pronoun σύ is positioned before the verb. This is done to indicate topical prominence. In the previous verse, the topic of Naomi’s sentences is Boaz—he is what the sentences are about. However, in v. 3, Boaz is no longer the topic. Instead, the sentences are about Ruth. Thus, to explicitly indicate this switch of topics and to draw attention to the new topic, the translator places σύ in a position that marks topical prominence. Hebrew, like Greek, is a VSO language that allows pragmatic positioning before verbs.17 Also like Greek, it encodes the subject of the verb into the verb form itself, but

I recommend the interested reader to start with the discussions in Runge 2010: 181–313, and Levinsohn 2000: 1–67, and then to move to Lambrecht 1998. 17 See the excellent study in Moshavi (2010), as well as the overview of pragmatic ordering in van der Merwe, Naudé, and Kroeze (2017). 16

90

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

can use an explicit constituent if it is felt that one is necessary. Because of this overlap between the two languages, the books of the Greek OT often mirror the word order and constituents of their Vorlagen.18 Thus, when such mirroring does not occur, it merits investigation. In Ruth 3:3, the MT begins with ‫ורחצת‬. There is no preposed subject and, therefore, there is no reason, as far as the lexemes themselves are concerned, for the translator of LXX Ruth to have used σύ. It would thus seem that the translator, aware of the topic switch in Naomi’s discourse, felt that an explicit preposed constituent was merited. Note also that the translator uses δέ instead of καί. These two features go handin-hand. Between vv. 2 and 3, there are discontinuities of topic and kind of action. By using δέ, the translator marks a new, distinct discourse segment. By preposing σύ, the translator establishes the topic for the following discourse. These choices shed light on a translator who was translating not just the words before him but was making decisions based on the nature of the information relative to the discourse and with respect to the structure of the discourse.

Historic Present The historic present is a verb form that breaks all of the rules. It is a present imperfective verb that occurs in narrative contexts where a past perfective verb is expected (Robar 2016: 329, 331). As such, it is a pragmatic device meant to grab the reader’s attention. According to Levinsohn (2000: 200–13), the historic present highlights either the speech or act to which it refers or, more often, the event that follows the speech or act to which it refers. With regard to highlighting generally, Levinsohn (2000: 197) writes, “Sentences are typically highlighted when they relate to a climax or when a particularly significant development or a change of direction occurs in the story.” Thus, their occurrences in the Septuagint, particularly if not motivated by a similar verb form in the underlying Hebrew, provide insight into what a translator considered climactic or significant in a story. For instance, in Exod. 2:5, the Pharaoh’s daughter goes down to the Nile to bathe and finds a basket floating in the river. In v. 6, an historic present is used to highlight following material: Exod. 2:6

ἀνοίξασα δὲ ὁρᾷ παιδίον κλαῖον ἐν τῇ θίβει, καὶ ἐφείσατο αὐτοῦ ἡ θυγάτηρ Φαραὼ καὶ ἔφη Ἀπὸ τῶν παιδίων τῶν Ἐβραίων τοῦτο Now, when she opened it, she saw a child crying in the basket, and Pharaoh’s daughter spared him and said, “This is one of the Hebrews’ children!” ‫ותפתח ותראהו את הילד והנה נער בכה ותחמל עליו ותאמר מילדי העברים זה׃‬ She opened it and saw him, the child. A boy crying! She had compassion on him and said, “This is one of the Hebrews’ children!”

By using ὁρᾷ, a present imperfective verb in a past perfective context, the translator grabs his reader’s attention, indicating a forthcoming significant development. While it could

Given this, despite the preponderance of claims within Septuagint scholarship to the contrary, it is often expected and unsurprising when a Greek translation mirrors the word order of its Vorlage.

18

THE SEPTUAGINT AND DISCOURSE GRAMMAR 91

be argued that the significant development, from the perspective of Pharaoh’s daughter, is seeing a child in a basket, such is not news to the reader of Exodus. What is more significant to the reader, given Pharaoh’s decree in Exod. 1:22 that every male child born to the Hebrews be thrown into the Nile, is that she spared the boy. Because of this and based on Levinsohn’s argument that historic presents usually point to the event after the speech or act to which they refer, it is reasonable to posit that ὁρᾷ points beyond itself and highlights the action Pharaoh’s daughter takes after seeing the child. The choice to use an historic present here is all the more significant when the underlying Hebrew is considered. The verb translated by ὁρᾷ is ‫ותראהו‬, a wayyiqtōl (with an object suffix that is not present in LXX Exodus) that expresses a past perfective action, exactly what is expected in context. The translator could have formally and suitably represented ‫ ותראהו‬with the aorist indicative εἶδεν. Owing to this, one is forced to ask what motivated the historic present. The only sensible answer is that the translator was aware of the forthcoming significant development in the narrative and considered it important enough to highlight, even though there is no such highlighting in the underlying Hebrew. This certainly provides further support for Aejmelaeus’s (1982: 180) claim that the translator of LXX Exodus was “the most competent of the Pentateuchal translators” whose work provides “adequate evidence of his consideration of the demands of the Greek language and mastery of large units of texts.” However, features such as this, should a sufficient amount of them be found and examined, could paint a fuller picture of a translator who felt free to employ pragmatic devices in conventional Koine idiom and thereby, at least in these instances, produced a text that is more than a translation but is also, in part, his own literary creation.

The Benefits of Discourse Grammar for the Study of the Septuagint By utilizing discourse grammar, we are able to flesh out more fully linguistic analyses of the books of the Greek OT. It adds a formal means of investigating pragmatic features of language that are motivated by discourse considerations and that contribute to the meaning of a text just as much as semantic and syntactic features. A discourse-grammatical analysis therefore further supports the work of Septuagintalists, providing them with different kinds of questions to explore. Moreover, by engaging in discourse grammar, we are treating the books of the Greek OT as texts in their own right and reading them as such. That is to say, by investigating the structure and flow of a discourse, by asking how discourse considerations motivated certain choices, we are reading the text qua text. This approach, then, before asking questions of translation technique, encourages us to read and interpret the Greek OT independently and in a more robust manner. The study of translation technique, in particular, stands to benefit. Discoursegrammatical analysis adds a dimension to the study that typically has not been present, at least not formally. It complements the work that has been done, thereby providing more detailed pictures of the translators, their work, and how they understood the flow of the discourse and the structure of the text. The more complete a study of translation technique is, the better one is able to then speak to issues of a Septuagint translator’s theology or to issues of textual criticism.

92

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

ONGOING RESEARCH QUESTIONS The examples provided above are merely a sampling of discourse grammar and its application to the study of the Septuagint. Little has been done in this area. Whether starting from the Greek and investigating a certain discourse feature or starting from a discourse feature in Hebrew and investigating how a translator, or if a translator, renders it, there are a plethora of questions to ask—too many to try to list here. What is crucial to understand is that this is an interdisciplinary task. In addition to one’s knowledge of primary and secondary sources in Greek, Hebrew, Hebrew Bible, and Septuagint, one must read widely and deeply in linguistics. This will provide the theoretical foundations and cross-linguistic insights necessary to produce linguistically viable and well-formed analyses. This is all the more crucial because the applications of discourse grammar and of the methodologies intrinsically connected to it—that is, cognitive linguistics and functional linguistics—to Koine Greek and Biblical Hebrew are barely out of their infancy. As such, it is not for everyone in biblical studies. However, for those who are interested in and excited by the Septuagint and who are energized at the thought of immersing themselves in and engaging with modern linguistic scholarship, they will find here meaningful work. There is an entire corpus to explore and innumerable questions to ask of it.

RECOMMENDED RESOURCES 1. Aejmelaeus (2007a). This is a good starting place for thinking about the study of Septuagint translation technique. While it approaches the study only in terms of semantics and syntax, Aejmelaeus’s scholarship is among the best in the field. 2. Dooly and Levinsohn (2001). Dooley and Levinsohn’s manual provides a helpful first step into discourse grammar. It teaches one how to think linguistically about discourse and provides an introduction to the sorts of questions one should ask of a discourse. 3. Fresch (2015). My dissertation demonstrates the benefits of a discourse-grammatical analysis to the study of the Septuagint. 4. Levinsohn (2000). This is an excellent introduction that is designed to teach its reader how to do discourse-grammatical analysis in Koine texts. It covers a wide variety of topics. 5. Runge (2010). This book is similar to Levinsohn’s above, though Runge is more accessible. It does not cover as many topics, but it has more detailed explanations and more numerous examples. I would encourage the interested reader to pick up Runge first and then read Levinsohn. 6. Hughes and McCarthy (1998). This is a helpful introduction to the concerns of discourse grammar and why it is necessary.

CHAPTER 6

The Septuagint and Greek Style EBERHARD BONS

The style of the biblical writings in general is a topic that has occupied readers of the Bible ever since antiquity.1 As for the Septuagint in particular, it cannot be denied that philological and exegetical research has made great progress in the last decades. In fact, numerous studies have been published since the 1980s, including research tools like lexicons as well as commentaries and studies dealing with various issues such as vocabulary, textual criticism, text history, translation technique, and theological questions. However, in-depth investigations dealing with style and rhetoric of the various writings of the Septuagint are still in their infancy (Léonas 2016: 373). This chapter addresses the challenges involved in such investigations, introduces key methodological considerations, and provides examples to point toward further research.

INTRODUCTION In his Prologus in libris Salomonis, Jerome, translator and commentator of the Bible (347–420 CE), observes with regard to the Wisdom of Solomon: apud Hebraeos nusquam est, quin et ipse stilus graecam eloquentiam redolet: “[this book] is not contained in the Hebrew writings [of the Bible]; rather, its style itself is redolent of Greek eloquence” (Weber 1994: 957). Modern commentators agree that the Wisdom of Solomon, one of the latest books of the Septuagint, was written in Greek from the beginning, the author being familiar with Greek style and rhetoric (Aitken 2015d: 404; Mazzinghi 2019: 18– 19). But the situation is completely different for the books of the Septuagint translated from Hebrew or Aramaic source texts. Once again, some of Jerome’s reflections enable us to gain a better understanding of the issue at hand. Skilled himself in ancient rhetoric, the Stridonite argues that the style of most of the books of the Bible does not meet the expectations of rhetorically educated readers whose mother tongue was either Greek or Latin. Blinded by their own elocutionary abilities, as he maintains, they are far from appreciating the obvious simplicitas of biblical texts. Rather than judging the value of Holy Scripture on the basis

I wish to express my sincere thanks to my colleagues with whom I was able to discuss several aspects of this chapter: Prof. Dr. Christoph Kugelmeier (Saarbrücken), Dr. Antonella Bellantuono (Strasbourg), Dr. Daniela Scialabba (Rome), and Laura Bigoni (Strasbourg).

1

94

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

of the “majesty of ideas” contained in it, these readers reject it because of its allegedly low stylistic quality.2 In the prologue to his translation of the book of Jeremiah, Jerome goes even further when seeking to explain the origin and the raison d’être of the simplicitas of biblical texts. His judgment seems quite harsh, on the one hand, when he concludes that—compared with Isaiah, Hosea, and other prophetic books—the book of Jeremiah appears to be less artful, more “rustic.” On the other hand, he claims that the rhetorical qualities of the language of a given biblical book are not decisive in themselves. Yet, on the level of ideas, as Jerome emphasizes, the book of Jeremiah is on par with the others already mentioned, since it is the same spirit that prompted the prophet to foretell the future. However, the simplicitas of his language is due to the fact that Jeremiah was born and grown up in a rural milieu, which probably means that he did not have the appropriate rhetorical training (Weber 1994: 1166). The simplicitas as a striking feature of Holy Scripture, as opposed to eloquium both of biblical authors and of the ancient translators of the Bible, is a kind of topos in Jerome’s writings.3 He even goes as far as saying that the raison d’être of the simplicitas lies in the fact that the message of the Bible as a whole was meant to be understandable for both educated and uneducated people. This applies not only to the original texts, whose putatively simple language was taken to be intentional on the part of the authors, but also to the translations of the Bible available in his time.4

METHODOLOGICAL CONSIDERATIONS If the simplicitas is an overarching feature of the biblical writings, what about the rhetorical and stylistic qualities of this first translation of the texts that became included in the Hebrew Bible? Several methodological considerations can help answer this question.

Hebraisms and Translation Greek The Septuagint, specifically the books translated from Hebrew or Aramaic, is notorious for its countless Hebraisms on different levels and its so-called translation Greek. As has been pointed out by various scholars in the last decades (e.g., Mussies 1976: 1048; Walser 2016: 224–5), Septuagint syntax is well known for phenomena like its paratactic style (Usener 2011: 44), the rareness of participle constructions like the genitive absolute, the use of nominal clauses, and the predominantly word-for-word correspondence with each element of the source text (e.g., enclitic pronouns). As for Septuagint vocabulary, the translators were inclined to render some typical features of the Hebrew language more or less literally, which results in expressions like εἰμί or γίνομαι + εἰς meaning “to become,” ἐπιστρέφω meaning “to do something again,” or προστίθημι meaning “to continue to do something.” Needless to say, most of these features of the Septuagint text are to be explained against the Hebrew or Aramaic source text, whose elements are reproduced in Greek, although they are apparently missing or are at least infrequent in contemporary Koine writings. Of

“simplicitatemque scripturae sanctae, non ex maiestate sensuum, sed ex uerborum iudicant uilitate” (CCL 76: 409). 3 CCL 76: 450. 4 Thus Jerome in his Letter 53 addressed to Paulinus of Nola, §10 (CSEL 54: 463): nolo offendaris in scripturis sanctis simplicitate et quasi uilitate uerborum, quae uel uitio interpretum uel de industria sic prolatae sunt, ut rusticam contionem facilius instruerent et in una eadem que sententia aliter doctus, aliter audiret indoctus. 2

THE SEPTUAGINT AND GREEK STYLE 95

course, there are exceptions, even in the books translated rather literally, as with the use of participle constructions while the Hebrew text has finite verbs each time. Conjunctive participles are also frequent, usually in the nominative, especially in the Pentateuch (cf. Aejmelaeus 1993: 10; Walser 2001: 18–110).5 Less frequent, however, albeit attested around 200 times in the translated books, is the genitive absolute (Muraoka 2016: §31h), a characteristic feature of Greek syntax.6 Nevertheless, these and other phenomena do not significantly change the overall picture: the style of the translated books of the Septuagint is largely influenced by its Hebrew or Aramaic source texts, especially as far as syntax is concerned.

Traces of Stylistic and Rhetorical Skill? The “translational” character of the Greek of most of the books of the Septuagint is an undeniable fact. But this is not the whole story. In fact, a crucial question remains open: Did the various Greek-speaking translators of Jewish origin slavishly imitate the alleged simplicitas of the style of their source texts, especially on the level of word order, syntax, lexical choice, and rhetorical devices? Did they never take the liberty of diverging from their Hebrew or Aramaic Vorlage, giving us a glimpse at their stylistic and rhetorical skill, as suggested by some recent authors (e.g., Lee 2018: 82–3)? Admittedly, many features that are common in contemporary non-translation Greek are not easily detectable in the Septuagint (Walser 2016: 225). But are they really absent from the translated texts? A glance at an example will help form an answer. In Gen. 1:2 the earth is said to be in the state of ‫( תהו ובהו‬thw wbhw), enigmatic words over which much ink has been spilt (see, e.g., Dines 1995: 441–2; Bührer 2014: 102–4). Nowadays English translations typically use “a formless void” (NRSV), “without form and void” (NKJB), or the like. But the Septuagint opts for a different approach, selecting two verbal adjectives, one of which is a hapax legomenon in the Greek Bible: ἀόρατος καὶ ἀκατασκεύαστος (“invisible and unformed”) (NETS). To be sure, many essential questions about this approach remain open and cannot be addressed fully here (for a more detailed study of the problem, see, e.g., Schmitt 1974: 150–1; Rösel 1994: 31–3). But the most important questions are: Did the Alexandrian translator really understand the Hebrew phrase? To what degree does the Greek rendering match the meaning of the latter, or does it diverge completely from it? What exactly did the translator mean by ἀόρατος καὶ ἀκατασκεύαστος? Perhaps that the earth was still invisible because of being covered by water (Gen. 1:9) and that it was “not yet properly organized” (Dines 1995: 144)? Finally, is the Greek wording influenced by Platonic or more recent Greek philosophy? Whatever the answers to these questions, one conclusion is inescapable: The Greek phrase reproduces the sound effect of the Hebrew ‫( תהו ובהו‬MT tōhû wābōhû), albeit in a different way, by choosing two verbal adjectives characterized by alliteration (the alpha privative) and assonance (the ending in -ος; Dines 1995: 444). Is the case of the translation of ‫ תהו ובהו‬an isolated phenomenon in the Septuagint? Do other texts of the Septuagint provide further examples of linguistic choices demonstrating sensitivity and ingenuity on the part of the translators? Or is the simplicitas an overall feature of the Septuagint version of the Bible, the translators only seldom displaying their E.g., Gen. 15:15 (ταφείς); Num. 13:26 (πορευθέντες); Deut. 1:41 (ἀναβάντες); but attested elsewhere as well, e.g., Ruth 1:18 (ἰδοῦσα) and 4:18 (ἀγαπήσασα); Amos 5:8 (ἐκχέων) and 5:12 (ἐκκλίνοντες); Sir. 51:13 (ὤν). 6 E.g., Gen. 23:10 (ἀκουόντων); Jer. 35:9 (ἐλθόντος); Ps. 103[104]:28-29 (δόντος … ἀποστρέψαντος); Isa 58:9 (λαλοῦντος). 5

96

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

translational skill (Walser 2008: 456)? Admittedly, it is difficult to give a clear-cut answer since in-depth studies of the vocabulary and style of the Septuagint as a whole, or at least of its more important parts, are still lacking.

Is the Septuagint a Mere Calque of Its Source Texts? Moreover, we should bear in mind that recent investigation of the Septuagint is heavily influenced by two opposing currents: the interlinear paradigm theory and the study of translation technique. In recent decades, some scholars have argued that the Septuagint, at least its more literally translated texts, represents a sort of interlinear translation of the underlying Hebrew or Aramaic Vorlage, each element of the Greek translation rendering an analogous element of its source text as exactly as possible. On this view, the Greek text would be quite devoid of elements of interpretation. In other words, the translators normally did not leave their own marks on the translation.7 If this theory were correct, the Septuagint would be no more than a word-for-word translation of its Hebrew or Aramaic parent text, and there would be no room for any innovation on the part of the translator including the level of style. On the other hand, since the 1980s much research has been done in order to reconstruct the specific translation technique underlying each book or larger units of the Septuagint (e.g., the Twelve Prophets). The study of translation technique is not limited to more or less “technical” issues, like the rendering of the Hebrew imperfect or perfect or the infinitive absolute. A more intricate problem is the explanation of lexical choice, analysis of syntax, and identification of rhetorical devices. These studies have yielded a wide range of results, varying from book to book, concerning, for example, the choice of Greek equivalents for rare or enigmatic Hebrew words, the preference of translators for particular Greek terminology, their use of standard or exceptional equivalents, and their tendency to harmonize the vocabulary, especially with the result of creating a greater stylistic homogeneity (see, e.g., Dhont 2018 for the book of Job; Bons and Brucker 2019: 307–10 for the Psalter). In sum, these studies make a significant contribution to a better understanding of the linguistic profile of each book of the Septuagint on the one hand. On the other, it becomes possible to assess the manner in which the translator, rather than translating his Vorlage “mechanically” and word-by-word, gives it his own personal signature. Admittedly, the interlinear paradigm takes into account the fact that the Septuagint— at least the majority of its books—is a very word-for-word translation of its Hebrew or Aramaic source texts. That is why the translation itself is characterized to a great extent by a simplicitas similar to its Vorlage. Nonetheless, this approach cannot explain a wide range of phenomena of the Greek text of the Bible insofar as it excludes a priori any attempt on the part of the translators to provide more than a mere word-for-word translation. It was Jerome who already observed that the Septuagint included many additions and omissions (Letter 57).8 But that is not all. The detailed comparisons of the Septuagint and the Hebrew Bible carried out over the last decades have revealed a wealth E.g., Pietersma (2006a: 38), who states that “the primary reason for a word’s presence in such a translated text is to represent its Hebrew counterpart, rather than its appropriateness to the new context that is being created. The primary cognitive process is thus that Greek X is deemed a good match for Hebrew Y. In other words, prototypically, suitability in the Greek context is a secondary consideration, not a primary one.” 8 Letter 57,11 (CSEL 54: 522): Longum est nunc euoluere quanta Septuaginta de suo addiderint, quanta dimiserint. 7

THE SEPTUAGINT AND GREEK STYLE 97

of other results, namely, surprising divergences between the two versions of the Bible that require an explanation. In this context, it is necessary to widen the horizon by taking into consideration data of different origin, namely, texts of various genres of Greek literature, as well as the evidence of papyri and the inscriptions. To quote only one example: In Ps. 85[86]:5 God is called ἐπιεικής (“fair, gentle”), which is a hapax legomenon in the Greek Psalter, while the MT offers the adjective ‫“( סלח‬forgiving”). How can we explain such a rendering? In this case, a thorough investigation of the uses of ἐπιεικής in non-biblical literature can shed new light. In particular, the word “has the nuance of goodness and moderation, virtues that only a person in a superior condition can show by exercising his power. In fact, the author does not wish to be forgiven for something that he has done but is trying to win for himself the favour of God by praising his qualities” (Bellantuono 2019: 165).

The Necessity of Investigations of Style and Rhetoric of the Septuagint In recent decades, scholars have focused on discerning the effect that the numerous divergences between Hebrew and Greek texts have on the meaning of the target text as a whole. To what extent does the Greek translation have a profile of its own, despite being a translation? To what extent does the translator modify the meaning of his source text? To what extent do these divergences concern theological issues? Finally, do these phenomena reflect tendencies of Jewish theology of the Hellenistic era?9 Important though these questions are, they should not prevent us from tackling the problem of style: What about the stylistic and rhetorical skill of the translators? As already mentioned above, an in-depth study of this problem is still lacking. Therefore, the following considerations, which by no means claim to be exhaustive, aim to draw attention to various relevant phenomena giving us some insight into the workshop of the translators. For our purposes, it is sufficient to introduce two categories that prove useful for tracing the stylistic and rhetorical skills of translators: choice and deviation (see, e.g., Renkema 2004: 149). On the one hand, the translators have at their disposal a wealth of words, expressions, and syntactic structures in the target language so that they are capable of making choices in their work. On the other hand, they do not translate their source text “mechanically,” using nothing except standard equivalents as they slavishly follow the Hebrew or Aramaic word order. Rather, their individual style is determined by a certain freedom that consists in deviating from an expected wording or procedure. In other words, although the Hebrew or Aramaic source text is somewhat “recognizable” behind its Greek translation, the translators often go their own way, giving the text its own character in the target language. Bearing in mind that the categories used in the next two sections are not entirely exclusive of one another, the first will address phenomena like vocabulary, word order, and syntax. The second section deals with rhetorical features, some of which exceed the level of the clause, namely, alliteration and paronomasia. The examples chosen are taken from books or sections whose translators tend to stick to their Vorlage. Thus, the translation is not the result of a profound rearrangement of the words and phrases of the parent text but of a careful relecture, or rephrasing, which only cautiously intervenes in the syntactical structure of the text. Nonetheless, some slight differences are obvious.

On these issues, see Glenny and Dhont in the present volume.

9

98

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

WORD ORDER, VOCABULARY, AND SYNTAX The fundamental differences between an Indo-European language like Greek and a Semitic language like Hebrew or Aramaic are notorious, namely, on the levels of word formation, word order, and syntax. Yet the translators take advantage of some characteristics of the Greek language to make the translation sound more “natural,” in spite of rendering quite literally their respective Vorlagen. This might seem contradictory. Nonetheless, a thorough comparison of the versions of the biblical text shows various procedures: The translators avoid standard equivalents, they do not follow the Hebrew or Aramaic word order, and at times they create syntactic structures that cannot be explained by interference from the respective source text. The following three examples illustrate that the translators had a wide range of possibilities when it came to rendering a source text, even if it was not extremely puzzling or enigmatic.

Proverbs 31:16 The first example is taken from the so-called Praise of the good wife in Proverbs 31:10-31. On close inspection, the Septuagint version of this passage is not at all a word-for-word translation of the Hebrew source text, which was probably similar, if not identical, with the consonants of the later Masoretic Text. This applies also for verse 16, whose specific wording deserves attention: Prov. 31:16 ‫ זממה שׂדה ותקחהו‬She considers a field and buys it, ‫ מפרי כפיה נטעה כרם‬with the fruit of her hands she plants a vineyard (NRSV)

10

θεωρήσασα γεώργιον ἐπρίατο

After considering a field, she bought it,

ἀπὸ δὲ καρπῶν χειρῶν αὐτῆς κατεφύτευσεν κτῆμα

and with the fruits of her hands she planted a possession (NETS modified)

To begin with, on the level of syntax, the Hebrew text has two coordinated verbs in the first line. The Septuagint, however, employs a conjunctive participle in the aorist that expresses the idea of a relative past time with regard to the finite verb (BDF §339): After considering the field, the woman decides to acquire it. The object of both actions— considering and buying—being identical (i.e., the field), the Septuagint skips the personal pronoun referring to the field that is present in the Hebrew verb ‫“( ותקחהו‬and she bought it”). In general, this phenomenon is characteristic of Greek style, especially when the pronoun would refer to a noun that has just been named (Menge, Thierfelder, and Wiesner 1999: §95a).11 As for the vocabulary in this passage, four words deserve attention. First, the verb θεωρέω is one of the rather uncommon verbs in the translated books of the Septuagint, where it is sometimes used to render verbs like ‫“( ראה‬to see”; e.g., Ps. 21[22]:7) or ‫“( חזה‬to look,” “to see in a vision”; e.g., Ps. 26[27]:4). Only in Prov. 31:16 does θεωρέω render the Hebrew verb ‫“( זמם‬to plan,” “to consider”). It is difficult to decide whether the translator was familiar with this verb or not, since in the only other occurrence of the verb

Qere; see Fox (2015: 394). The phenomenon is attested elsewhere in Hebrew also; see Muraoka (2016: §74a).

10 11

THE SEPTUAGINT AND GREEK STYLE 99

in the book (Prov. 30:32), the Septuagint diverges significantly from the Hebrew text, rendering ‫ זמם‬with ἐκτείνω (“to stretch out”; see also Fox 2015: 388). Perhaps the choice of θεωρέω in Prov. 31:16 is at best explained as a contextual guess insofar as the range of meaning of the verb covers both the ideas of “seeing” and of “considering.” Only if, after inspection, the field appears to be appropriate for the woman’s purpose does she decide to acquire it. That leads us directly to the next point. Concerning the verb πρίαμαι (“to purchase”), it is first of all remarkable that the Hebrew verb ‫“( לקח‬to take,” “to acquire”) is not rendered with its standard equivalent λαμβάνω (“to take;” e.g., Prov. 7:20; 8:10) but with a rare word in the Septuagint. In fact, πρίαμαι occurs only in the Joseph story of the book of Genesis when it comes to sending the brothers of Joseph in Egypt in order to buy corn. The Hebrew equivalent is each time ‫“( ׁשבר‬to buy grain”; Gen. 42:2-3, 10; 43:2, 20). In Prov. 31:16, the translator does not opt for a more common verb, namely, κτάομαι (“to get,” “to acquire”; e.g., a property in Gen. 25:10; Ruth 4:10; Jer. 39[32]:7-9). But how do we explain the use of πρίαμαι in Prov. 31:16? To be sure, the verb is attested from Homer, throughout classical and until Hellenistic and later Greek literature, as the overviews in the lexicons show. However, the verb becomes a technical term in Hellenistic papyri, especially in Egyptian purchase and sale agreements where the transfer of ownership is expressed with parallel formulas stating the seller and buyer, respectively, as “x ἀπέδοτο” and “y ἐπρίατο” (e.g., BGU 3.996, lines 1–5; Pathyris, c. 112 B.C.E; P.Amh. 2.51, lines 22–7; Pathyris, 88 BCE; see Rupprecht 1994: 115). Hence, if the Septuagint of the book of Proverbs had an Egyptian Hellenistic setting (see Aitken and Cuppi 2015: 344), the use of the exact verbal form ἐπρίατο in Prov. 31:16 would have specific overtones: The translator uses a technical term borrowed from contemporary juridical documents. In so doing, he emphasizes that the woman does not only “buy” the field, with the details of the transfer of ownership remaining open. Quite the opposite is the case: In the light of the papyri, one could infer that the woman is involved in an official purchase contract that is made in the presence of the competent civil authorities and, by way of consequence, has legal effect. Finally, rather than choosing ἀγρός, (“field”), the standard equivalent for ‫“( שׂדה‬field”; e.g., Gen. 2:5; Exod. 8:9; Deut. 5:29) and a word familiar to the translator of the book of Proverbs (see Prov. 24:27), he opts for γεώργιον, “an area of land used for cultivation” (BDAG, ad loc.), just like in Prov. 24:30. In this latter verse, which diverges much from the Hebrew text, the Septuagint makes a comparison between a foolish man and his fields and vineyards insofar as their poor condition indicates the abilities of the person who cultivates them, as the following verse in the Septuagint version explains: If one leaves the vineyard to itself, it becomes barren and overgrown with weeds. In the light of this short passage, the choice of γεώργιον in Prov. 31:16 becomes more understandable. The point here is that the woman acquires the field with the very specific purpose of cultivating it. As such she is the stark opposite of the incompetent farmer or winegrower who neglects his γεώργιον. Returning to this specific word, it is extremely rare in the Septuagint but attested here and there in inscriptions (IStratonikeia 501, line 9; Lagina, Caria [Asia Minor], 323 BCE), papyri (P.Tebt. 1.72, line 370, Kerkeosiris, Arsinoites, 114/113 BCE; see also Moulton and Milligan 1930: 125) and Greek literature, but not before Strabo (Geogr. 14.5.6). Nevertheless, the choice of this noun shows again that the translator does not follow a trodden path choosing a standard equivalent of ‫שׂדה‬. In the Hebrew text, the woman is said to plant a vineyard, while the Septuagint reads κτῆμα (“possession,” “property”). Of course, the translator is familiar with the Hebrew noun ‫כרם‬, rendered with ἀμπελών (“vineyard”) in Prov. 24:30. The noun κτῆμα has

100

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

surely the connotation of a territory, as emerges from Prov. 23:10: μὴ μεταθῇς ὅρια αἰώνια, εἰς δὲ κτῆμα ὀρφανῶν μὴ εἰσέλθῃς (“do not relocate ancient borders, nor enter the possession of orphans”; NETS). Moreover, the idea that the woman cultivates the κτῆμα is confirmed by the verb καταφυτεύω (“to plant”). Consequently, the second line of the verse continues the topic of agriculture, including vine-growing. Once again, the Egyptian papyri provide precious material that could shed some light on the verse. In fact, a κτῆμα seems to be a sort of farm that produces, among other agricultural goods, wine (see, e.g., P.Rev., col. 37, lines 16–18, Arsinoites [?], 259–258 BCE; P.Col. 4.19276, lines 8–12, Philadelphia, 247 BCE; see also Preisigke 1915: 113). As a result, the noun κτῆμα appears to have a more general meaning (see also Jüngling, von Lips, and Scoralick 2011: 2000) than ἀμπελών, which could denote a single vineyard. Thus, according to the Septuagint text of Prov. 31:16, the woman is presented, not as a kind of amateur winemaker, but as a person whose enterprise and creativity enable her to run a farm. As an interim conclusion we can state that, in Prov. 31:16, the translator introduces very specific vocabulary that reflects, at least partially, his Hellenistic Egyptian background. It is noteworthy that three very common Hebrew words—‫“( שׂדה‬field”), ‫“( לקח‬to take”), and ‫“( כרם‬vineyard”)—are not rendered with their Greek standard equivalents, but with terms that are either rare (γεώργιον) or can be situated in a typically Egyptian milieu. Furthermore, the syntax of the verbs sounds more “natural” insofar as the Septuagint introduces a participle construction and skips the pronoun. Last but not least, the second line of the verse shows a rhetorical feature that will be dealt with below, the alliteration: καρπῶν χειρῶν αὐτῆς κατεφύτευσεν κτῆμα.

DanielOG 3:9-14 The following example is taken from the Old Greek version of Daniel 3 according to Rahlfs and Hanhart (2006), a translation that is known for its tendency for so-called dynamic translation. More specifically, that means that the Old Greek translation exhibits elements of formal equivalence, mostly following its parent text (see McLay 1994: 193 for Dan.OG 1:1-10). However, some differences are obvious, especially on the level of word order. Furthermore, additions and omissions as well as uncommon Septuagint equivalents should be noted. In Dan.OG 3:9 the three young Hebrews exiled in Babylonia—Shadrach, Meshach, and Abednego—are denounced to King Nebuchadnezzar by certain Chaldeans whose discourse is introduced as follows: Dan.OG 3:9 ‫ ענו ואמרין לנבוכדנצר‬They responded and said (NASB)

καὶ ὑπολαβόντες εἶπον

And they answered and said (NETS)

The typically Hebrew formula ‫ ויאמר‬x ‫“( ויען‬and x answered and said”) is usually reproduced in Greek by ἀπεκρίθη x καὶ εἶπεν (e.g., Exod. 4:1; Josh. 7:20; Joel 2:19; see also BDF §420). The same holds true in the Aramaic texts of the book of Daniel, where the corresponding Aramaic formula consisting of finite forms of ‫ ענה‬followed by ‫ אמר‬is rendered in the same way in Greek (e.g., Dan.OG 2:5, 7-8, 10; 3:14). However, unlike these passages, Dan.OG 3:9 introduces the discourse not with ἀποκρίνομαι (“to answer”),

THE SEPTUAGINT AND GREEK STYLE 101

but with ὑπολαμβάνω (“to answer”). Except for around twenty-five occurrences in the book of Job (2:4; 4:1; 6:1; etc.) where the stereotyped formula ὑπολαβὼν δὲ x λέγει (“then x said in answering”) introduces a discourse by one of the speakers in the dialogues between Job and his friends, this phrase is attested only once more in the book of Daniel, in Dan.OG 3:95(28). In other words, the OG translator of the book of Daniel normally adopts the usual Septuagint translation of the Hebrew or Aramaic formula quoted, on the one hand. On the other, at least in two cases, he deviates from the standard translation but follows Greek usage. In fact, the formula ὑπολαβὼν εἶπεν is attested quite frequently in nonbiblical Greek (e.g., Thucydides, Hist. 2.72), and it is also employed in cases where the discourse introduced by ὑπολαβὼν εἶπεν or the like is not at all an explicit answer to a previous statement or a question, as in Dan.OG 3:9 (see, e.g., Xenophon, Anab. 2.1.15). As for biblical writings, the formula occurs later in the New Testament, in Lk. 10:30. Nonetheless, as the example of Dan.OG 3:9 illustrates, the translator of the book of Daniel did not always feel bound by a set of standard equivalents borrowed from previous translations. In fact, he appears to be familiar with Greek style, on the one hand, but, on the other, this does not prevent him from employing the standard Septuagint equivalents. Interestingly, in the so-called Theodotion translation of the passage, a formula corresponding to καὶ ὑπολαβόντες εἶπον is lacking (see also the synopsis by Koch and Rösel 2000: 64–5). The central part of the denunciation is introduced as follows: Dan.OG 3:12 ‫ איתי גברין יהודאין‬There are certain Jews (NRSV)

εἰσὶ δέ τινες ἄνδρες Ιουδαῖοι

But there are certain Judean men (NETS)

Where does the word “certain” in the English translations come from? No doubt it derives from the OG Septuagint text, rather than from the Aramaic. In overall terms, indefinite pronouns are quite rare in the translated books of the Septuagint (Muraoka 2016: §10), the plural form τινες corresponding here and there to ‫( אנׁשים‬literally “men”) but used with the meaning “some” (e.g., Exod. 16:20; Jer. 44[37]:10; see Thackeray 1909: 45). Returning to Dan.OG 3:12, the use of the indefinite pronoun does not at all mean that the speaker is not aware of the identity of the “certain Judean men” (see Muraoka 2016: §10b). On the contrary, their identity will be revealed in precisely the same verse. So what is the exact function of the phrase εἰσὶ δέ τινες ἄνδρες Ιουδαῖοι? To my mind, by inserting δέ τινες in its specific context the OG translator increases the tension caused by the reminder of the royal decree quoted in the preceding verses: There are certain persons who are not willing to obey the king’s order to worship the golden statue (for a similar use of the indefinite pronoun, see Demosthenes, Aristocr. 142; see also Kühner and Gerth 1904: I, §470.3). In so doing, the translator draws attention to the three young men who are not yet identified but who will soon emerge from anonymity. In sum, the translator makes use of a typically Greek formula in order to nuance an idea already present in the Aramaic text. This aspect becomes all the more apparent when one compares the Old Greek translation with that attributed to Theodotion that follows closely the Aramaic text: εἰσὶν ἄνδρες Ιουδαῖοι (“there are Judean men”).

102

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

The denunciation continues as follows: Dan.OG 3:12 ‫ גבריא אלך לא־ׂשמו‬These pay no heed ‫ עלך מלכא טעם‬to you, O king ‫ לאלהיך לא פלחין‬They do not serve your gods ‫ הקימת לא סגדין‬and they do not ‫ ולצלם דהבא די‬worship the golden statue that you have set up (NRSV)

οἱ ἄνθρωποι ἐκεῖνοι οὐκ ἐφοβήθησάν σου τὴν ἐντολὴν

These people do not fear your command

καὶ τῷ εἰδώλῳ σου οὐκ ἐλάτρευσαν

and they do not serve your idol

καὶ τῇ εἰκόνι σου τῇ χρυσῇ ᾗ ἔστησας οὐ προσεκύνησαν

and they do not do obeisance to your gold image, which you have set up (NETS)

The extant Aramaic text of the first line is difficult to understand, especially the meaning of the verb ‫ ׂשים‬and the noun ‫( טעם‬see, e.g., Koch 2005: 250). Be this as it may, the translator gives the Greek text a shape that is different from the Aramaic text in two respects: First, he creates a chiastic text (for this rhetorical device see Kühner and Gerth 1904: II, §607), which no longer mentions the king and his gods (see also Meadowcroft 1995: 152; McLay 1994: 245). The chiastic character of the first two lines is enhanced by the pre-position of the enclitic pronoun σου: σου τὴν ἐντολήν instead of τὴν ἐντολήν σου. An analogous word order is impossible in Aramaic. However, given the fact that the Greek wording does not correspond to the morpheme order of the Aramaic text, one could infer that the pre-position of σου is due to emphasis (see Kühner and Gerth 1904: II, §606; Muraoka 2016: §41ab). It is precisely the royal command that the young Israelites do not fear as can be illustrated by the following diagram: οὐκ ἐφοβήθησάν   σου   τὴν ἐντολὴν καὶ τῷ εἰδώλῳ   σου οὐκ ἐλάτρευσαν Furthermore, in rendering ‫“( לאלהיך‬your gods”) or perhaps a singular form of the same noun (see the Qere) with the dative of εἴδωλον (“idol”), rather than an equivalent for “god” (τῷ θεῷ) or “gods” (τοῖς θεοῖς), the translator adjusts the last lines of the verse to each other, creating a sort of crescendo: The “idol” and the “golden image” are one and the same, both terms being used as synonyms, as emerges from the Old Greek version of Dan. 3:18 as well. Thus, the last line of the verse develops what has been already said in the second last line: The image is golden and has been erected at the king’s will. Nebuchadnezzar is not willing to let the three recalcitrant young men get away, but summons them to justify themselves. Once more, the translator opts for a more specific word. Rather than “saying” (‫ )אמר‬Nebuchadnezzar “orders” that the three Judean men be brought (Dan.OG 3:13), employing προστάσσω, a favorite word of the OG translator he already used in Dan.OG 2:14 for ‫אמר‬:

THE SEPTUAGINT AND GREEK STYLE 103

Dan.OG 3:13-14 …‫ באדין נבוכדנצר‬Then ‫ אמר להיתיה‬Nebuchadnezzar … ‫ לׁשדרך מיׁשך ועבד נגו‬commanded that Shadrach, Meshach, and Abednego be brought in (NRSV)

τότε Ναβουχοδονοσορ … προσέταξεν ἀγαγεῖν τὸν Σεδραχ Μισαχ Αβδεναγω

Then Nabouchodonosor … ordered that Sedrach, Misach, Abdenago be brought in (NETS)

That the translator allows himself to slightly nuance the text emerges also from the introduction of Nebuchadnezzar’s encounter with the three young men (Dan.OG 3:14). In this case, the translator makes use of a specific feature of Greek syntax that has no equivalent in Aramaic: Dan.OG 3:14 ‫ ענה נבכדנצר ואמר להון‬Nebuchadnezzar said to them (NRSV)

οὓς καὶ συνιδὼν Ναβουχοδονοσορ ὁ βασιλεὺς εἶπεν αὐτοῖς

So when King Nabouchodonosor saw them, he said to them (NETS)

From the point of view of the narrative, the first words of the Greek translation serve as a kind of transition. What will happen after the three young men have been presented to king Nabouchodonosor? So, as if the translator would not let the king speak immediately, he inserts a phrase that cannot be explained on the basis of the Aramaic text: Before taking them to task, the king “perceives” the young men accused of transgressing his decree. Once more, it should be emphasized that the verb συνοράω (“to see,” “to comprehend”) does not at all belong to the standard vocabulary of the translated texts of the Septuagint (see also McLay 1994: 248). But this is not the only argument in favor of the translator’s signature. There is also a syntactical phenomenon that is not due to an Aramaic Vorlage, even if it would have been misunderstood. In fact, from a syntactical point of view, the phrase οὓς καὶ συνιδὼν Ναβουχοδονοσορ seems to introduce a relative clause (McLay 1994: 239). However, it is equivalent to a coordinate clause, the relative pronoun οὕς having the function of a demonstrative (Smyth 1920: §2490). In nonbiblical Greek literature, as well as in the New Testament, this syntactical phenomenon, called “relative connection” (van Emde Boas 2019: §50.16), is attested (e.g., 2 Tim. 4:15; see BDF §293.3c), while in the Septuagint it seems to be quite rare. Be this as it may, this is another example of the translator’s openness to introducing elements of Greek vocabulary and syntax that do not belong to the standard features of the language of the Septuagint. As a preliminary conclusion, it can be said that the OG translator of Daniel has a certain range of syntactic and lexical possibilities at his disposal to reproduce his source in a more idiomatic Greek. This reality becomes even clearer when comparing the OG with the translation attributed to Theodotion, where, for example, ‫ענה נבכדנצר ואמר להון‬ (literally, “and Nebuchadnezzar answered and said to them”) in Dan.OG 3:14 is rendered quite literally by καὶ ἀπεκρίθη Ναβουχοδονοσορ καὶ εἶπεν αὐτοῖς (“and Nabouchodonosor answered and said to them”; for further details, see McLay 1994: 233–59).

104

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Amos 6:4 Concerning word order, because the Greek language is inflected, it can separate words that, from the perspective of syntax, should be connected. A similar word order is not possible in Hebrew and Aramaic. This leads here and there to the phenomenon of hyperbaton (see Andersen 2000: 121–2). It is attested in various biblical books, such as in Psalms:12 τίς ἐστιν ἄνθρωπος ὁ θέλων ζωὴν ἀγαπῶν ἡμέρας ἰδεῖν ἀγαθάς (Ps 33[34]:13) “What person is he who wants life, coveting to see good days” (NETS) κρεῖσσον ὀλίγον τῷ δικαίῳ ὑπὲρ πλοῦτον ἁμαρτωλῶν πολύν (Ps 36[37]:16) “Better is a little that the righteous has than the great wealth of sinners” (NETS) πάντα πρὸς σὲ προσδοκῶσιν δοῦναι τὴν τροφὴν αὐτοῖς εὔκαιρον (Ps 103[104]:27) “All look to you to give them food in due season” (NETS)

This specific feature of Greek language can be found all over the books of the Septuagint, including texts for which it is not certain whether or not there was a Semitic original (see, e.g., Pss. Sol. 13:3; 17:19, 43; see Joosten 2015: 40).13 An interesting example is Amos 6:4, where the Septuagint has a plus vis-à-vis the Hebrew text: Amos 6:4 ‫ ואכלים כרים מצאן‬and [they] eat lambs ‫ ועגלים מתוך מרבק‬from the flock, and calves from the stall (NRSV)

καὶ ἔσθοντες ἐρίφους ἐκ ποιμνίων καὶ μοσχάρια ἐκ μέσου βουκολίων γαλαθηνά

And [they] eat kids from the flocks and suckling calves from the midst of the herds (NETS)

Whatever the translator may have understood the last word of the Hebrew text, ‫מרבק‬ (“fattening”), there is no doubt that the adjective γαλαθηνά (“suckling”) refers to μοσχάρια (“calves”) as its governing noun, but is separated from it by ἐκ μέσου βουκολίων (“from the midst of the herds”).

ALLITERATION AND PARONOMASIA The Greek language contains some phenomena that are largely foreign to Semitic languages, in particular a more extensive use of prefixes and the possibility of forming different types of words—nouns, adjectives, adverbs, verbs—from a single lexical stem. These features enable translators to create various rhetorical features. A striking example is Job 2:3, where the protagonist of the book is described as follows:14 ἄνθρωπος ἄκακος, ἀληθινός, ἄμεμπτος, θεοσεβής, ἀπεχόμενος ἀπὸ παντὸς κακοῦ ἔτι δὲ ἔχεται ἀκακίας “An innocent, genuine, blameless, religious man, staying away from all wrong, and he still maintains his innocence” (NETS, slightly modified)

For more details, see Bons (2011: 70–2). For more examples, see Muraoka (2016: 484–5); Dhont (2018: 156–60). 14 See also Dhont (2018: 162). 12 13

THE SEPTUAGINT AND GREEK STYLE 105

Not only is the alliteration with the letter alpha a salient feature of the verse, but so is the wordplay that links ἄκακος, ἀπὸ παντὸς κακοῦ, and ἔχεται ἀκακίας, that is, words denoting integrity and its opposite. Examples where several words form an alliteration are attested elsewhere as well, as in Sir. 51:13:15 πρὶν ἢ πλανηθῆναί με ἐζήτησα σοφίαν προφανῶς ἐν προσευχῇ μου “When I was still young, before I wandered, I sought wisdom plainly in my prayer” (NETS)

However, examples where only two words are concerned are more frequent (see also Dines 2016: 375–7; Backfish 2019: passim), as in the phrase ἀδίκων καὶ ἀνόμων καταγελάσῃ (“you will laugh at the unrighteous and lawless”) in Job 5:22. A somewhat similar phenomenon is the so-called homoioteleuton, where the respective words have identical endings, as in Amos 6:11 καὶ πατάξει τὸν οἶκον τὸν μέγαν θλάσμασιν καὶ τὸν οἶκον τὸν μικρὸν ῥάγμασιν (“he shall strike the great house with bruises and the little house with lacerations”; NETS) and Hab. 2:16 διασαλεύθητι καὶ σείσθητι (“shake, and quake”; NETS; see also Mulroney 2016: 93–4). Finally, both phenomena occur side by side, as in the adjectives in Gen. 1:2 quoted above, ἀόρατος καὶ ἀκατασκεύαστος. as well as in Ps. 49[50]:13 κρέα ταύρων ἢ αἷμα τράγων (“flesh of bulls or … blood of he-goats”; NETS). These and other examples show that the translators understood the art of literary embellishment, although this art is not equally developed in all the books of the Septuagint. In some cases, phenomena like alliteration do not only affect the translation of two or more closely connected words. There is some evidence for saying that the respective words have not been selected at random, but with the aim of giving a specific character to a smaller literary unit. In what follows, I will take two examples from different literary genres.

Psalm 91[92]:7 ‫ איׁש־בער לא ידע‬The dullard cannot know ‫ וכסיל לא־יבין את־זאת‬the stupid cannot understand this (NRSV)

ἀνὴρ ἄφρων οὐ γνώσεται

A foolish man cannot know

καὶ ἀσύνετος οὐ συνήσει ταῦτα

and a stupid one will not understand these things (NETS)

As in the Greek translation of Gen. 1:2, the translator of the Psalter employs two adjectives formed with an alpha privative, ἄφρων and ἀσύνετος. Two comments need to be made. First, in the Greek language there is at least one positive adjective that would have been fitting for rendering one of the Hebrew adjectives: μωρóς (“dull,” “stupid”). This word is not at all unknown to the translators (e.g., Deut. 32:6; Sir. 18:18; Isa. 19:11; 32:6). The Septuagint Psalter, however, employs it only once, in Ps. 93[94]:8, where the Masoretic Text reads ‫“( כסיל‬stupid”). Therefore, the choice of ἄφρων and ἀσύνετος might be best explained by the interest of creating an alliteration not only between the two adjectives but also with ἀνήρ.

For the specific vocabulary, see Scialabba (2019: 380–1).

15

106

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Secondly, the two adjectives ‫ כסיל‬and ‫ בער‬are translated in Ps. 48[49]:11a with ἄφρων (“fool”) and ἄνους (“without understanding”), respectively (for details see Bons 2011: 72–3). In Ps. 91[92]:7, only one of them is retained. The second, however, is replaced by the quite rare Septuagint word ἀσύνετος (“stupid”), which is not at all a standard equivalent for ‫ כסיל‬or ‫בער‬. Perhaps this choice suggested itself because the Greek verb usually employed to translate the Hebrew verb ‫“( בין‬to understand”) is συνίημι, also in the Psalms (Ps. 5:2; 18[19]:3, etc.), and it occurs in the same verse. Thus, by opting for ἀσύνετος, the translator not only creates another rhetorical device—namely, paranomasia, or the “recurrence of the same word or word stem in close proximity” (BDF §488)—but also strengthens the idea of the impossibility of understanding. In other words, the ἀσύνετος is unable to understand what is said in verse 8: The apparent success of sinners and unlawful people does not last long.

Jeremiah 2:6 In Jer. 2:6, God accuses the Israelites of no longer remembering the God who once led them through the desert after their Exodus from Egypt. Similar to other biblical texts where the desert is described with adjectives formed with alpha privative (e.g., ἐν γῇ ἐρήμῳ καὶ ἀβάτῳ καὶ ἀνύδρῳ in Ps 62[63]:2), the desert is characterized as follows: ‫ במדבר‬In the wilderness ‫ בארץ ערבה וׁשוחה‬In a land of deserts and pits ‫ בארץ ציה וצלמות‬In a land of drought and deep darkness ‫ בארץ לא־עבר בה איׁש‬In a land that no one passes through ‫ ולא־יׁשב אדם ׁשם‬Where no one lives (NRSV)

ἐν τῇ ἐρήμῳ

In the wilderness

ἐν γῇ ἀπείρῳ καὶ ἀβάτῳ

In a land boundless and trackless

ἐν γῇ ἀνύδρῳ καὶ ἀκάρπῳ

In a land waterless and fruitless

ἐν γῇ ἐν ᾗ οὐ διώδευσεν ἐν αὐτῇ οὐθὲν

In a land that nothing passed in it

καὶ οὐ κατῴκησεν ἐκεῖ υἱὸς ἀνθρώπου

And no person settled there? (NETS)

The divergences between the Hebrew text and the Septuagint are conspicuous, namely, the rendering of the negated noun ‫“( איׁש‬no one”) by οὐθέν (“nothing”) in the fourth line (for text-critical issues, see also Walser 2012: 205–6). In so doing, the Septuagint obviously underlines the inhospitality and harshness of the desert (Finsterbusch and Jacoby 2016: 46 n. 67). Furthermore, as for the explanation of the word “wilderness” (‫)מדבר‬, the translator creates two-word pairs, each of which consists of two adjectives formed with the alpha privative. The first word-pair focuses on the complete absence of any reference point in the desert insofar as it seems “infinite” and “untrodden” (ἀπείρῳ καὶ ἀβάτῳ). The second word pair, however, emphasizes the idea of infertility insofar as the lack of water makes any growth of vegetation impossible (ἀνύδρῳ καὶ ἀκάρπῳ). Of course, the question arises whether the translator was familiar with the Hebrew words (provided that his Vorlage was identical with the later Masoretic Text). Without entering into too many details, the following observations can be made: First, the meaning “boundless” or “infinite” of the adjective ἄπειρος is not unknown to nonbiblical Greek (e.g., Herodotus, Hist. 5.9; Aristotle, Mot. an. 699b17), but it is not attested elsewhere in the Septuagint. In Jer. 2:6, the Hebrew equivalent is ‫“( ערבה‬desert”), while in Jer. 17:6 the same Hebrew noun is translated with ἔρημος (“desert”).

THE SEPTUAGINT AND GREEK STYLE 107

Second, the word pair γῆ ἄνυδρος καὶ ἄβατος appears as equivalent of ‫“( ארץ ציה וערבה‬a land of drought and a desert”; NRSV) in Jer. 28:43 (MT Jer. 51:43) as well. Thus, at least the translation of ‫ ציה‬is concordant, with the translator opting for ἄνυδρος in both occurrences. Third, as for ἄβατος, the Hebrew equivalent in Jer. 2:6 is ‫“( ׁשוחה‬pit”), a noun that occurs once more in Jer. 18:20 but is translated differently in the Septuagint. It cannot be ruled out that the translator of the book of Jeremiah, who is likely to be ignorant of the meaning of the word ‫ׁשוחה‬, borrowed the Greek equivalent from Lev. 16:22, where the Hebrew text reads the rare word ‫“( גזרה‬infertile land” (see Pietersma 2006b: 412). Perhaps the translator of the book of Jeremiah, who no doubt had a preference for ἄβατος (sixteen further occurrences of the word!), deemed it a fitting translation in Jer. 2:6. Fourth, the most stunning divergence is between ‫“( צלמות‬darkness”) and ἄκαρπος (“fruitless”), an adjective used nowhere else in the translated books of the Septuagint. The standard Greek equivalent of ‫ צלמות‬is σκιὰ θανάτου (“shadow of death”),16 a translation known also to the translator of the book of Jeremiah (see Jer. 13:16) and familiar to the translators of other biblical books (e.g., Isa. 9:1; Amos 5:8; Ps. 22[23]:4; Job 3:5). Taken together, these observations lead to the conclusion that the translator deviated from the usual paths and wanted to give the translation a shape of its own. This hypothesis is supported by another line of evidence: On the one hand, the first pair of adjectives, ἄπειρος and ἄβατος, would anticipate the idea alluded to in line three, namely, that the desert is not suitable for crossing. On the other hand, the second pair of adjectives, ἄνυδρος and ἄκαρπος, would prepare the idea developed in line four, namely, that the desert is not suitable for living. Thus, this small textual unit is built according to the following pattern: a – b – a’ – b’. In sum, this example shows that in Jer. 2:6 the translator is far from translating his Hebrew Vorlage “mechanically” (provided that it was at least similar to the Masoretic consonantal text). Rather, in carefully selecting non-standard Greek equivalents consisting of adjectives formed with an alpha privative, the translator not only creates a fourfold alliteration, but also shapes this small unit differently so as to give it a greater homogeneity and a clearer structure.

ONGOING RESEARCH QUESTIONS It is a truism that language and style of the Septuagint books, even within the chapters of one and the same book, differ depending on the overall translation technique adopted by the respective translator (see Prestel 2016: 50–4). These differences emerge all the more sharply when one compares two translations of the same book, such as the Old Greek of the book of Daniel and the translation attributed to Theodotion. Consequently, scholars attempt to describe and classify the various degrees of literalism and the different nuances between a “free” and a “literal” translation. As for the latter, the tendency of translating word by word and using stereotyped equivalents—even at the risk of creating absurd translations like inquiring “after the peace of the war” (εἰς εἰρήνην τοῦ πολέμου) in 2 Kgdms 11:7)—is well known and has led to the numerous Hebraisms in Septuagint texts. For this reason, the language of the Septuagint has been considered “simple” ever since The translators understood the word as a kind of status constructus of ‫“( צל‬shadow”) and ‫“( מות‬death”). The Masoretes vocalized the word accordingly. However, it is by no means a compound word with the meaning mentioned, but a noun meaning “darkness” or the like; see the dictionaries, e.g. GKC18: 1120.

16

108

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

antiquity. However, even if a Septuagint text is considered rather literal—perceiving the  underlying Semitic Vorlage shining through the translation—here and there the translators show us a glimpse of their stylistic and rhetorical skill. Nevertheless, research on this issue has to go on. As has been highlighted throughout this chapter, research on stylistic and rhetorical features of the Septuagint is still in its infancy. This is certainly due to the fact that the Septuagint has often been studied from the point of view of the Hebrew text. In fact, especially in the field of textual criticism of the Hebrew Bible, the Septuagint has often been used as a sort of quarry that is only expected to provide textual variants when the Hebrew biblical text seemed to be difficult, enigmatic, or even erroneous. Nevertheless, a paradigm shift is necessary, namely, an investigation of the Septuagint texts in their own right. In other words, the question at stake is not whether a Hebrew word has an equivalent in the Septuagint and, if so, does it more-or-less match the meaning of the corresponding Hebrew word. Rather, the crucial question is how the Septuagint translators made use of Greek language. This opens up a new field for research that has so far been investigated only very incompletely.17 This concerns the three areas already mentioned— vocabulary, syntax, and rhetorical devices—although these three categories cannot always be sharply distinguished from one another, as can be illustrated by our analysis on Jer. 2:6 above. As for vocabulary, it is first of all important to explain why a translator could have employed relatively uncommon words and what function they are supposed to have in the context. As for syntax, there is no doubt that the Septuagint typically does not deviate from the Hebrew or Aramaic source text. Nevertheless, there are exceptions that need to be detected, which can be done with the aid of research tools that have been available for a long time (e.g., BDF) and more recent ones (e.g., Muraoka 2016). As for rhetorical devices, it would be worthwhile to examine texts that can be understood as discourse in the broadest sense: not only speeches but also prayers and prophetic texts.

RECOMMENDED RESOURCES 1. Dhont (2018). A careful analysis of the typical stylistic and rhetorical features of the book of Job, whose approach is important for the investigation of the language of the Septuagint in general. 2. Dines (2016). A short overview of different stylistic features of the Septuagint taken from various literary corpora, namely, the Pentateuch and the Twelve Prophets. 3. Lee (2018). A very detailed and groundbreaking investigation of the language of the Septuagint Pentateuch that explains its specific features against its linguistic background: the Hebrew Bible, on the one hand, and the language of the Koine on the other. 4. Léonas (2016). An important article dealing, on the one hand, with history of research and, on the other, the conceptual framework required in order to describe stylistic features of the Septuagint. 5. Muraoka (2016). An in-depth study of the syntax of the Septuagint in its broader sense: word order. 6. Walser (2001). A careful investigation of the syntax of the Septuagint and Jewish Hellenistic literature against the background of non-Jewish literature of the Hellenistic era. The author focuses on the use of participles, conjunctions, and particles.

For the book of Daniel, see the monographs by McLay (1994) and Meadowcroft (1995); for Job, see the recent monograph by Dhont (2018).

17

CHAPTER 7

The Septuagint and Biblical Intertextuality MYRTO THEOCHAROUS

INTRODUCTION When authors decide to produce a text they will inevitably employ words, meanings, ideas, images, metaphors, expressions, idioms, and other linguistic tools that they know from their own cultural matrix in order to present it to their current readers in a way that will presumably make their product understandable. Indeed, they have no other means of producing a text apart from drawing upon their own ways of expression. It is, therefore, unavoidable that their product will contain expressions that are also present in other texts available to them and presumably to their readers, whether those are available in written or oral form. These inevitable connections between texts are what make reading understandable, and the scholarly recognition of this state of affairs led to the coining of the term “intertextuality” by Julia Kristeva in an attempt to explain how texts work. Kristeva was originally interested in examining a philosophy of how meaning was generated in language. Moving beyond theories of fixed meaning, she argued, along the lines of Mikhail Bakhtin, that texts exist in dialogue with other texts, those of the writer, the addressee, and the contemporary or earlier cultural context, effectively transforming diachrony into synchrony (1980: 64–5). She writes, “Any text is constructed as a mosaic of quotations; any text is the absorption and transformation of another” (Kristeva 1980: 66).1

Intertextuality in Biblical Studies The term “intertextuality” was then borrowed from the field of literary criticism by biblical scholars with whom it had lost its philosophical function. Biblical scholars now tend to use the term when they locate and describe how a biblical author employs other texts (known as “intertexts”) in their composition. A specific biblical text may appear to be an echo, allusion, or quote of other texts in or outside scripture, and scholarly interest in these phenomena are now primarily hermeneutical in nature. Intertextual phenomena raise questions such as, “How does the new composition interpret the intertext it alludes to?” and “Does the intertext retain its understood original meaning, or is it transformed?”

For more details on how the term “intertextuality” arose in the field of literary criticism, see Aichele and Phillips (1995).

1

110

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

When intertextuality began to inform biblical scholarship, Old Testament scholars were at first interested in how relatively later OT texts used earlier ones within the canonical corpus. This exercise often helped scholars evaluate the chronological arrangement of biblical texts. But interest soon extended beyond inner-OT intertextuality to the use of other ancient Near Eastern texts by biblical ones as well (e.g., Alter 1981; Fishbane 1985; Alter 1992; Nogalski 1996; de Moor 1998; Schultz 1999; Polaski 2001; Strazicich 2007). As a result of this development in biblical scholarship, the term “intertextuality” is also loosely used as a synonym for “echo” or “allusion.” As Peter Miscall (1992: 44) describes, “intertextuality” is a covering term for all possible relations that can be established between texts. The relations can be based on anything from quotes and direct references to indirect allusions to common words and even letters to dependence on language itself. Scholars have not really been explicit about the criteria used to identify these textual connections. One of the best sets of criteria for working with biblical material has been proposed by the New Testament scholar Richard Hays (2005: 34–45), who was mainly interested in the use of the Old Testament in the Pauline corpus.2 These criteria are summarized as follows: (1) Availability: The source of the intertextual reading must be available to the writer who uses it. (2) Volume: This refers mainly to how “loud” the intertextual reading is, that is, the degree of verbatim repetition of words and syntactical patterns. Also, it refers to how familiar the precursor text is to the later author and how much rhetorical stress that author placed upon the phrase(s) in question. (3) Recurrence or Clustering: Also called “multiple attestation.” How often does the author elsewhere cite or allude to the same passage? (4) Thematic Coherence: How well does the alleged intertextual reading fit into the line of argument the author is developing? (5) Historical Plausibility: Could the author have intended such an intertextual reading and would it be meaningful to his readers? This guards against anachronistic readings. (6) History of Interpretation: This refers to the subsequent history of the reading of a text and whether subsequent readers have noted the same intertextual readings. Other readers in the tradition may help to alert us to intertextual readings we may have missed. (7) Satisfaction: Does the proposed intertextual reading illuminate the surrounding discourse and the document or corpus as a whole?3

Intertextuality and Translated Texts So far we have looked at the scholarly interest in tracing intertextuality in original compositions. For the most part, people do not usually expect translators to function as authors, since the main task of translators is thought to be a “faithful” reproduction

Interest in the use of the OT in the NT has become, and continues to be, very popular and has generated works such as Beale and Carson (2007). 3 The criteria are restated and elaborated from Hays (1989: 29–33). 2

THE SEPTUAGINT AND BIBLICAL INTERTEXTUALITY 111

of their source text into another language. Despite this presupposition, scholars have increasingly become aware that translators are not slaves to their source text, but unavoidably contribute—either intentionally or unintentionally—to the production of additional meaning in their translation that may not have been present in their source text. In some way or another, they transform the text. Van der Louw’s study on transformations in translation is very informative here. He defines transformations, or shifts, as “micro-level changes that occur in the transfer from one language to another” (2008a: 109). Of course, transformations in translation can take place at the macro level as well but what we observe most often, particularly in examining intertextuality in the Septuagint, is small shifts that nevertheless manage to create new meanings. He makes the following observations: [W]hy do translators apply transformations? The answer is obvious: because a literal translation does not work! Literal translation is always the easiest and fastest method. Even the so-called “free translator” proceeds literally most of the time, at least in prose. A transformation is used to solve the translational problem that arises from a literal rendering. This has an important methodological implication. Behind each transformation stands a literal rendering that has been rejected. Here, I would add that even though sometimes the literal rendering is intentionally rejected, at other times it is simply unknown to the translator. One would think that such transformations are true only of dynamic or paraphrastic translations, but scholars have noted that even the most word-for-word translations involve some degree of interpretation (e.g., Barr 1979). As with any reader of the Hebrew scriptures, their translators are also readers and interpreters of their text, and in the case of the Septuagint we have indeed the first witness to the reception or interpretive reading of the Hebrew scriptures. Given these authorial features of the translator, it is therefore legitimate for scholars to attempt a study of intertextuality in translated texts. As we shall see below, there are additional challenges in tracing intertextuality in translated texts that a scholar who studies original compositions does not have to face. The primary challenge arises from having to evaluate intertextual features produced by one of two literary agents—the author and the translator—and tracing how those features arose. While an author has no constraints in incorporating other texts into their composition, a translator is limited by the source text as far as the extent of interpretive freedom is concerned: the translator’s work must represent the source text if it is to be regarded as a translation. Yet it is often the case that the translator will subtly “manipulate” elements in the source text in order to introduce an intertextual reading, rather than simply add extra words or phrases.4

Intertextuality in the Septuagint One of the primary reasons that the study of intertextuality began to be applied to the LXX was interest in understanding the hermeneutics of the translators and gaining some insight into their theological perspectives. What did their reading of the Hebrew scriptures—revealed in the manner with which they translated, including their word

The latter is more commonly encountered in the Targums than in the Septuagint.

4

112

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

choices, deviations, and so on—say about the way they viewed their source text? (see, e.g., Koenig 1982; Dorival 1995). Joachim Schaper (2006) in particular was interested in seeing whether deviations or specific word choices in Greek Isaiah provided a window into the translator’s messianic expectations by observing how the translator connected messianic texts in the book.5 For other scholars, such as Thackeray, the study of intertextuality in the Septuagint (similar words and expressions) was a means of identifying whether one translator worked in more than one book of the canon (Thackeray 1903a, 1903b, 1903c).6 For other scholars who are more interested in the dating and chronological order of Septuagint books, the study of intertextuality helps evaluate which Septuagint translators were aware of other earlier Septuagint translations and thus made use of Greek expressions found in them.7

METHODOLOGICAL CONSIDERATIONS AND CRITERIA The Translator’s Vorlage and Target Text The first criterion that Jennifer Dines (2007: 5) offers for identifying intertextuality has to do with the Septuagint translator’s Vorlage: A resemblance between two texts must not be explicable as independent renderings of similar Vorlagen. In other words, before claiming to have discovered the use of intertextuality by the Septuagint translator, we must ensure that the translator is not simply following the Vorlage available to him but is instead, in one way or another, intruding. Therefore, the relationship between the translator and his Vorlage must be established. As mentioned previously, studying intertextuality in translation comes with greater challenges than those of studying original compositions, since we are dealing with two literary agents: the author and the translator. It is, therefore, a prerequisite for scholars to distinguish the two voices and make sure that their observations reflect the translator’s use of intertextuality and not that of the underlying author. Intertextuality in the underlying source text, and specifically his Vorlage, that is simply reproduced in the translation would not count as the translator’s personal intertextual contribution. In order for us to describe with accuracy the relationship between the author and the translator, we must attempt to get as close as possible to the voice of the translator as well as to the voice of the author. For this reason, thorough text-critical work on the source text as well as on the target text has to take place. Karen Jobes and Moisés Silva (2015: 328) alert us to this prerequisite for pondering any theological motivation (or ideological, if I may add) behind the translation. They urge us to ensure that both the original words that translated the Semitic parent text and any additions to or omissions from that original translation have been identified. This work allows the theological viewpoint (Tendenz) of the original translators to be distinguished from that of subsequent revisers and copyists. If the theological trait in question is indeed

Similarly Ngunga (2012). See also van der Kooij (2001: 229–47). 7 See, e.g., Ziegler (1934: 103–34); Cook (2001); Van der Vorm-Croughs (2001); Williams (2001); Seeligmann ([1948] 2004: 222–30); Cook (2006a); Dines (2007); Hacham (2007); van der Kooij (2008); Perkins (2009); Cook (2010a); Dogniez (2010). 5 6

THE SEPTUAGINT AND BIBLICAL INTERTEXTUALITY 113

part of the original translation, then one must try to discern if it was introduced by the Greek translator or if it was already present in the Hebrew Vorlage. Due to the complex history of the reception of the Septuagint, it is important that one examines available variants of the Greek text in order to decide which one is closest to what the translator may have written. Publications such as the Göttingen edition of the Septuagint are extremely useful for making these variants available in the critical apparatus. The same approach should be taken with the Hebrew text. One cannot assume that the Hebrew Vorlage used by the Septuagint translator is identical to the MT we have in modern printed editions such as the BHS, especially with respect to books with extensive variations such as Jeremiah or Ezekiel. It is important to examine available Masoretic variants, other ancient versions, and extant manuscripts from the Judean desert. Regarding the Hebrew Vorlage of the Septuagint as a whole, on the basis of frequent interchanges between ‫( ד‬dalet) and ‫( ר‬resh), as well as ‫( י‬yod) and ‫( ו‬wāw), reflected in the readings of the Greek translators, it is believed to have been written in the Aramaic square script, probably very similar to that of many of the Qumran scrolls (Tov 1993: 117). Moreover, there is not much evidence of scriptio continua for Hebrew texts, as Alan Millard (1970) has concluded. It is possible, however, that a translator’s Hebrew source text may have had unclear handwriting, or parts of it may have been damaged, which presumably made the task of reading much harder for the translator (Gelston 2002: 499–500). One example of a case of a different Vorlage is found in Amos 1:3, where the phrase “pregnant women” represents a plus in the Septuagint vis-à-vis the MT. Some scholars saw here an intertextual allusion to 4 Kgdms 8:12, where similar wording is found, or perhaps influence from the nearby context in Amos 1:13. However, the discovery of a fragment in Qumran of Amos 1:3 (5QAmos) reveals a Hebrew text that mentions “pregnant women.” I have argued elsewhere that the Septuagint translator of Amos must have had in his hands a Hebrew manuscript similar to 5QAmos. In this case intertextuality must have been introduced into the Septuagint Hebrew Vorlage of Amos 1:3 by the Hebrew scribe who was influenced by Amos 1:13 and attempted to harmonize the two verses. This harmonization, therefore, predates the Greek translator’s work. The Septuagint translator simply reproduces into Greek the intertextuality he sees in his Hebrew text (Theocharous 2012: 109–16).

The Translator’s Enhancement of Intertextuality While the intertextuality found in the Hebrew Vorlage that is simply reproduced in the translation would not count as the translator’s personal intertextual contribution, the enhancement of intertextuality in the source text would. Such an enhancement by the Septuagint translator would show that the translator had actually detected an existing intertextual link and actively contributed to it in its transference to the target text. Cécile Dogniez’s study on Deutero-Zechariah, for example, focuses on this particular aim: to discover whether the Greek translator had recognized intertextual links already found in the MT, even though she includes examples that arise with the Septuagint translator (2005: 82–3). An example of enhanced intertextuality is offered by Robert Hanhart (1992: 360–1), who demonstrates how the translator of Amos 9:13 alludes to the Greek text of Lev. 26:5. While, in my own work in the Twelve Prophets, influence from Greek versions of biblical books was not detectable, Hanhart shows how the translator of Amos was familiar

114

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

with the Greek translation of Leviticus and used it to enhance the existing intertextuality of the Hebrew: Lev. 26:5

Amos 9:13

καὶ καταλήμψεται ὑμῖν ὁ ἀλοητὸς τὸν τρύγητον

καὶ καταλήμψεται ὁ ἀλοητὸς τὸν τρύγητον

OG

And for you the threshing shall overtake the vintage

And the threshing shall overtake the vintage

‫והׂשיג לכם דיׁש את־בציר‬

‫ונגׁש חורׁש בקצר‬

MT

Your threshing shall overtake the vintage (NRSV)

the one who plows shall overtake the one who reaps (NRSV)

While the Hebrew formulation is different in these two texts, the Greek translator of Amos recognizes the allusion and renders the verse exactly as it has been rendered in LXX Leviticus. The Greek translator of the Twelve Prophets never uses καταλαμβάνω for ‫נגׁש‬. Building on Hanhart’s criteria and supplementing them, Ross (2017) has examined the inner-Greek dependence of Judg. 19:23 on Gen. 19:7, demonstrating how the language of the Greek Pentateuch was well known to the Greek translator of Judges and strong enough as to exert influence on his translation choices. Again, the intertextual links are already there in the Hebrew text and demonstrably enhanced by the Septuagint translator. According to James Barr (2003: 541), such cases where the Septuagint renderings cannot be accounted for by common translation practice are the most convincing.

The Translator and His Vocabulary Emanuel Tov (1981) speaks of another type of intertextuality. He has argued that Greek renderings adopted by the translators of the Greek Pentateuch influenced subsequent Greek translators of other biblical books. In other words, the Greek Pentateuch had functioned as a lexicon for subsequent translators who would consult their predecessors’ choices when working on their own translations. This would be a case of “loose” intertextuality since we are not dealing with two distinct texts where one is influencing another, but we are dealing with an entire list of expressions that translators presumably adopted in bulk. Even though one could not make a one-to-one correspondence between specific dependent texts, if Emanuel Tov’s proposal is correct, dependence between the Pentateuch and subsequent texts would still be valid. However, one need not go so far as to endorse Emanuel Tov’s thesis wholesale. It can be demonstrated that in most cases the Greek translators could have come up with those equivalences independently of the Greek Pentateuch, owing to the fact that the bulk of the vocabulary that was chosen reflected the Greek available to the translators in their Hellenistic milieu (Theocharous 2012: 23–66). We have already mentioned that one of the scholarly aims in the study of intertextuality is to see whether later Greek translators were influenced by the Greek renderings chosen by earlier Greek translators. Apart from the Pentateuch that is believed to have been translated first, scholars are not certain about the order of translation of subsequent books. Intertextuality would theoretically be able to give clues as to the chronological order in which each biblical book was translated. However, it is not always easy to determine whether the translator had an earlier Greek rendering in mind, which he borrowed (as

THE SEPTUAGINT AND BIBLICAL INTERTEXTUALITY 115

seen in the examples of Hanhart and Ross), or whether he was influenced by another Hebrew text that he himself translates, resulting in a Greek rendering similar or identical to the Greek intertext. Cécile Dogniez (2005: 89) alerts us to this fact, noting that even when Greek lexical links are present, this could be coincidental if these were words in common use. The Greek Pentateuch, however, is relevant even in our examination of intertextual cases between other biblical books. The reason is that the Pentateuch is believed to have been translated first (see Letter of Aristeas) and is thus regarded as the predecessor of the Greek translations of the rest of the books. Jennifer Dines (2007: 5) stresses the importance of this fact for those cases where a similarity exists between two texts outside the Pentateuch. If that similarity also happens to be found in the Pentateuch, then the possibility of independent influence from the Pentateuch on each of those texts is fairly high. She states her criterion as follows: “a resemblance between two texts must not be traceable to a passage in the Pentateuch used independently by each translator.”

The Translator and His Translation Technique In another criterion for identifying intertextuality, Dines (2007: 5) states that “a resemblance between two texts must not be the outcome of each translator’s normal translational practice.” It is, therefore, imperative to determine the technique each Septuagint translator has, and, according to Anneli Aejmelaeus (2001: 532), it is the study of translation technique that designates “the relationship between the text of the translation and its Vorlage.” However, one should not think of the Greek translators as having a pre-determined approach to the text that they adhere to strictly. It is not until Aquila that one may talk about a preconceived technique.8 Moreover, one should also remember that the Septuagint is not a single translation but a collection of translations with evidence of different types of translation technique. The study of translation technique informs all the stages in the process of identifying intertextuality, and especially the text-critical stage. Emanuel Tov was interested in translation technique in order to determine the usefulness of the translation for textcritical purposes and to this end he developed a list of criteria to measure a text’s literalness (2015a: 22–5). The importance of approaching the Septuagint in this way was also stressed by van der Louw (2007: 8), who states, “If one wants to determine whether a ‘deviation’ stems from a different Vorlage, it should first be excluded that the ‘deviation’ has its roots in translational factors. In order to do so we must know which techniques were used, how and why.” For example, if a translator tends to follow his source text very strictly, then a deviation raises suspicions of a possible variant Vorlage rather than a sudden resorting to a freer translation technique. Emanuel Tov (2015a: 19) describes LXX Job as a very free translation, so he is not surprised when ‫ עם‬is translated as ἄνθρωποι in 12:2. However, the occurrence of the unusual equivalence ‫ עם‬and ἄνθρωπος in Isa. 36:11 is surprising and, according to Tov, may reflect a different Hebrew Vorlage because the historical sections of Isaiah (chs. 36–39) are rendered with relative literalness while elsewhere in Isaiah the Septuagint appears more free.

However, studies in Aquila show that he also made use of various approaches in interpreting his text. See Edwards (2012: 87–106).

8

116

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Studying a translator’s technique, for example, will show whether a translator knows a Hebrew word that occurs more than once in his text. If he does know it, then when a deviation occurs in the rendering of the particular word, it is very likely to have had the purpose of contributing in some way to stylistic goals, adding a theological point or perhaps functioning intertextually. Where it becomes apparent that the translator does not know a Hebrew word, his guesses are also informative. Again, he may be filling in his gap from other texts that he would deem as parallel or relevant to the passage before him. An example of the translator’s ignorance of a word and attempt to interpret it by reference to other texts is found in Amos 5:24. The Hebrew phrase is ‫ כנחל איתן‬that should have been rendered “like an ever-flowing stream.” The Septuagint translator renders the phrase by ὡς χειμάρρους ἄβατος (“like an impassable wadi”). Impassability of the eschatological river is the key characteristic of Ezekiel’s vision in Chapter 47, specifically v. 5, and it seems that this eschatological image provided the source from which the Septuagint translator drew in order to interpret his text (Theocharous 2012: 212–23).

Deviations in the Septuagint Translation The place to look for possibilities of intertextuality, for the most part, is in those instances where the translation deviates from the source text. But one needs to look for other explanations for those deviations and eliminate other possibilities first before interpreting such deviations as intertextual. What other factors are there that give rise to deviations? James Barr (1979: 287) says that deviation in the target text may take different forms: “it may be an original Greek composition of the translator, a reminiscence of some other scriptural passage, a quotation of a non-biblical Greek proverb, or something of the kind.” Add to this list the possibility that the apparent deviation observed in the Septuagint in fact arose from a variant Hebrew Vorlage not available to us, as discussed previously. How do such deviations come about? In the majority of cases, the Hebrew lexeme finds its Greek counterpart relatively easily; for example, ‫ איׁש‬corresponds to ἀνήρ. If for some reason, the translator decides to reject the correspondence he has been in the habit of adopting, then there seems to be an intentional deviation. As van der Louw (2008a: 110–11) has said, a deviation from the Hebrew text would normally happen when the literal rendering of the text is rejected for some reason. However, at times the translator faces difficult words and/or syntactic relationships that he has to work harder to decipher, including ones that he knows will not work in the target language. Here he finds himself in unknown territory, a complexity he has to resolve. He has to explain the unknown by the familiar. He will have to render in the target language what he suspects the text is saying, that is, to whatever he thinks the text refers. One example where the Septuagint translator rejects the literal translation of the text for the reason that it would not work in communicating the supposed meaning of the source text is found in Amos 6:6. The Septuagint translator uses οἱ πίνοντες τὸν διυλισμένον οἶνον for the Hebrew ‫הׁשתים במזרקי יין‬. The literal translation should have been “those who drink from wine bowls.” However, the translator knows that ‫ מזרק‬is a sacrificial bowl that has an exclusively cultic use (see LXX Zech. 9:15, 14:20). How, then, could sacrificial bowls be “wine bowls” at the same time? The construct chain would need to be clarified so that Amos’s accusation is communicated to the reader. If he were to use φιάλη (“bowl”) and translate “those who drink from wine bowls,” that would hardly make for a prophetic accusation. The translator has to find a way to communicate in the target language something that would analogically be an accusation

THE SEPTUAGINT AND BIBLICAL INTERTEXTUALITY 117

that Amos would have made against his audience. By using οἱ πίνοντες τὸν διυλισμένον οἶνον, the translator suggests that an elite group of people is indulging in the best quality of wine, thus showing how Amos is attacking excessive luxuries (Theocharous 2012: 131–42). One should not rush to make judgments upon the translator’s motivation for these divergences. There is more than one possibility available to the translator for solving a problem, and all options should be scrutinized before claiming that ideological or theological motivation underlies a transformation. In the majority of cases, the norms of the target language demand a certain transformation. An example of one such demand in the target language is found in Zech. 11:17, where the LXX reads: ὁ βραχίων αὐτοῦ ξηραινόμενος ξηρανθήσεται

His arm withering shall wither

καὶ ὁ ὀφθαλμὸς ὁ δεξιὸς αὐτοῦ ἐκτυφλούμενος ἐκτυφλωθήσεται

His right eye blinded shall be blinded! (NETS)

What the Septuagint translator has done was to pair participles with future passive verbs in the third person singular of the same root. The respective Hebrew, however, consists of pairs of infinitive absolutes with yiqtōl verbs: ‫ זרעו יבוׁש תיבׁש ועין ימינו כהה תכהה‬Let his arm be completely withered, his right eye utterly blinded! (NRSV)

The infinitive absolute in Hebrew, when it accompanies a cognate verb, “affirms and intensifies the authenticity or conviction of the verbal action” (Arnold and Choi 2003: 74–5). The way the Septuagint commonly reflects this Hebrew construction is not by direct imitation but by the use of what is called the “intensive” participle. The participle is “used along with a finite form of the same verb, to convey the intensive force that is accomplished in Hebrew by the addition of the infinitive to the finite verb” (Conybeare and Stock [1905] 1995: §81). Often, the reason behind a transformation may have been a simple misreading of the text (such as mistaking a ‫[ ד‬dalet] for a ‫[ ר‬resh]). But even in those occasions one needs to be careful because the translator may manipulate consonants in order to produce the meaning he feels was intended. One example is the reading ‫ גזי המלמך‬as ‫ גוג המלך‬by the LXX translator in Amos 7:1, thus rendering “Gog, the king” rather than “the king’s mowings.” The eschatological expectations associated with the figure of Gog were strong at the time the Septuagint translation was produced and, in addition to the military reference at the end of Amos 6, must have stirred the translator to see this vision as referring to king Gog and turn the three consonants of ‫ גזי‬into ‫( גוג‬Theocharous 2012: 228–30). Lastly, we must keep in mind that when we are trying to determine the direction of influence between two texts, it is the text where the Hebrew is manipulated that would be dependent on the other. Jennifer Dines (2007: 5) explains that “one rendering must fit its presumed Hebrew, and thus its context, more appropriately than the other, so that dependence is plausible only in one direction.” Only after various possibilities not pointing to interpretive coloring by the translator have been eliminated may one look for an explanation in the translator’s interpretation of the biblical text. Only then should the use of intertextuality be considered, and indeed, we are justified to consider it because it was a common practice in ancient writings.

118

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Intertextuality as Common Ancient Practice Various scholars have observed this practice in the Septuagint as well as in other ancient texts. Marguerite Harl (2001: 192) notes that at times a translator decides to incorporate “analogical (‘intertextual’) interpretations, due to the links with parallel passages elsewhere in the Septuagint. This method of interpreting a passage by reference to another one within the same work has been practiced in antiquity for all great writings.” Similarly, Robert Hanhart (1992: 358–9) observes that sometimes the translators explained the Hebrew original from analogous formulations in the OT witnesses themselves; it is thus a question concerning the possibility of illuminating such analogous statements by translating in a way that was not explicitly represented in the statement of the original … The act of translation leads necessarily to the question about analogous formulations in the context of the original as a whole as received by the translator. These parallel passages may be quotations, thematic allusions, and sometimes a single word association between two texts may place them in a parallel relationship in the eyes of the ancient reader or translator, as Friedrich Avemarie (2009) has observed in Qumran writings and the NT. George Brooke (1998: 56) has also examined intertextuality in Qumran texts and in the NT and notes that, the scrolls found at Qumran attest how scribes in copying its books often behaved intertextually themselves, introducing phraseology that was reminiscent of other passages of scripture. This may happen both deliberately as two scriptural texts with related subject matter are associated with one another; or it may happen unconsciously as the idiomatic phraseology of one passage comes to influence the scribe as he works on another. Many scholars have concluded that the Septuagint translators, like authors of Qumran writings and NT texts, did not have as limited a scope of context as modern interpreters do. Emanuel Tov (2015a: 27) writes that, “it should be stressed that the translators’ concept of context was more comprehensive than ours. They referred not only to the relationship between the words in their immediate context but also to remote contexts. Furthermore, the translator might introduce any idea that the source text called to mind.”9

ONGOING RESEARCH QUESTIONS Since intertextuality as a technical term has been used in a variety of literary areas and with different aims each time, it is imperative that scholars embarking on the study of intertextuality first define what exactly they are discussing and what they regard as intertextual. Indeed, there are many areas in which this line of research could expand. First, there is intertextual influence at the lexical level. Has a certain translator created equivalences between Hebrew and Greek words in a way that has influenced subsequent translators, as Emanuel Tov (1981), and others before him, have suggested? Second, at the level of a phrase or idiomatic expression, one could also examine whether the coining of such phrases was produced by a Greek translator and subsequently

Also, Brooke (1985: 8–79, esp. 28 and 36); Jones (1995: 9); Troxel (2008: 291).

9

THE SEPTUAGINT AND BIBLICAL INTERTEXTUALITY 119

influenced others, or whether these stock phrases were already circulating in the translator’s world (Theocharous 2012: 67–106). Third, there is the category of intertextuality that is activated on the basis of so-called catchwords. Catchwords are certain words or phrases in the Hebrew Vorlage that manage to generate for the translator a connection with other biblical texts where the same words or phrases are found. The Septuagint translator, having been reminded of other biblical passages by means of one or more of these catchwords, let the passage or passages he remembered influence the way he translated his source text (Theocharous 2012: 107–95). Lastly, intertextuality may take the form of allusions to specific biblical stories, events, or characters, but it is not triggered by a shared catchword between the Septuagint Vorlage and another Hebrew text, as was the case in the previous category. The pull does not come from the translator’s Vorlage, but “in these cases the intertext exerts a stronger pull than does the Hebrew text being translated” (Theocharous 2012: 196; see also 197–239). For future scholars interested in this field, there is still a lot of ground to be covered with respect to the study of intertextuality in the Septuagint, both on inner-biblical influences and on intertextual influences from extra-biblical Greek texts. Another related area for future research would involve the study of intertextuality in Septuagint reception. The language of the Septuagint may have created different connections in the mind of patristic writers than would have been made merely from the study of the Hebrew text.

RECOMMENDED RESOURCES 1. Cook (2006a: 119–34). Cook explains how intertextuality works in the Septuagint translation with particular examples and proposes that translators made use of external exegetical traditions in an endeavor to make the text clear to their Jewish audience. 2. Dogniez (2005: 81–96). Dogniez investigates intertextual echoes in the Greek text of Zechariah and raises important methodological questions with respect to intertextuality coming from the Hebrew text, from other Greek texts, or simply from misunderstandings and coincidences. 3. Schaper (2006: 371–80). Schaper offers an example of how intertextuality may reveal the translator’s messianic interpretation of texts. 4. Theocharous (2012). The purpose of this work is to examine a broad spectrum of intertextuality under various divisions of types of intertextuality in the Greek text of the Twelve Prophets. Cases that appear intertextual are thoroughly examined to determine whether there is indeed some influence on the translator or whether other factors would offer better explanations for the phenomena observed.

120

PART THREE

Text

122

CHAPTER 8

The Septuagint and Textual Criticism of the Greek Versions JOSÉ MANUEL CAÑAS REÍLLO

INTRODUCTION Textual criticism of the Septuagint has two possible orientations. One views the text as a means of reconstructing the Hebrew text of which it is a translation (e.g., Tov 2015a). The other orientation views the text of the Septuagint itself as its object, aiming to reconstruct its oldest stage and describe the different stages of its textual history. This chapter will deal with the latter.1 The basis for textual criticism is analysis of the available sources, of which there are three types, in order of importance: 1. The direct tradition (Greek manuscripts); 2. The indirect tradition (quotations of the Septuagint by later Greek authors); and 3. The secondary versions translated in antiquity from the Septuagint tradition. Textual criticism thus requires many disciplines, such as paleography, codicology, and linguistics, as well as a detailed study of and familiarity with the range of Septuagint translations styles. Of course, textual criticism of the Septuagint must also account for the Hebrew background of the Greek text recognizing that, although the Septuagint was mostly transmitted independently, the textual history of Hebrew Bible and its Greek translation did not unfold in total isolation from one another. There were constant interactions between the two traditions, and these interactions must be described as best as possible in order to understand how, why, and under what conditions the Septuagint text evolved. Moreover, some issues in Septuagint textual criticism, such as the Kaige recension and the relevance of the Qumran documents, require recourse to the Hebrew for reference.2 Finally, knowledge of the other languages in which the secondary versions were composed—such as Latin, Syriac, Coptic, Armenian, Georgian, Ethiopic, and Arabic—is also important for Septuagint textual criticism. On the other orientation—the use of the Septuagint for textual criticism of the Hebrew Bible—see Screnock in the present volume. 2 On the relationship of the Septuagint to Qumran, see Kotzé in the present volume. 1

124

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

APPROACHES AND THE HISTORY OF RESEARCH The discipline of textual criticism is almost as old as the text itself, in that it has been present from the moment the first Greek manuscripts were copied. The copying process inevitably introduces changes in the text, whether intentional or unintentional. Textual criticism attempts to restore the text prior to such changes whenever possible. The first well-known example of Septuagint textual criticism is Origen’s Hexapla, which applied the parameters of the Alexandrian method—originally created for secular literary Greek texts—to the Greek scriptures. Origen found that the contemporary Septuagint manuscripts differed not only among themselves but also with the extant Hebrew text itself. He sought, therefore, to restore the Septuagint text to an earlier stage by means of Hebrew as a point of reference. The Hexapla was thus a precursor of modern textual criticism because, although it did not incorporate a critical apparatus with variant readings like modern editions, it did provide the reader with guidelines for comparing Origen’s received Septuagint to other Greek texts (especially Aquila, Symmachus, and Theodotion) and with the Hebrew (Neuschäfer 1983: 86–103).3 Ironically, rather than restoring the text of the Septuagint, the Hexapla contaminated its textual history and ultimately made it very difficult in certain cases—such as Greek Judges—for modern textual critics to reconstruct the oldest stages of the text of the Septuagint.4 The first printed edition of the Septuagint was the Greek text of the Complutensian Polyglot (1514–17). In the sixteenth century two other main editions were also issued: the Aldine (1518) and the Sixtine (1587). All three show some attempts at textual criticism, although we do not always have a clear idea of the guidelines used by the editors. Unlike the Vulgate editions, in which a very stable textus receptus predominates starting from the editio princeps (Gutenberg’s edition), in early modern Septuagint editions there is great diversity resulting from the use of different manuscripts as their textual basis. The Greek text of the Complutensian was based on manuscripts from Rome and Venice that attest the Lucianic recension. The Aldine was also based on manuscripts, but its text differs from that of the Complutensian. The Sixtine presents a text based on the Aldine but revised against Vaticanus (B) and other manuscripts (Kreuzer 2018: 130–1). The Sixtine was thus the first attempt to provide a truly critical edition of the Septuagint, as it incorporates annotations with the scholia of Petrus Morinus (Pierre Morin) and Flaminius Nobilius. These scholia are the “varias lectiones variasque interpretationibus” (“variant readings and variant interpretations”) that had been collected by Morinus since 1578. They attest marginal notes drawn from old manuscripts with the texts of Aquila, Symmachus, and Theodotion, from the Old Latin versions, and from the Fathers. These notes were included anonymously in the 1587 Sixtine, and in 1588 were reprinted in the Latin edition of the Bible as part of the annotations of Flaminius Nobilius (Jellicoe 1968: 127–8). The Sixtine edition became the textus receptus in later centuries, reprinted in the editions by Holmes and Parsons (1798–1824) and Tischendorf (1850), and it prevailed as the accepted Septuagint text until the mid-nineteenth century. At that point, the Sixtine fell into disfavor owing to the development of Lagarde’s eclectic method of Septuagint textual criticism, which posited that all extant manuscripts preserve a mixed text (Jellicoe 1968: 6–7). With this new development, other factors came into play in Septuagint textual criticism, such as the importance of translation style and weighing variant readings by On the Septuagint and the major recensions, see Mäkipelto in the present volume. On the Septuagint and the Hexapla, see Gentry in the present volume.

3 4

TEXTUAL CRITICISM OF THE GREEK VERSIONS 125

means of quality and not quantity. At the same time, other factors diminished, such as the importance ascribed to the Hebrew text. Lagarde also posited the reconstruction of the Greek recensions following Jerome’s trifaria varietas (see Prologus Sancti Hieronymi in Libro Paralipomenon in Weber 1994: 9–12, 546) as the key to Septuagint textual criticism (Klein 1974: 9–10). These recensions include the Hesychian (from Alexandria), that of Lucian Martyr (from Constantinople), and the Palestinian (in Origen’s recension). Pursuing this hypothesis, Lagarde (1883) published the text of the Lucianic recension as the first step of his project. Lagarde’s approach to textual criticism was the foundation for the project based at the Septuaginta-Unternehmen in Göttingen, Germany, originally headed by Alfred Rahlfs. This project was for decades carried out within the frame of the Göttingen Academy of Sciences and Humanities (1908–2015) and now continues as the Kommission zur Edition und Erforschung der Septuaginta. Its main task has been the reconstruction of the earliest text of the Septuagint in the form of an eclectic edition with a critical apparatus, though this task is in fact a departure from Lagarde’s initial proposal. Still, Lagarde’s heritage is still part of the so-called Göttingen Septuagint, as well as Margolis’s edition of the Greek Book of Joshua (Margolis 1931–8; Margolis 1992). Margolis tried to establish his edition of Greek Joshua on the basis of the geographical distribution attested in Jerome’s trifaria varietas, and thus established five recensions (Margolis 1992: vii–ix): the Egyptian (E), the Syrian (S), the Palestinian (P), the Constantinopolitan (C), and a collection of other textual groups and individual manuscripts (M). This text-critical procedure, while interesting, assumes a geographical and temporal relationship among texts and textual groups that cannot be proven on the basis of known Greek textual transmission. Although it was never finished, it is also necessary to mention the Cambridge edition of Brooke and McLean (1906–40), who opted for another text-critical approach. This project instead presented a diplomatic text, in this case Vaticanus (B), accompanied by a critical apparatus containing the evidence of other Greek witnesses and secondary versions. Until the late 1940s the only available witnesses for textual criticism of the Septuagint, aside from the major codices, were Greek manuscripts, numerous papyri, and quotations by Flavius ​​Josephus and Greek Christian authors. Only two fragmentary witnesses were known to date earlier than the Christian era: P.Ryl. 458 (Ra 957) from the second century BCE, and P.Fouad 266 (Ra 848), c. 50 BCE. Any textual comparison with Hebrew had to be done with the only text available at that time, namely, the Masoretic Text (MT), a standardized text more than a thousand years younger than the Septuagint. However, with the discovery of the Qumran manuscripts, scholars obtained numerous witnesses to a Hebrew Text far older than the MT and, perhaps, closer to the Septuagint Vorlagen. These new sources radically changed the textual landscape of the Septuagint and all previous text-critical research had to be reconsidered. Old problems were resolved, but at the same time new problems arose, breathing new life into this area of research and putting Septuagint textual criticism back at the center of scholarly attention.

RESEARCH METHOD The Sources As a point of contrast, textual criticism of classical Greek literature deals with a very small number of witnesses. Moreover, they are mostly literary compositions, not translations. For such texts, copyist mistakes constitute the primary basis for a textual critic to establish

126

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

groups, making the task of producing a stemma codicum more or less straightforward. A stemma codicum is a picture of the history of a text in the form of a tree diagram with branches representing collective points of departure from the known codices. When the stemma codicum is completed, it results in a textual archetype that, in theory, is the original (reconstructed) Greek text on the basis of all extant evidence. But in the case of the Septuagint, the situation is very different. The scholar is faced with an enormous amount of manuscripts that may be dated anywhere from antiquity to the end of the middle ages, making the construction of a stemma codicum almost impossible. Generally speaking, witnesses to the Septuagint show great typological diversity. There are two main kinds of written Greek sources: papyri, and book-format (codex) volumes. The latter can be divided into uncial (the oldest evidence) and minuscule. Distinguishing format, material, and style of writing (paleography) in sources is important because each has distinct characteristics and problems related to textual transmission. The typology of copyist mistakes in an uncial manuscript, for example, is not the same as that found in minuscules, and a significant portion of copying errors in the middle ages originated from the reading of uncials. Similarly, textual transmission in papyri shows that some larger scale literary variations in the manuscript tradition may have resulted from misplacing pieces in the process of arranging the papyrus roll. As in all textual traditions in antiquity, copyists sometimes intended to correct and/or improve their copy, despite the Septuagint often being considered sacred. Thus, the copyists often changed what they considered a mistake by updating or standardizing the spelling of a text, introducing unintentional changes by conceptual association (e.g., a proper name for pronoun), as well as introducing marginal glosses into the body of the text, which sometimes produced doublets. Copyists also made mistakes in the ordinary copying process, such as haplography, dittography, omissions due to homeoteleuton and homeoarcton, as well as mistakes due to visually or mentally confusing letters. Sometimes the latter two may be explained in terms of the internal mental pronunciation of the copyist in the reading process, which could lead to itacism or the interchange of long and short vowels (ω/ο and η/ε) or single and double consonants. Detecting such phenomena and explaining their causes and origins are tasks of first importance in Septuagint textual criticism. It is also necessary to be aware that, in older stages of transmission, texts were copied without word division (scriptio continua), and so abbreviations abounded, and the text often lacked accents and aspiration marks. It is during this older period of the textual history of the Septuagint that many of the copyist mistakes perpetuated in later Greek manuscripts originated. Hence, the textual and chronological (but not always critical) importance of uncials and papyri. In order for any variant to have critical value it is essential to explain its origin and development. In this connection, Walters (1973) has dealt in detail with the typology of copying mistakes and the tools to address them. Variations in the formal characteristics of Septuagint manuscripts also have criticaltextual implications. The Septuagint was transmitted in two basic formats in the manuscript tradition: one containing the biblical text only, and the other containing the biblical text accompanied by patristic commentaries. The latter are known as catena manuscripts. For a general study on the catenae, see the still-valuable work of Karo and Lietzmann (1902). Publication of these catena texts is slow and difficult work, but it is also important for Septuagint textual criticism, since these sources may preserve material attested nowhere else. Certain catenae have introduced changes within textual history insofar as the biblical text has been influenced by citations in the catena commentary (accurate or not) and, over the course of the transmission of a given catena, vice versa.

TEXTUAL CRITICISM OF THE GREEK VERSIONS 127

Editions and Recensions The text of the Septuagint and its transmission are characterized by plurality and fluidity. Textual diversity is present in all books, although to different extents. Each book presents its own peculiarities, and, above all, the process of producing critical editions has shown that the textual history varies from book to book, and even within a single book. The tendency to adapt or emend the Greek text toward Hebrew is particularly common. An example here is the Kaige recension, in contrast to the Christian recensions such as the Lucianic/Antiochene and Hexaplaric, which lack the Hebraizing impulse. Consequently, there is often a dual tendency within Septuagint textual transmission, as in the Historical Books where, on the one hand, there is Codex Alexandrinus (A) with its non-Hebraizing Lucianic/Antiochene and Hexaplaric texts. Yet within the same books there is, on the other hand, evidence for the Hebraizing Kaige and Semi-Kaige recensions (Kreuzer 2016: 56–61). Besides the main textual groups (e.g., the Lucianic and Hexaplaric recensions) in some books of the Octateuch (Genesis to Deuteronomy and Ruth), the Editio Maior of Göttingen has isolated several minor textual types that vary from book to book. An example is the so-called t-group. From Genesis to Ruth the t-group text is found in MSS 74 76 134. MS 106 appears in Ruth as t-group evidence, but in Deuteronomy it is included in the d-group. In Genesis, Exodus, Leviticus, and Deuteronomy, MS 370 is part of the t-group, and to it MS 799 is added for Genesis, Deuteronomy, and Ruth. However, MS 134 in 2 Chronicles is included in the so-called a-recension, while in Judges it is included in the Lucianic group. In the deuterocanonical books the picture changes considerably. In those books there is no Hexaplaric or Kaige-type recension. Instead, in 1 and 2 Maccabees there is a basic duality between a Lucianic text that is often difficult to identify and the so-called q-recension. In this group is found MS 134, but it is part of the a-group in 2 Chronicles and of the t-group in the Octateuch. The upshot of this detailed information is to highlight the need to individualize the textual study of each book rather than generalize about the value of a given witness for the entire Septuagint. For the status quaestionis of textual criticism book by book, Aitken (2015b) and Kreuzer (2016, 2019) provide essential data.

The Septuagint as Translation and Adaptation A second difference with respect to the study of classical Greek literature is that, for Septuagint textual criticism, the working method must be adapted to two types of texts: translations from the Hebrew (i.e., protocanonical), and writings originally composed in Greek (as is the case with most of the deuterocanonical books). For books translated from Hebrew, textual criticism must deal with several factors at once, such as the approximate date of translation (if known), the characteristics of the Hebrew Vorlage, the style of translation into Greek, and even whether it is more of an adaptation than a translation. Distinguishing between the Greek Pentateuch and all other books is fundamental, because the former was a venue for experimentation in vocabulary choice and translation method that would be imitated and developed in later translations. It is also important to take into account whether the Hebrew text that lies behind a given translation is narrative or poetic, and the theological role that each book of the Septuagint plays within the whole of the Bible. For instance, the theological significance of the books of the Pentateuch, the Psalter, or the Prophets is not the same as that of Ecclesiastes. Generally, the translation technique chosen for books with narrative content tends to be

128

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

more word-for-word in style; for books of a legal or cultic nature, such as Leviticus and Deuteronomy, a very stereotyped translation is more common. In general, for the Poetic, Wisdom, and Prophetic books, the translators adopted a much more relaxed posture, developing a set of more freely translated texts, such as in Job or Proverbs. Both the linguistic features and the translation technique of each book require detailed study as a fundamental part of the work of textual criticism. The relationship of the Greek text to the Hebrew is not limited to the initial moment at which the translation of each book occurred. Ever since the pre-Christian era, Hebrew has constantly influenced the history of the Greek text, and many of the changes detected in the latter may be due to comparison with and modification of the former. A clear example of this phenomenon is the Kaige recension, a general movement of revising the Greek text toward a Hebrew source during different periods of the textual history of the Septuagint. The limits of this tendency are still not well defined, and the particularities vary from book to book in the same way that its manifestation in textual history varies case to case. Of course, this does not mean that every attempt to revise a Greek text toward Hebrew has to be considered part of the Kaige recension, particularly as this label still has a broad and imprecise sense.

Indirect Tradition: Patristic Quotations by Greek Writers A third feature of these texts is that they were of central importance to the culture in which they developed and were used. The Septuagint provided a variety of themes and images that later writers incorporated into their works as allusions or quotations, for example. Biblical quotations in Greek authors also allow scholars to situate the use of a certain text-type in a geographic region and temporal period. Thus, quotations from the Septuagint in Philo of Alexandria or in Flavius ​​Josephus, for example, demonstrate that these authors knew a certain textual form that was pre-Hexaplaric—in the case of Flavius Josephus, one very close to a Lucianic text and similar to what appears in the New Testament (Spottorno 2013). In the same way, the citations of Theodoret of Cyrus and John Chrysostom provide fundamental criteria for isolating the Antiochene text in the editions of Samuel, Kings, and Chronicles. The use of biblical quotations in textual criticism poses methodological problems and requires some precautions. Preliminary study is necessary to determine whether the quotations tend to be literal—whether taken directly from a written source or from memory—or adapted to the context (i.e., allusions). For textual criticism, the former— literal quotations drawn directly from a written source—have the highest value. Allusions, on the other hand, must always be considered with caution. A special problem in this area is presented by the writing of Flavius Josephus, ​​ where literal quotations are scarce, but some allusions and proper names show points of agreement with the Lucianic recension, or, in this case, Proto-Lucianic. That is, Josephus must have known a pre-Hexaplaric form of the Septuagint. For the list of patristic quotations arranged by reference, a very useful resource is the Biblia Patristica (BiPa 1975–2001). However, given the date of its publication, its information must be consulted alongside the most recent editions.

Secondary Versions In early Christianity the Septuagint served as a basis for translations into other languages, such as Latin, Coptic, Ethiopian, Armenian, Georgian, and Arabic. In other cases, only a part of the Septuagint (particularly Origen’s recension) served as the base text for a

TEXTUAL CRITICISM OF THE GREEK VERSIONS 129

translation, as with the Syro-Hexapla. The versions are generally word-for-word in style, and thus confirm the testimony of the Greek recensions and facilitate more precise knowledge of the textual history of the Septuagint. The edition of Holmes and Parsons (1798–1827) already incorporated into its critical apparatus the evidence of these versions (plus the Slavonic or Paleo-Slavic). In the fascicles of the Editio Maior of Göttingen these versions play a more fundamental role, especially since J. W. Wevers demonstrated their value in his fascicles of the Greek Pentateuch. The use of these versions also requires caution since they have their own textual history that is not often well understood. The ideal situation would be to have highquality critical editions for all of them. However, this has not yet been accomplished, and it is often necessary to collate manuscripts of the versions to supplement the testimony of extant editions, which in some cases are very old and unreliable. The most valuable secondary version for Septuagint textual criticism is the Vetus Latina, sometimes called the Old Latin version. For this version, there are up-to-date critical editions for some biblical books, published in the Vetus Latina series (1954–), but for other books the researcher must resort to other editions of manuscripts (in some cases very old) and to quotations of Latin writers, including, for example, Tertullian, Cyprian, and Lucifer of Cagliari (each of whom offers pre-Hexaplaric evidence). One may also consult Augustine, Rufinus, and Jerome as well (for their value to the textual history of the Latin Bible and indirectly to the Septuagint also). Additionally, other versions, such as the Coptic, the Syro-Hexapla, the Ethiopian, the Armenian, the Georgian, and, in some cases, the Arabic, are important, not as evidence for the original text of the Septuagint itself, but for its textual history and especially for the status of the recensions in the fourth and fifth centuries, such as the Hexaplaric and the Lucianic.5 The critical value of the versions lies in their support for variant readings attested by Greek manuscripts or the recensions. They must not be used to restore Greek readings that are not attested by the manuscript tradition itself. Less cautious practices prevailed in the past as, for example, in the critical apparatus of Holmes and Parsons (1798–1827). For contemporary textual criticism this procedure is unacceptable. The Greek evidence is the only sure basis for restoring the text.6

Unity and Plurality One of the peculiarities of the Septuagint text is its plurality—that is, the evidence of different textual types generated in the process of its transmission. These types are the result of ancient editorial processes carried out for different purposes, such as adaptation or emendation in relation to a given Hebrew text (the Kaige recension), modifying linguistic style and updating vocabulary (the Lucianic/Antiochene recension), or for comparing the text to Hebrew and other known Greek texts (the Hexaplaric recension). Rigorous analyses of collations of Greek manuscripts can isolate textual groups in terms of agreements and divergences. For instance, MSS 19 82 93 108 and 127 transmit the text of the Antiochene recension in Samuel (Fernández Marcos and Busto Sáiz 1989), and MSS Z and 700 are added to this group for Kings (Fernández Marcos and Busto Sáiz 1992), while in Chronicles (Fernández Marcos and Busto Sáiz 1996) the group is reduced On the use of the Coptic and Armenian versions in the textual criticism of the Septuagint, see Wevers (1985) and Cox (1985), respectively. For an updated bibliography on the eastern versions of the Bible, an interesting overview is that of Van Esbroeck (1998). 6 On the Septuagint and the secondary versions, see Cox in the present volume. 5

130

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

to MSS 19 93 108 and 121. In Daniel, the Lucianic text is found in MSS 22 36 48 51 96 231 and 763 (Ziegler 1999), while in Esther it is attested in MSS 19 93 108 and 319 (Hanhart 1983a). The textual character of the manuscripts varies from book to book. Isolating textual groups is the first and fundamental step of textual criticism, as it will establish a framework for stratifying the textual history of a Septuagint book, which may then be confirmed or modified by other data. The evidence provided by the manuscripts is not enough for textual criticism. Although it is to be considered decisive, these direct witnesses must be compared with indirect evidence that complements it and helps to confirm the stratification of the Greek text into groups. The recensions often impede this goal, however, especially in books where influence from the Kaige or Hexaplaric recensions is pervasive within the greater part of the manuscript tradition. Research has found in the Lucianic/Antiochene text a pre-Hexaplaric stage (proto-Lucianic) that—when it has the support of the Vetus Latina evidence—constitutes a pathway to the earliest stages of the Septuagint text (e.g., the book of Judges). Detailed knowledge of the recensions—their origins, development, and significance in the textual history of a given book—is extremely important for textual criticism of the Septuagint. Without the versional evidence, which is sometimes difficult, if not impossible, to obtain, full access to the textual history of the Septuagint is not possible. This is what Lagarde realized in the nineteenth century when, recognizing that all Greek manuscripts present eclectic texts, he proposed the recensions as the solution.

ONGOING RESEARCH QUESTIONS Septuagint and Qumran The discovery of the Qumran texts had two distinct consequences for Septuagint textual criticism: an internal one for the textual history of the Septuagint as a version of the Hebrew Bible, and an external one for Septuagint as translation in relation to its Hebrew source text. The Qumran documents provided new Greek witnesses from the preChristian period that have considerably changed the textual landscape. Prior to Qumran, the Rylands and Fouad papyri mentioned above were the only available pre-Christian texts. But now many biblical books have such early witnesses. For example, from only two caves (4 and 7) new evidence emerged for Exodus, Leviticus, Numbers, Deuteronomy, and the Letter of Jeremiah. Scholars agree that the Qumran texts support Lagarde’s hypothesis against Kahle concerning the early textual history of the Septuagint, and that there were revisions long before Origen produced his Hexapla. These revisions were not therefore an exclusively Christian endeavor, but also occurred in the Jewish milieu. One of the greatest developments for textual criticism of the Septuagint took place with the discovery of the Naḥal Ḥever manuscripts, with which Barthélemy (1963) identified characteristics of the Kaige recension. This opened new possibilities for explaining old text-critical questions, particularly with reference to the role of the B codex in the books of Samuel and Kings, in Judges, and to the text of Daniel traditionally attributed to Theodotion. From the point of view of external implications, it is important to note that most of the Qumran documents are writings in Hebrew and Aramaic, and that for some biblical books they provide a previously unknown text in these languages. These witnesses are written in the Hebrew consonantal alphabet, texts very similar to those that the Septuagint

TEXTUAL CRITICISM OF THE GREEK VERSIONS 131

translators must have used as Vorlagen. This Hebrew text sometimes attests differences from MT, and in such cases the Septuagint sometimes supports the former against the latter. This phenomenon has partially overturned an old assumption that Septuagint readings different from MT could be explained as a result of a different exegetical tradition. Nevertheless, the differences between the Septuagint and MT involve not only textual variants, but are sometimes literary in nature, as in Samuel and Jeremiah. In select cases, there are indications that the Greek translation was carried out at a time when its Hebrew Vorlage had not yet been fully established.

The Linguistic Factor An aspect of framing the Greek text as the focus of Septuagint textual criticism—that is, not only as a witness for Hebrew Bible textual criticism—has been the incorporation of linguistics. Before the twentieth century, the only available evidence for linguistic study of Septuagint was Hellenistic Greek literature. That situation changed with the discovery of the Egyptian documentary papyri, whose language, unlike literary texts, was nonliterary Greek. This discovery provided a new body of evidence for research of the language of the Septuagint using texts written in the same period and in the same geographical context as the first Septuagint translations, namely, Egypt. This development prompted reevaluation of many previous theories, such as the notion of ‘Biblical Greek,” often applied too freely to the language of the Septuagint without taking into account that it is a multifaceted reality in itself with many variations. The most recent research has identified the Greek of the Septuagint as the Greek in conventional use at the time when it was translated from Hebrew, but with the limitations of a text translated from another language. As a result of this research, the degree of perceived Semitic influence upon the Greek of the Septuagint has attenuated. Many of the phenomena involved in semantic change, for example, were under way for Greek words attested in both the Classical and Hellenistic periods, and thus in the Septuagint could be explained as result of internal developments of the Greek language. Still, it is true that in some cases the explanation for lexical semantic change is found in the influence of a Semitic (Hebrew or Aramaic) source. This perspective has profound significance for the textual history of the Septuagint, especially in analyzing vocabulary changes in textual transmission, for instance, in the Kaige recension. In this respect, the research of John A. L. Lee (1983) in the late 1960s on the vocabulary of the Greek Pentateuch, and, more recently, the work of James K. Aitken (2014b) and his “Cambridge School” (Ross 2018a) on the syntax and lexicon of the Septuagint within the history of Greek as a language have had a deep influence upon textual criticism. Lee proved that his method of dating the vocabulary of the Pentateuch would also be valid for Judges, where A would show older vocabulary than B (Lee 1983: 148). Following the same methodology, Ross (2018b) has recently proved that the same observation is also applicable to Judges.

The Lucianic/Antiochene Text The long history of research on the Lucianic text, sometimes valued more for Septuagint textual criticism by some (e.g., Lagarde) than others (e.g., Rahlfs), took a turn in the 1980s  with the identification of the so-called Antiochian Recension by the “Madrid School” of textual criticism of the Spanish National Research Council (CSIC). Traditionally, scholars would isolate the Lucianic recension in a group of manuscripts and in the Greek text of the Complutensian Polyglot. As mentioned above, Lagarde (1883)

132

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

published the first attempt at an edition of the Lucianic text as a means to access older stages of the Greek text in the Octateuch and Historical books, but subsequent research has shown that there are differences from book to book, that a given manuscript does not always contain a Lucianic text for each book, and that even within a particular book there may be changes from one section to another. As a result, scholars now recognize that it is very difficult to isolate a Lucianic text in the Octateuch, Ruth being the first book that shows the clear presence of this type of text. Current critical editions have detected the text of the Lucianic recension in the Prophets, Maccabees, Judith, 1-2 Ezra and Job, and with less clarity in Wisdom and Sirach. The reconstruction of the Lucianic/Antiochene recension and its different stages (especially proto-Lucianic) rests upon two types of criteria: internal and external. Internal criteria include the tendency to introduce certain stylistic and linguistic corrections, and, in some cases, changes of content, but with variations from book to book. External criteria include the support of the indirect tradition (e.g., quotations of Theodoret and Chrysostom) and secondary versions such as the Vetus Latina. The books where textual reconstruction is possible are Samuel, Kings, and Chronicles. In these books the manuscripts that preserve the Lucianic/Antiochene recension constitute a well-defined group. The most recent research has also postulated in the Lucianic/Antiochene text of these books not only a revision, but a recension that would adhere to specific guidelines, such as the elimination of Hellenistic Greek forms by replacing them with Attic forms, variations in vocabulary, elimination of Semitisms, and the (re)introduction of the perfect and the historical present verb forms as a result of the Atticist movement of the second century BCE (Fernández Marcos and Busto Sáiz 1989: xxviii–xxx). There was also literary reworking, for example, to extend the book of 2 Samuel up through the death of David and the beginning of Solomon’s kingdom in 1 Kings 2:22; to omit the account in 1 Kings 16:34 of the construction of Jericho; the elimination of the story of Jehoshaphat from 1 Kings 22:41-52 since he had already appeared in 1 Kings 16:28a–h; and the introduction of a negative judgment on King Asa in 1 Kings 15:23. There are even stylistic modifications apparently meant to fulfill what has been predicted previously in the narrative (Fernández Marcos and Busto Sáiz 1989: xxx–xxxii; Fernández Marcos 2013: 63–71). Present too are tendencies to clarify the content, corrections of Midrashic character, and combination of alternative readings that give rise to doublets. Scholars have given this edition/recension the label “Antiochene” because it is related to the biblical school at Antioch and is a text prepared for public reading in the Christian context. But caution is necessary to avoid haphazardly applying the Antiochene label— originally used for Samuel, Kings, and Chronicles—to the Lucianic text of other books like Joshua and Judges, since for those books current research does not allow such an identification. Research on the Lucianic/Antiochene recension is currently one of the topics generating the most literature in the field of Septuagint. Its importance goes beyond this recension itself, though, as mentioned previously, it can provide a pathway to oldest stages of the text. For example, recently Kreuzer (2013: 51–3) has focused on the relationship between the Lucianic/Antiochene and Kaige recensions.

The Septuagint and the Secondary Versions As discussed above, not all the secondary versions have the same weight for current textual criticism. Their value depends on the status quaestionis for each of them and

TEXTUAL CRITICISM OF THE GREEK VERSIONS 133

on the availability of modern editions that can be used as the main textual basis for comparison. Thus, sometimes the Septuagint textual critic is forced to undertake textual study of a given version themselves before using it. This challenge has too often hindered incorporating some of these versions into Septuagint textual criticism, and scholars have not always been able to fully benefit from their potential. This is the case for the Vetus Latina, the Ethiopian, the Armenian, and the Georgian versions. Ever since the edition of the Antiochene text for the Historical Books, the Vetus Latina has become an essential element for current research. It is important indirect evidence for the reconstruction of the Antiochene text, and it provides a very solid foundation for accessing pre-Hexaplaric stages of the Septuagint—especially the proto-Lucianic—that are very close to the OG text. This is the case for Samuel, Kings, and Chronicles. Most recently, the support of the Vetus Latina for Lucianic/Antiochene readings has allowed scholars to extend its important role to other books, such as Judges (Fernández Marcos 2011: 6*–9*). The agreement of the Lucianic text with Vetus Latina can provide the pathway to old stages of Septuagint textual history. However, it is important to recognize that the transmission and textual history of the Vetus Latina is very unstable and not fully understood. Its use in Septuagint textual criticism requires, therefore, a previous study of the relevant Vetus Latina textual history that, for some books of the Old Testament, is available in the series Vetus Latina (1954–). In recent years, the Armenian version has also acquired extraordinary importance in textual criticism of the Septuagint (Cox 1985). It constitutes important evidence for the reconstruction of Origen’s Hexapla, especially of the Aristarchan symbols (obeli and asterisks), since it is a decisive witness together with the Greek manuscripts and the SyroHexapla (Cox 1993). Its evidence is a minor contribution to the reconstruction of the oldest Septuagint text, but it can also be used as a control to confirm intermediate stages of Septuagint transmission, such as the Lucianic and Hexaplaric recensions. The Ethiopic version had a very uncertain role in Septuagint textual criticism until the mid-twentieth century. For many years there was little agreement upon the identification of its Vorlage. But current research accepts that the Vorlage of known Ethiopian texts was in fact a Greek text that in the fourteenth and fifteenth centuries was revised against an Arabic version. Until the end of the twentieth century there were no high-quality critical editions, and the available manuscripts were essentially those preserved in European libraries, most of them very late. However, the Ethiopian Manuscript Microfilm Library project (EMML) has greatly changed the landscape of evidence, bringing to light manuscripts preserved in Ethiopia, in some cases older and of better textual quality than those previously known. This development has positioned the Ethiopian version within Septuagint textual criticism as a valuable witness, not for reconstructing the OG text, but rather as a criterion for confirming the data provided by Greek manuscripts relevant to the place of the recensions in textual history. There are no uniform criteria for the role of the Arabic versions in the published volumes of the Editio Maior of Göttingen. These versions have not been taken into account in Esther (Hanhart 1983a) or 2 Chronicles (Hanhart 2014), but they have been collated for Daniel (Ziegler 1999) and Jeremiah (Ziegler 1957). In other volumes, some editors like Rahlfs and Hanhart, following Lagarde, held that the Arabic versions had no critical value for the Septuagint. However, since the work of Wevers on the Greek Pentateuch, the Arabic versions have obtained a greater role. But the evidence they offer requires very careful preliminary study to ensure it was a direct translation from Greek, a judgment that is rarely easy to make.

134

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

RECOMMENDED RESOURCES 1. Rahlfs (1914), Rahlfs and Fraenkel (2004). These resources are catalogues providing detailed information on all known Greek manuscripts of the Septuagint, with descriptions and, when possible, characterization of their texts. The numbering used for the designation of manuscripts is that used in the Editio Maior of the Göttingen Septuagint. 2. Fernández Marcos (2000). An overview of the Septuagint that provides a detailed description of the main problems in current research, especially in the field of recensions and textual transmission. 3. Wevers (2003). An approach to Septuagint textual criticism by one of the editors of the Editio Maior of the Septuaginta-Unternehmen of Göttingen. It explains in detail the process of textual study prior to the establishment of a critical edition. 4. Aitken (2014b). For linguistic factors in Septuagint textual criticism, see especially chapters 1 and 2, respectively, “The Current Study of Septuagint Vocabulary” and “Documentary Evidence in Biblical Lexicography.” 5. Kreuzer (2016)/Kreuzer (2019). The most up-to-date overview of the Septuagint, its research problems, and the status quaestionis book-to-book, available in both English (2019) and German (2016). In particular, the chapters “Entstehung und Überlieferung der Septuagint” (“The Origins and Transmission of the Septuagint”) and “Überblick zu den Textzeugen der Septuaginta” (“Overview of Textual Witnesses to the Septuagint”) provide detailed panoramas of Septuagint textual history and issues in current research. The remaining chapters offer overviews of current research and future studies for each book of the Septuagint. 6. Schäfer (2016). Especially the following sections: Teil B,II: “Rahlfs als ‚Vorarbeiter‘ der Septuaginta-Edition” (1898–1921); Teil B, 3.3. “Untersuchungen zu den Kirchenschrifstellerzitaten und Tochter-übersetzungen”; Teil C, I: “Rahlfs als ‚Textkritiker‘ der Septuaginta”; Teil C, II: “Rahlfs als ‚Editionstechniker‘ der Septuaginta.” These essays provide an overview of the genesis and development of modern textual criticism of the Septuagint, specifically as applied in the Editio Maior of Göttingen.

CHAPTER 9

The Septuagint and Textual Criticism of the Hebrew Bible JOHN SCRENOCK

INTRODUCTION AND HISTORY OF SCHOLARSHIP The Septuagint’s usefulness for Hebrew Bible (HB) textual criticism has been recognized— and debated—for a long time (Fernández Marcos 1987; Tov 2015a: 37–9). The basic principles of the modern approach were already articulated in the nineteenth century by scholars like S. R. Driver and Julius Wellhausen (Wellhausen 1871: 1–33; Driver 1890: xxxvi–xl; cf. Fernández Marcos 1987: 60–3). They argued that we should first undertake textual criticism of the Old Greek (OG)1 itself to make sure we are dealing with the original translation, then we should eliminate differences that stem from the translator, at which point, the remaining evidence can be used in HB textual criticism. Before the discovery of the Dead Sea Scrolls, the use of the OG to reconstruct Hebrew variants was difficult to verify. When the manuscript evidence provided by the scrolls came to light, however, many reconstructions of alternative Hebrew texts on the basis of the Greek version were vindicated (Fernández Marcos 1987: 66–8; Tov 1999c). In 1 Sam. 1:23 in the Masoretic Text (MT), for example, Elkanah states ‫“( יקם יהוה את דברו‬may the LORD establish his word”). In the mid-nineteenth century, O. Thenius used the OG’s τὸ ἐξελθὸν ἐκ τοῦ στόματός σου (“what goes out of your mouth”) to reconstruct the variant ‫את היצא‬ ‫“( מפיך‬what goes out of your mouth”) for MT ‫( את דברו‬Tov 1999c: 289). Over a century later, we discovered that 4QSamuela reads ‫היוצא מפיך‬, which is remarkably close to Thenius’s retroversion. In cases of major variation between the MT and OG, our estimation of the value of the OG as a reliable text-critical witness also increased after the scrolls. A clear example is with the case of Jeremiah, where the OG text is significantly shorter than the MT (Ulrich 1999: 91). Before the discovery of the scrolls, we could not be sure whether the OG took liberties in its translation or whether it closely followed a variant Vorlage. But

I use the term “Old Greek” (OG) to refer generally to the first Greek translations of books of the HB and their manuscript tradition, excluding recensions like Aquila and Theodotion. I do not use “Septuagint” because I am not referring to a particular project of translating the Pentateuch. I do not mean “Old Greek” to refer exclusively to the (reconstructed) original translation of each book, unless in context the distinction is made.

1

136

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

4QJeremiahb and 4QJeremiahd proved that the shorter OG version was in fact based on a Hebrew Vorlage that circulated during the Second Temple period. Of course, Dead Sea Scrolls manuscripts are not extant to confirm every single divergence from the MT in the OG. What the scrolls have shown us, however, is that the OG is generally a reliable witness. In similar cases where we lack Hebrew evidence, we must take the OG seriously. The OG account of the construction of the tabernacle in Exodus 35–40, for example, is shorter than the MT—parallel to the example of Jeremiah, except without corroborating evidence from a Dead Sea Scroll.2

RESEARCH METHODS For the purpose of using the OG for HB textual criticism, it is helpful to think about OG data in terms of competing explanations.3 The relationship between the words of a particular OG witness and a particular Hebrew witness can always be explained in a variety of ways. Since Greek is not Hebrew, the target text will always differ from the source text; this is a fact of translation. As such, strictly linguistic aspects of transfer from Hebrew into Greek make up the majority of differences between a Greek reading and a Hebrew reading. However, even when basic aspects of translation are taken into account, there often appear to be further differences that require explanation. When the Greek text is compared to the Hebrew, then, we want to know how much of the difference stems from translation into Greek, and whether any elements of difference are better explained by a Hebrew source text that differed from the Hebrew witness at hand (often MT), or by textual development within the transmission of the OG itself. Given the complexity of the overall translation process, there are many ways we can explain elements of an OG reading that seem to go beyond the typical mechanics of translation for any given OG book. The following typology represents the range of potential explanations for the relationship between a Greek and Hebrew text: 1. The Hebrew Vorlage of the Greek witness differs from the Hebrew text. 2. The Greek is a representation of the Hebrew text in a different linguistic and cultural medium. This involves three distinct but connected processes: 2a. The translator decodes the Hebrew source. 2b. The translator encodes elements of the Hebrew source into Greek. 2c. The translator may change the content of the source. 3. The Greek reading derives solely because of inner-OG textual development and is not reflective of any Hebrew Vorlage. This typology is meant to be broad and systematic, marking distinct categories for  phenomena. Some, if not all, of the categories, however, may be involved in

Most scholars take the OG version of the text to be earlier (Aejmelaeus 2007a: 107–21; Tov 2012: 316–17; Ulrich 2012: 217). Others, however, continue to doubt the reliability of the OG (Gurtner 2013: 26). 3 For an earlier statement of methodology based on the notion of competing explanations, see Screnock 2017: 31–8. Readers will note that elsewhere in that book, and in a recent publication (Screnock 2018), I propose a slightly different approach. In the present chapter, I offer a synthesis of methodology more in line with consensus views, noting some of my points of departure but generally leaving them in the other publications, which readers can consult. 2

TEXTUAL CRITICISM OF THE HEBREW BIBLE 137

particular cases. The point is that the full range of possibilities should be considered when analyzing the OG. We do not have access to an original Greek translation and the particular Hebrew text that it translated. Greek manuscripts vary from one another, as do manuscripts of the HB. The typology above is meant to be flexible in this regard, not referring to the OG translation and the Hebrew source text, but to the words found in a particular OG witness compared to those found in a particular HB witness—whether we are using a single manuscript (e.g., the Leningrad Codex or Codex Sinaiticus) or a critical reconstruction (e.g., an HBCE or Göttingen edition). This is particularly important on the Hebrew side of things, since the point of this methodology is to recover potential variant Hebrew readings. As such, in the following discussion I refer to ‘Hebrew witnesses” instead of “the Hebrew source text.” One final distinction is important: in our endeavor to use the OG for HB textual criticism, we must be concerned with the OG at the point of its production rather than its reception, since this is the point at which the Hebrew Vorlage of the OG was involved.4 Before considering each category in more detail, I will use a single example to illustrate how each aspect may be involved in an individual case. In Ps. 104:10 (103:10 OG), most manuscripts of the OG read ἀνὰ μέσον τῶν ὀρέων διελεύσονται ὕδατα (“waters will pass between the mountains”), while all Hebrew manuscripts have ‫“( בין הרים יהלכון‬they go about between the mountains”).5 There are two distinct issues that could be explored here: (1) What Hebrew verb is represented by διελεύσονται; and (2) Why does the OG have the word “waters” where the Hebrew witnesses do not? Leaving aside the former question and focusing on the latter, we could explain the difference along the lines of any of the categories noted above.6 The first possibility (category #1) is that the Hebrew Vorlage had the plus ‫“( םימ‬waters”), a reading that differs from our extant manuscripts. The second possibility (category #2a) is that the Vorlage may not have included the plus, but the translator may have read and understood the text as including ‫( מים‬i.e., at the moment of reading the text, without any conscious effort to change the text for any reason). This unconscious change could have happened for several reasons: He may have had the phrase ‫“( על הרים יעמדו מים‬the waters stand upon the mountains”) still in his mind from v. 6, he may have expected another “water” noun as in the preceding clause, where ‫“( מעינים‬springs”) is the subject, and/or he may have accidentally read the last one or two words by looking at v. 6 (an ocular skip). The third possibility (category #2b) is that the translator may have added ὕδατα (“waters”) at the point of transferring the meaning into Greek. In this scenario, he would have read the phrase as ‫( ּבֵ ין הָ ִרים יְ הַ ּלֵכּון‬as in the MT), understanding from context that the

On this distinction, see Pietersma 2006d: 50–2. If we want to use the OG for textual criticism of the HB, we must aim at the Hebrew behind the OG at the point when it was translated. As such, the approach of scholars like Alma Brodersen (2017), who uses the OG as an ancient witness but often treats it as a standalone text in Greek, is inadequate. 5 There is some slight variation: 4QPsd has ‫ ההרים‬with the article and the verb ‫ יהלכו‬without paragogic nun, but both are inconsequential to the meaning of the phrase; in all other respects the reading is identical to our other Hebrew witness, the MT. 6 I discuss the former question in Screnock 2018: 251–2. 4

138

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

subject must be some kind of water, and thus at the point of rendering the phrase in Greek he added an explicit subject, ὕδατα. This would be an instance of less isomorphic, more target-oriented translation technique—though the one-to-one representation of the preceding three words would be in keeping with the general isomorphism of OG Psalms. The fourth possibility (category #2c) is that the translator may have thought, “I need to add something to the text here.” Perhaps he wanted to clarify that it was water, not some other element of creation, going between the mountains, or perhaps he saw an opportunity for heightened poetic parallelism, tying this line more clearly to ‫ על הרים יעמדו מים‬in v. 6. Whatever the reason, in this scenario the impetus for change would not be his natural reading of the text at the very moment of reading, nor the particular translation technique used to render Hebrew using Greek, but a more conscious and explicit desire to change the text in Greek.7 The fifth and final possibility (category #3) is that the original translation may not have included ὕδατα at all. Although most manuscripts of the OG do have “waters,” some have a minus as in the Hebrew. If the minus were the text of the original translation, we could explain the plus as resulting from transmission of the OG. In this scenario, a scribe might have skipped up to στήσονται (“they will stand”) in v. 6, or even just the string of letters (in scripta continua) ΣΟΝΤΑΙ—they are identical to the end of διελεύσονται in v. 10—and continued copying ὕδατα as in v. 6. In considering the presence of ὕδατα in the OG, then, it is necessary to consider a variety of possible explanations, each of which are able to account for the difference, but which have varying levels of explanatory power. In the following paragraphs, I further explore each point of the typology.

1) The Greek and Hebrew Differ Because of the Vorlage of the OG A word, phrase, or passage in the OG may differ from a Hebrew manuscript because the Hebrew translated by the OG was different. In such cases, the difference exists in Hebrew, before and distinct from the process of translation. In Exod. 1:1, for example, OG Ιακωβ τῷ πατρὶ αὐτῶν (“Jacob, their father”) differs from MT ‫“( יעקב‬Jacob”) because it translates different Hebrew, ‫“( יעקב אביהם‬Jacob, their father”; Screnock 2017: 124). When a variant Vorlage is responsible for some level of difference between the OG text and a Hebrew manuscript, the OG is a reliable witness to a variant Hebrew text and we can therefore use it for HB textual criticism. As such, determining whether differences belong to this category is the ultimate goal of the entire process described in this chapter. These are the cases we want to isolate. Some argue we should assume that the MT is the Vorlage of the OG unless proven otherwise,8 while others maintain that we should be less reticent to postulate a variant Vorlage (Barr 2002; Aejmelaeus 2007a: 81–5, though cf. Barr 1968: 245; Metso and Ulrich 2003: 253; Peters 2010: 121–34). While the term Vorlage is often used to refer to

Perhaps we might think of this sort of change coming about for theological or literary reasons that are only understood by the translator somewhere below the surface of conscious and explicit thought, in which case this point (#2c) would bleed into translation technique (#2b). If, for example, the community to which he belonged had a strong emphasis on the theological importance of water (an implausible scenario but helpful to illustrate the point), and he therefore added “waters” as an automatic, community-conditioned reflex. 8 Wevers, for example, often followed this principle (1997: 438). Two others who have articulated similar principles, but do not follow or prioritize it to the same degree, are Tov (2015a: 44) and Aejmelaeus (2007a: 79), though note her position on deliberate content changes (2007a: 81–5). 7

TEXTUAL CRITICISM OF THE HEBREW BIBLE 139

a physical, written Hebrew manuscript that the OG translator looked at when translating, the German term Vorlage can refer more broadly to the concept of a template or model. Indeed, the specific source text used by an OG translator may not always have been a physical, written text; it may instead have been the recitation aloud of such a text (Van der Louw 2008b), and we could even imagine a translator translating from memory (Screnock 2017: 81–2). Screnock (2017: 72–92) has suggested that the true source text for translation is a mental version of the text existing in the translator’s mind when he reads a written text, hears a recited text, or recalls a memorized text. The most important aspect for considering whether a variant Vorlage is responsible for a difference is the retroversion of Greek into Hebrew (cf. Tov 2015a: 62–111; Screnock 2018: 248–52). Retroversions should be based on a rigorous understanding of the general translation technique of the OG book and its approach to the particular issues in view (see “2b Encoding” below). If an OG book consistently represents aspects of the Hebrew isomorphically, we can use established patterns to determine the equivalents that likely stand behind a particular Greek reading. Though some scholars advocate the use of scholarly intuition (Tov 2015a: 78–80), retroversion without corroborating evidence should be avoided (Aejmelaeus 2007a: 74; Screnock 2018: 250–3). Retroversions should keep statistical significance and probability in mind. It is not enough for a given equivalence to occur elsewhere in the OG. Rather, an equivalence should be the most probable option for a retroversion, based on a statistically significant amount of data (Screnock 2018: 250–2). All aspects of language—especially morphology, morpho-syntax, syntax, and vocabulary in isomorphic translations9—should be considered with respect to potential equivalences set up by the OG (Margolis 1910: 304, 311–12; Screnock 2018: 248–9; cf. Tov 2015a: 65–6 on vocabulary equivalences). In Exod. 1:10, for example, the OG has καὶ ἐκπολεμήσαντες ἡμᾶς ἐξελεύσονται (“and after making war with us, they [= Israel] will go out”) where all Hebrew witnesses have ‫“( ונלחם בנו ועלה‬and he [= Israel] will fight with us and go up”). A few differences are apparent: the first verb in Greek is a participle, whereas the Hebrew texts have a finite verb; this verb takes an accusative noun complement in the Greek while in the Hebrew it takes a bêt prepositional phrase; there is no conjunction before the second verb in Greek, where Hebrew has wāw; and the lexemes ἐξέρχομαι (“to go out”) and ‫“( עלה‬to go up”) have different semantic ranges. How does OG Exodus typically deal with these aspects of language—the form of the verb, the syntax of verbal complementation, the use of wāw in sequential actions, and the lexeme ‫ ?עלה‬Answers to these questions are crucial for accurately retroverting the Hebrew behind καὶ ἐκπολεμήσαντες ἡμᾶς ἐξελεύσονται. Taking the first three issues in order: OG Exodus regularly uses a participle to translate a Hebrew finite verb within a sequence of actions (Aejmelaeus 2007a: 5), regularly follows the syntax of Greek for verbal complementation instead of woodenly rendering Hebrew syntax (Screnock 2017: 111–12), and drops the conjunction in protasis-apodosis and temporally situated clauses (Aejmelaeus 2007a: 91; Screnock 2017: 114). Clearly, on these three points it makes sense to retrovert the Vorlage of the OG as identical to the Hebrew witnesses. It would be reckless to retrovert something like ‫—וְ נִ לְ חָ ם אֹ תָ נּו ָעלָה‬a participle with direct object and no conjunction before the second verb—both because of While these are the most commonly represented aspects of language in isomorphic OG translations, we should also be attentive to higher-level features like information structure, literary structure, and poetics, which may also be intentionally represented in the OG translation.

9

140

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the clear translation patterns just noted, and because it is ungrammatical Hebrew (on the latter point, see Tov 2015a: 84–6). For the fourth issue, however, when we investigate the manner in which OG Exodus represents ‫עלה‬, we find that ἐξέρχομαι usually represents ‫יצא‬, not ‫( עלה‬Screnock 2017: 121). Because this is a demonstrable translation pattern, we should reconstruct the verb as ‫יצא‬, a variant with the extant Hebrew witnesses. For vocabulary equivalences, Hatch and Redpath’s concordance (1998) is a valuable aid, and a backward concordance based on Hatch-Redpath should be consulted for research that begins with Hebrew (dos Santos 1973; Muraoka 1998). For vocabulary and grammar equivalences, computer software can also be helpful if used carefully, as well as the CATSS database (Tov and Polak 2009). None of these aids, however, can substitute for in-depth studies on individual issues (cf. the bibliography in Tov 2015a: 75–8). With vocabulary, for example, different contexts may impact translation choices, and with syntax, general translation trends may have exceptions for specific syntactic constructions. We should seek to be as detailed as possible, not assuming that the OG always follows a general, contextually insensitive method (Maurais 2020). It is appropriate to use a single Hebrew witness like MT or SP to establish translation trends. Even though some of the cases included may involve a variant Vorlage, a single Hebrew witness adequately approximates the Vorlage of the OG when a large set of data (i.e., statistically significant data) is involved. Retroversion based on equivalences elsewhere in the OG works well for isomorphic OG texts. With highly target-oriented translations like OG Job, however, retroversion is harder (Tov 2015a: 66, 96), if not impossible, because the target text does not give clear access to the source text. If we are unable to retrovert the OG into Hebrew, we cannot use it for HB textual criticism (Aejmelaeus 2007a: 78). When considering whether a variant Vorlage is responsible for a difference in the OG, one simultaneously considers also whether translation technique is responsible (i.e., category #2b, “Encoding”). Because proper retroversion relies on analysis of translation technique, consideration of the two go hand-in-hand (Tov 2015a: 48).

2a) The Greek and Hebrew Differ Because of How the OG Translator Decoded the Text Before the translator could translate the words of the OG Vorlage, he had to decode them, making some sense of what the Hebrew meant, even if only at the most basic level (Pietersma 2006a: 34–5; Lemmelijn 2009: 98–9; Englund Dimitrova 2010: 406; Stolze 2010: 141; Tov 2012: 118; 2015a: 49–50). Depending on the translator’s proficiency with Hebrew, decoding could involve simply vocalizing the Hebrew text—in cases like Ezra 8:27, where OG καφουρη simply transliterates the Hebrew, ‫“( כְ פֹ ֵרי‬bowls of”) (Tov 1999a: 204). Or the translator may have decoded the text by identifying lexical items without vocalizing them and then selected a Greek equivalent from a set of possible equivalents for words from the same root (Barr 1967b). In most cases, however, it is likely that the translator did vocalize the text (against Barr’s theory, see Screnock 2017: 82–4) and went beyond just vocalizing. The way the translator typically decoded the Vorlage was by reading it in the same manner that you are currently reading (Screnock 2017: 84, 2018: 232).10 Other strategies may have been employed (see below), but the core of decoding was reading. Note that, concerning translation in general, “while the translator’s reading of a text may be to some extent more thorough and deliberate than that of an ordinary reader, it is not likely to be markedly so” (Shreve et al. 1993: 21).

10

TEXTUAL CRITICISM OF THE HEBREW BIBLE 141

The OG may appear to differ from the Hebrew because of the way it decoded its Vorlage. First, the translator may have vocalized the consonants differently than the way scholars do today (Gentry 2006: 202–3), and/or differently than the text is vocalized in the MT. In Exod. 1:15, for example, OG ταῖς μαίαις τῶν Εβραίων (“to the midwives of the Hebrews”) understands the vocalized Hebrew text ‫“( ִלְמ יַ ְּל ֹדת ָה ִע ְב ִר ֹּית‬to the midwives of the Hebrews”) instead of the vocalization of the same consonants found in MT, ‫“( ַל ְמ יַ ְּל ֹדת ָה ִע ְב ִר ֹּית‬to the Hebrew midwives”).11 In addition to vocalization, the translator’s basic interpretation of the Hebrew text could potentially differ from the basic interpretation that scholars think is correct. Also included at the stage of decoding is the phenomenon of translator scribal error, whether stemming from visual, aural, and/or memory factors (Jobes and Silva 2000: 152–3; Gentry 2006: 203–6; Aejmelaeus 2007a: 80; Tov 2015a: 144–56). In 2 Sam. 5:20, for example, OG ἐκ τῶν ἐπάνω διακοπῶν (“from the upper breaches”) possibly stems from a misreading of ‫“( בבעל פרצים‬in Baal-perazim”) as ‫ממעל פרצים‬, mistaking the initial two bets as mems (Tov 2015a: 149). Differences in vocalization and basic interpretation, as well as scribal error committed by the translator, all occur at the point of the translator decoding the text, before translation into Greek. If we fail to understand how the OG decoded the Hebrew, we are tempted to falsely attribute other activities to the translator. For example, the Hebrew text of Sirach 51 has the verb ‫“( תעיתי‬I erred”) in v. 13—or at least this is how the manuscript 11QPsa has been transcribed and understood by scholars (Sanders 1965: 80). OG πλανηθῆναί με is not an idiomatic rendering that uses an infinitive construction instead of a finite verb. Rather, it reads the Hebrew text as ‫“( תעותי‬my erring”), isomorphically rendering the Hebrew infinitive followed by a first-person pronoun with the same grammar in Greek (Skehan and Di Lella 1987: 574). Yod and wāw were often confused, and in this column of 11QPsa the two are virtually indistinguishable. The OG text, then, attests to the same graphic representation of the word in 11QPsa. Scholars and the OG simply read the word differently. We cannot discount the Greek translation of the Hebrew on the basis of our reading of the text. The translator may very well read the text in a less-than-preferable way, but this does not entail that the translation based on that reading is revisional or misguided.12 A fascinating aspect of the OG is the occasional interference cause by Aramaic and the native Hebrew spoken by translators (Goshen-Gottstein 1963: 140; Joosten 2003 Loiseau 2016; Byun 2017). Translators may have unintentionally read the Hebrew Vorlage through the lens of one of these languages and, at times, they purposefully used these languages as a strategy for making sense of the Vorlage. In Ps. 60:10 (59:10 OG), for example, OG τῆς ἐλπίδος μου (“my hope”) for Hebrew ‫“( ַרחְ צִ י‬my washbasin”) reflects the meaning of the Aramaic root ‫רחץ‬, “to trust,” instead of the meaning of the Hebrew (Gentry 2006: 200–1). Though the Greek translation is, of course, affected, the phenomenon occurs with the Hebrew text before it is rendered as Greek. A related issue is the translator’s understanding of the text (or lack thereof ), which includes both his competence in Hebrew and his ability to use contextual clues to interpret

Some scholars (Childs 1974: 6; Wevers 1990: 7), assuming the vocalization in MT is the vocalization, misunderstand the OG at this point. 12 Sanders (1965: 83), for example, states, “There can be little doubt that [the OG] presents an interpretive recension of [Sirach 51].” To my mind, however, the OG presents an isomorphic and source-oriented translation that is not “interpretive” (cf. Screnock 2017: 34). For discussion of further examples, see Screnock (2017: 34–5) and Pietersma (2006d: 71). 11

142

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

a text. To a greater or lesser extent depending on the OG book, the translator may have found the text difficult to understand (Barr 1968: 249, 253–4, 266–7; Tov 1999a; 2015a: 178–83; Gelston 2006: 47–58; Glenny 2007: 534–43; Aejmelaeus 2007a: 80; Lemmelijn 2009: 99–103). When this occurred, he could use a variety of strategies to overcome the difficulty, including transliterating the Hebrew, relying on context to smooth over the difficulty, or drawing on Aramaic or contemporary Hebrew for potential meaning.13 Though the use of Aramaic or spoken Hebrew as a strategy for reading a difficult text is not part of the normal reading process, it nevertheless occurs on the level of the Hebrew text, not on the level of transfer into Greek. It is one way to make sense of and decode the Hebrew Vorlage. Similarly, translators sometimes drew on etymology to find meaning for the text—whether consciously as a strategy for decoding, or unconsciously as part of the reading process (Barr 1968: 253–5; Tov 1999a: 216–18; Tov 2015a: 188–97). For example, for 2 Chr. 3:10 ‫“( מעשה צעצעים‬a work of casting”) the OG translates ἔργον ἐκ ξύλων (“a work from wood”), deriving the meaning of ‫צעצעים‬, it seems, from ‫“( עץ‬wood”) (Tov 2015a: 190). It should be noted that, although the presence of etymological exegesis and Aramaic interference is often taken as a sign of the translator not understanding the text, in reality the translator may have been engaging in typical interpretive practices of the ancient world.14 In any of the situations just described, the point of understanding the translator’s decoding of the text is to help us retrieve the Vorlage (Tov 2015a: 189). If the translator read the text as Aramaic, or vocalized the text differently, or changed the meaning of the word based on perceived etymology, this does not entail a true variant, nor does it exclude one. The goal is to understand the consonantal text used by the OG translator or, in some cases, the tradition of vocalization followed by the OG translator, if we believe he was familiar with established ways of reading the text (Tov 2015a: 118).15 Though all of the phenomena discussed above occur during the activity of translating, none of them stem from the actual moment of transfer from one language to another. All of these phenomena occur in reading of the Hebrew source text. Emanuel Tov has discussed the phenomenon of Hebrew variation that exists only in the mind of the translator (e.g., 2015a: 98–9, 134–5, 178; cf. Screnock 2018: 235–6). Screnock (2018) argues that this notion applies to a large spectrum of OG data, particularly data that falls into the category of decoding. Long pluses in the OG, if they do not reflect a plus in the Vorlage of the OG, can be conceptualized as small-scale pseudo-translation—composition in Greek that feels, whether intentionally or not, like it represents a Hebrew source text.16

2b) The Greek and Hebrew Differ Because of How the OG Is Encoded Because Greek is not Hebrew, translation entails difference. As a result, most of the differences between the OG and a Hebrew manuscript stem from encoding, that is,

For examples of how an OG translator may have overcome various difficulties in the Vorlage, see Glenny (2007). 14 Arie van der Kooij (2017) argues this point well. Cf. Tov’s comparison of translator “maneuvering” with midrashic interpretive practices (2015a: 179–80). 15 In some books at least, the tradition of reading the text preserved in the OG reflects ancient ways of understanding the text, like the reading tradition behind the vocalization of the MT. 16 See Screnock (2017: 79) for an example of small-scale pseudo-translation in n. 107. On pseudo-translation, see Toury (1995: 40–52) and Davila (2005). 13

TEXTUAL CRITICISM OF THE HEBREW BIBLE 143

from “translation technique.” Many have rightly emphasized, therefore, the need to understand the translation technique of an OG book before considering its text-critical value (Margolis 1910: 302–3; Barr 1968: 250–9; Wevers 1985: 20–4; Aejmelaeus 2007a: 72–4, 76–7; Lemmelijn 2009: 19, 23–4, 97, 103–4, 130–1; Tov 2015a: 18–32, 48–9, 64, 66, 96 Maurais 2020). In recent decades, Septuagintalists in North America have turned their focus by and large to the study of OG translation technique, making significant advances by drawing on work in translation studies (e.g., Wright 2003; Boyd-Taylor 2006a; Pietersma 2006d). After decoding the Hebrew text, the translator transfers the text to a Greek linguistic context, encoding it in Greek. While it is tempting to say that the “meaning” of the Hebrew is what the translator encodes in Greek, in translation there is no single element of the source text that must be transferred. Rather, the translator chooses—whether consciously or unconsciously, often by habit17—what aspects of the source text to represent. These might include the meaning of the text, word order, lexical semantics, distinct morphological items, the forms of verbs, and literary features. Moreover, these levels of representation can be aimed at different ranges or sizes of text, from the smallest (e.g., the gender of a single morphological item) to the largest (e.g., always using a single Greek lexeme to represent a given Hebrew lexeme across the entire book). So the translator can represent all sorts of linguistic elements at every level of language. The decisions made by the translator about what elements from the Hebrew to represent and how to represent them are the substance of “translation technique.” The countless differences between the linguistic systems of Hebrew and Greek (Wevers 1985: 15–19; Wevers 1990: vii; Aejmelaeus 2007a: 76) created a barrier between the source text and its Greek speaking/reading target audience. The notion of translation technique refers to the translator’s strategy for surmounting that barrier. This definition of translation technique is very strict, dealing only with the linguistic aspects of the text. Though there are other aspects of the translation process (the other categories of this typology), this is the point at which the actual linguistic transfer from Hebrew to Greek occurs. It is important to note that, though they are discrete processes that are helpful to consider separately, decoding and encoding happen almost simultaneously in real time, and they mutually inform one another. Because it is possible for a translation to represent any elements from the whole range of linguistic and textual phenomena in the source text—but not all at once—the common descriptions of translation using terms like “faithful,” “literal,” and “consistent”—and their negative counterparts—are misleading. A translator who slavishly represents TenseAspect-Mood features in the verbal system could not at the same time consistently represent distinctions between verbal conjugations, and vice versa. Or to cite another example, there is often tension between verbal semantics and argument structure. Greek verbs often take a different argument structure than the Hebrew verbs they represent, forcing the translator to use the grammatical Greek construction instead of exact representation of the syntax of the Hebrew. In Exod. 1:10, for example, Greek κατασοφισώμεθα αὐτούς (“let us outwit them [= Israel]”) takes an accusative complement, rather than a prepositional phrase as in the Hebrew ‫“( נתחכמה לו‬let us deal shrewdly with him [= Israel]”) (for more examples, see Screnock 2017: 111–12).

Whether or not the process was deliberate, translators’ “translational behavior [followed] norms,” and is thus “subject to description after the fact” (Pietersma 2010: 7; cf. Screnock 2017: 30–1).

17

144

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Although there is no single way to “faithfully” translate a text, there was a typical practice of isomorphism taken in many OG books (Pietersma and Wright 2009: xiv–xv; Aejmelaeus 2007a: 295, 298). Specifically, isomorphic OG translations try to represent each morphological item in Hebrew with a corresponding item in Greek (Wright 1987; Boyd-Taylor 2006a: 25–7; Pietersma 2006d: 63–9; Pietersma 2007: 542–4), and to set up lexical equivalents that are maintained across entire texts (Wright 1987; Lust 1993; Pietersma 2007: 544; Boyd-Taylor 2010b; on various types of lexical equivalency, see Pietersma and Wright 2009: xvii–xviii). Thus, Ps. 104:2 (103:2 OG) ‫עֹ טֶ ה אֹור ּכַּׂשַ לְ מָ ה‬ (“wrapping [himself with] light like a garment”) is translated ἀναβαλλόμενος φῶς ὡς ἱμάτιον (“putting-on light like a garment”). There is one Greek item for each Hebrew item, of the same type and in the same order: participle, noun, preposition, noun. Furthermore, the nineteen occurrences of ‫“( אֹור‬light”) in the Psalms, including this one, are glossed each time by φῶς or φωτισμός (both mean “light”).18 In books like OG Exodus, Leviticus, and Psalms, this kind of isomorphism is pervasive. Instances of “free” translation technique certainly occur, even systematically, but these are departures from the overall norm.19 When these aspects of translation are taken into account, they often provide a more compelling explanation for differences between a Hebrew witness and the OG. For example, in a passage of Sirach considered by some to have sexual overtones, in the Hebrew ‫“( קנאתי בטוב‬I was zealous for pleasure”) the noun ‫“( טֹוב‬pleasure”) is translated τό ἀγαθόν (“the good”) (Sanders 1965: 83). Though the result may be the removal of sexual connotation, clearly the reason the OG translator chose τό ἀγαθόν is because ἀγαθός is the standard equivalent for ‫טוב‬. It would be unwise to attribute an ideological or theological motive to the translator for rendering the Hebrew in this way. The generally isomorphic character of the OG should, then, be considered when evaluating the difference between a Hebrew witness and the OG. This is not, however, completely sufficient for understanding the particular translation technique of individual OG books. As such, scholars must also study and pay attention to the translation patterns of the OG text in question (Margolis 1910: 306; Tov 2015a: 16–17).

2c) The Greek and Hebrew Differ because OG Makes a Content Change There were times when the OG translator produced Greek that altered the content of the source text in some way, beyond what was typically involved in the process of encoding (Jobes and Silva 2000: 153; Gentry 2006: 201–2; Glenny 2009; Lemmelijn 2009: 99; Tov 2012: 119–22; 2015a: 50–5). Such changes can range from small to extensive, and there is a significant amount of grey area between small content change and encoding: the translator may have added or removed words and phrases to help his audience understand the text, particularly if he was using a more target-oriented approach to translation.20 At other times, however, significant alterations involving addition, omission, and semantic alterations are rightly considered to go beyond the boundaries typically assigned to This is, of course, an oversimplification. Isomorphism involves other types of representation—for example, of high-level word order—and does not exclude the non-use of isomorphism, for example when verbal argument structure demands a departure (as noted above). Lexical isomorphism, in particular, is rarely as simple as a oneto-one relationship: one Greek word is often used regularly for two or even three Hebrew words, while one Hebrew word is often glossed by two or three Greek words. 19 See the discussion of the isomorphic translation profile of OG Exodus, despite its frequent use of free translation, in Screnock (2017: 43–4). 20 As such, the types of change I include in this category are often included in the concept of “translation technique.” As noted above, I prefer a stricter definition of “translation technique.” 18

TEXTUAL CRITICISM OF THE HEBREW BIBLE 145

translation. The changes can stem from a variety of factors, whether interpretive, exegetical, contextual, logical, or cultural. In this category, we include changes that result from a particular theological perspective, which the translator inserts into the translation though it is not present in the Vorlage. Though often appealed to by some scholars, such scenarios involving theology are in fact rare in most OG books (Pietersma 2006a; cf. Pietersma 2006d: 50–2; Douglas 2012). It is important to distinguish between the reasons for the translator changing the text and the changes themselves. Theological interpretation, for example, is one possible impetus for the translator to change the content of the text. However, it is also a potential impetus for the translator to read (decode) the text in a certain way. As such, theological interpretation is a useful aspect of translation to be aware of, but it does not fit into this typology as a distinct category of its own. By “content change,” then, I mean changes to the text—not reasons for change—that are made deliberately by the translator, not introduced via his natural reading of the Vorlage. It also important to note that, in real time, the insertion of content changes happens alongside the processes of decoding and encoding, though it is a conceptually discrete aspect of the translation process. It is reasonable to suppose that content changes can occur either during the process of encoding or during the process of decoding. As such, the translator may have changed the text either in Hebrew or in Greek. Consider an example: Gen. 9:22

‫וירא חם … את ערות אביו ויגד לשני אחיו‬ And Ham saw … the nakedness of his father, and he made it known to his two brothers. καὶ εἶδεν Χαμ … τὴν γύμνωσιν τοῦ πατρὸς αὐτοῦ καὶ ἐξελθὼν ἀνήγγειλεν τοῖς δυσὶν ἀδελφοῖς αὐτοῦ And Cham … saw the nakedness of his father, and after he had gone out he told his two brothers. (NETS)

The underlined word in the OG may have been added in Greek, “to improve [the text’s] readability from a linguistic and contextual point of view” (Tov 2015a: 50–1). Alternatively, the translator may have added the word in Hebrew (‫ )ויצא‬and then translated it into Greek (Screnock 2018: 242–3). When seeking an explanation for the relationship between the OG and a Hebrew witness, the explanations in this category carry the burden of proof. One should first account for translation technique, decoding, and a variant Vorlage before appealing to intentional content changes made by the translator (Barr 2002; Dines 2004: 132–3; Pietersma 2006a; Aejmelaeus 2007a: 81–5; Boyd-Taylor 2010b).

3) The Greek and Hebrew Differ Because of Inner-OG Textual Development Just as we should not assume the MT is the Vorlage of the OG, we also cannot assume that the Greek text of the OG according to a particular manuscript or edition is the actual original translation. Differences between one of these versions of the OG and a Hebrew witness to the HB, then, can stem from textual changes arising in the process of transmission of the Greek (Barr 1968: 247–8; Wevers 1985: 19–20; Gentry 2006: 195– 6; Aejmelaeus 2007a: 72, 74–5; Tov 2012: 122; Tov 2015a: 44, 49, 55–60, 88–9). In Judg. 5:8, for example, for the Hebrew text ‫“( מגן אם יראה ורמח‬a shield, if it appeared, or a

146

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

spear …”), most manuscripts have the following Greek: σκέπη νεανίδων σιρομαστῶν (“a shelter of girls, of lances”) (Tov 2015a: 58–9). It is difficult to imagine what sort of translation technique or theologizing would result in such a translation. Instead, during the process of transmission of the OG itself there has been a mis-division of the words, which should be σκέπην ἐὰν ἴδω σιρομαστῶν (“shelter, if I see it, of lances”). This reading, found in one Greek manuscript, is clearly the reading of the original translation. In the example from Judg. 5:8, most Greek witnesses are glaringly different from the Hebrew. Even if a Greek reading does not obviously differ from the Hebrew, however, we should still consider whether the Greek text contains inner-Greek textual development. If we do not, we may miss an important difference between the OG and a Hebrew witness. In Ps. 5:9, for example, the Hebrew ‫“( הושר לפני דרכך‬make your way straight before me”) seems to be rendered straightforwardly by the OG text of Rahlfs: κατεύθυνον ἐνώπιόν μου τὴν ὁδόν σου (“direct your way before me”).21 Rahlfs is following Codex Sinaiticus in this instance, but nearly all the other evidence for the OG—including Vaticanus, Alexandrinus, numerous other Greek manuscripts, and numerous Old Latin manuscripts—has a slightly different text: κατεύθυνον ἐνώπιόν σου τὴν ὁδόν μου (“direct my way before you”),22 reflecting ‫“( הושר לפנך דרכי‬make my way straight before you”). Considering the weight of evidence, Sinaiticus may reflect revision toward the Hebrew text preserved in the MT. Agreement with the MT is no doubt the reason Rahlfs opted for the reading in Sinaiticus in his critical text—on the basis that Sinaiticus better translates what he considered to be the source text (MT).23 We should not unquestioningly follow the edition of Rahlfs or even a critical edition in the Göttingen series—whether the Greek text of a witness or edition seems to translate the MT or not. Though we should be careful to engage OG text-critical evidence and consider what reading represents the original OG, this does not entail that we cannot ever use secondary OG readings. Different Hebrew witnesses continued to impact the Greek text during its transmission, as Greek manuscripts were often corrected based on Hebrew manuscripts (Screnock 2018: 247–8).

ONGOING RESEARCH QUESTIONS When using the OG for textual criticism of the HB, we should consider every aspect of this typology. In each individual case, we must consider whether differences are best explained by a variant Vorlage, the translator’s decoding of the Vorlage, the translator’s translation technique, content changes introduced by the translator, or inner-OG textual development. If the best explanation is a variant Vorlage, and if that Vorlage can be reconstructed, we can use the retroverted Hebrew in textual criticism of the Hebrew Bible. The most significant factor in this approach is how the individual scholar prioritizes potential explanations. Which of these five categories yields more probable explanations

This is the text translated by Pietersma in NETS, suggesting that he deems it to be the original OG. This is the text chosen by the Larger Cambridge Septuagint. 23 See, similarly, the logic of Pietersma (1990: e.g., 265–6): if a Greek witness translates the Hebrew (by which, he means MT) isomorphically, and an argument can be made that the Greek variation arose in the process of transmitting the Greek text, then the witness agreeing with the Hebrew (=MT) is earlier. Crucially, arguments for inner-Greek variation can include, for Pietersma, both obvious scribal error requiring Greek characters and more common scribal developments (e.g., conforming one text to the wording of a parallel text) that can occur in the transmission of Hebrew as well. 21 22

TEXTUAL CRITICISM OF THE HEBREW BIBLE 147

in most cases? As I have noted, some scholars prefer to explain differences as content change made by the translator, while others prefer to appeal to translation technique, and others to a variant Vorlage. It is often very difficult to tell the difference between a change made as a result of translation and a change made as a result of transmission in Hebrew (Goshen-Gottstein 1963: 130–1, 133–4; Tov 2015a: 178). There is a pervasive assumption that translators were more prone to misunderstand and change the Hebrew text than scribes24—thus putting the burden of proof on arguments for a variant Vorlage. I have recently questioned this assumption.25 Understanding translation technique should be our highest priority, but as a means for accurately reconstructing the Hebrew Vorlage, which is the most probable explanation for differences between the OG and a Hebrew witness in isomorphic translations (Metso and Ulrich 2003: 253).

RECOMMENDED RESOURCES 1. Aejmelaeus (2007a). This volume addresses OG translation technique and its text-critical value (see esp. 71–106). Aejmelaeus argues clearly and presents the most recent thinking on these matters. Some essays address specific issues, while others speak to broad topics, all with a methodological focus. 2. dos Santos, E. C. (1973). This is an extremely helpful resource for researching vocabulary equivalences. Unlike Muraoka’s reverse concordance of Hatch-Redpath (1998), dos Santos provides frequency counts, which are essential for quickly leveraging the data. 3. Goshen-Gottstein, M. H. (1963). In this essay, Goshen-Gottstein tackles the recurring question of attribution to translator or Vorlage, in the specific context of how to present the data in an edition of the Hebrew Bible. 4. Lemmelijn, B. (2009). Through study of a particular corpus, Lemmelijn presents an outstanding statement of general methodology on the use of the OG in HB textual criticism (see esp. 96–125). 5. Tov, E. (1999d). This volume contains numerous valuable studies on the OG that address the use of the OG for textual criticism and the translation process. 6. Tov, E. (2015a). Tov’s methodology for using the Septuagint in HB textual criticism dominates the field. Most text-critics, in particular, follow Tov in their engagement of the Septuagint. 7. Tov, E., and F. Polak (2009). The CATSS database provides a word-for-word alignment of Rahlfs’s OG text and the MT. There are brief annotations indicating the editors’ analysis of the OG, and helpful reconstructions. The database is best used in the context of computer software, where it can be searched according to various terms. 8. Wevers, J. W. (1985). This essay is a helpful statement on the issue of using the OG for textual criticism. 9. Wevers, J. W. (1990, 1993b, 1995, 1997, 1998). Wevers was the preeminent Septuagintalist in North America and his work continues to shape the field there. His Notes series is particularly strong on textual criticism of the OG itself and description of translation technique.

This point is noted by Tov 2015a: 183. Screnock (2018: 237–41; cf. 2017). This is not to say that OG translators were complete masters of the Hebrew language, as there is plenty of evidence that they were not. However, they may have known the language better than we often think they did, and scribes of the Hebrew text were capable of misunderstanding the text in the same ways as translators.

24 25

148

CHAPTER 10

The Septuagint and Qumran GIDEON R. KOTZÉ

INTRODUCTION The Dead Sea Scrolls, the hundreds of ancient manuscripts that have been discovered at various sites in the Judean wilderness region near the Dead Sea, are some of the most important resources for research on the languages, literature, practices, and ideas of early Judaism (Brooke 2019: 119–27). The manuscript finds from the caves at Qumran and Naḥal Ḥever in particular hold special significance for Septuagint studies.1 Of paramount importance are the copies in Greek, Aramaic, and Hebrew of literary writings included in Septuagint corpora, and some of the other compositions among the Qumran scrolls. They provide crucial evidence for the history of the Greek texts of early Jewish writings, information about ancient scribal practices in translating and copying manuscripts, and data relevant to the translators’ and copyists’ knowledge of the source text languages. They also provide details that shed light on the wordings of the source texts of the Greek translations and the impact of certain readings on the subject matter of passages in their Greek formulation. Furthermore, although the number of Greek manuscripts of literary writings from the Judean Desert is very small in comparison to those written in Hebrew and Aramaic, they raise intriguing questions concerning the use of the Greek language for literary purposes in Judea, such as whether they were produced and read there, by and for whom, and in which contexts (Greenspoon 1998: 101).2 These questions are of interest to Septuagint scholarship given the probable Judean provenance of at least some of the Greek translations of early Jewish writings (Tov 2015a: 203–4, 2015b: 365). Of the few fragmentary Greek manuscripts that were recovered from Qumran Caves 4 and 7, six are directly relevant to the study of the Septuagint: two copies of LXX Leviticus, and one each of LXX Exodus, LXX Numbers, LXX Deuteronomy, and the Letter of Jeremiah.3 In this chapter, I use the term “Septuagint” in two ways. First, to refer to the early Jewish literary works in Greek (translations and writings) included in codices such as Codex Sinaiticus, Codex Vaticanus, and Codex Alexandrinus, and second, to refer to the textual versions of these Greek literary works that are represented primarily (but not exclusively) by the codices. The abbreviation LXX also denotes these versions and not a specific stage in the textual history of the literary works (e.g., the Old Greek text). 2 Tov (2008: 340–1) provides statistics on the number of Greek texts from the Judean Desert, the percentages of the total number of texts they comprise at each site, and how many of the Greek texts are non-documentary writings. 3 The compositions represented by 4Q126 (4QUnidentified gr) and 4Q127 (4QpapParaExod gr) have not been determined with certainty, while much debate in the past has surrounded the identification of the texts of 7Q3–5, 6–19. O’Callaghan’s claim (e.g., 1972: 91–100) that fragments from Cave 7 preserve verses from New Testament documents have been refuted by other scholars. According to Tov (2008: 349), “the most likely assumption is that 7Q3–7 contain fragments of either the LXX of the Torah or Enoch.” 1

150

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Table 10.1 Details of the Septuagint Manuscripts from Qumran4 Sigla

Material

Date

Content

4Q119 4QLXXLeva Ra 801

Leather

End of second century BCE, no later than the first century BCE

Lev. 26:2-16

4Q120 4QpapLXXLevb Ra 802

Papyrus

First century BCE

Parts of Lev. 1:11-6:5

4Q121 4QLXXNum Ra 803

Leather

First century BCE–first century CE

Num. 3:40-3, 50-1(?); 4:1(?), 5-9, 11-16

4Q122 4QLXXDeut Ra 819

Leather

Second century BCE

Deut. 11:4

7Q1 7QpapLXXExod Ra 805

Papyrus

First century BCE

Exod. 28:4-7

7Q2 7QpapEpJer gr Ra 804

Papyrus

First century BCE

Ep. Jer. 43-44

It is noteworthy that, with the exception of 7Q2, all these Greek manuscripts from Qumran are Torah scrolls. They antedate the great uncial codices such as Sinaiticus, Vaticanus, and Alexandrinus by centuries. Their wordings were penned by Jewish scribes and provide a teasing glimpse of what the texts looked like before they were transmitted by Christian scribes and, especially, before the recensional work of Origen in the third century CE (Jobes and Silva 2015: 183). The wordings show remarkable similarities and differences when compared to other textual representatives of the Greek translations. Agreements in wording between the Greek scrolls from Qumran and other Greek manuscripts have led scholars to judge that they are representatives of the same translation and not evidence of independent translations. This judgment has determined the way the disagreements in wording have been interpreted. Some of the readings in the Greek Qumran manuscripts that differ from those in the critical Göttingen editions are unique, while others are also attested in a few later manuscripts. In certain cases, the variants in the Qumran manuscripts are regarded as better reflections of the Old Greek text than the readings selected for inclusion in the Göttingen editions. The latter would then be revisions or approximations to the Hebrew wording represented by the Masoretic text. At least some of the Old Greek readings in the Qumran manuscripts are thought to have been based on Hebrew wordings that differed from the Masoretic text. In other instances, however, the variants in the Qumran manuscripts are seen as revisions of the Old Greek text or as scribal errors. Revisions appear in places other than where the Greek text departs from its Hebrew counterpart(s). This implies that ancient scribes in pre-Christian times adapted Tov (2008: 350–61), Wevers (2005: 1–10), Rahlfs and Fraenkel (2004: 150–4), Fabry (2001: 139–45), Skehan, Ulrich and Sanderson (1992: 161–5, 167–86, 187–94, 195–7), Parsons (1992: 10–12), Baillet (1962a: 142–3, 1962b: 143), Kahle (1959: 223).

4

THE SEPTUAGINT AND QUMRAN 151

the wordings of Greek manuscripts for various reasons and not only to bring the Greek wording closer to the Hebrew (Greenspoon 1998: 109). Accordingly, because of their age and the nature of their wordings, the Greek manuscripts from Qumran are invaluable evidence for the study of the history of the Septuagint texts (especially their early history), the source texts of the Old Greek translations, and scribal activities in translating and copying the Greek versions of early Jewish writings. Another important source of knowledge about the early textual history of the Septuagint is the Greek Minor Prophets scroll from Naḥal Ḥever (8ḤevXII gr/Ra 943).5 This manuscript, of which only fragments have survived, preserves portions of the texts of Jonah, Micah, Nahum, Habakkuk, Zephaniah, and Zechariah written by two different scribes.6 It has been dated to the later part of the first century BCE on the basis of a paleographic examination of the scroll’s scripts (Parsons 1990: 19–26).7 Analyses of its wording have concluded that 8ḤevXII gr does not represent the Old Greek text or an alternative translation, but a revision of the Old Greek towards a Hebrew text that was similar to, but not identical with the Masoretic text (Barthélemy 1963: 196; Tov 1990a: 145–53). The revision entails modifications of the Old Greek wording such as the exact rendering of Hebrew sentence constituents, additions and omissions, changes of the forms of nouns, pronouns and verbs, stylistic rearrangements of words so that the Greek text would agree with the Hebrew, as well as the choice of lexical equivalents that are closer to the meanings of Hebrew adjectives, nouns, verbs, prepositions, and particles than the ones in the Old Greek text (Barthélemy 1963: 196–7; Tov 1990a: 131–40). Barthélemy (1963: 198–200) argues that 8ḤevXII gr exhibits the characteristics of, and therefore belongs to, the Kaige group of translations and revisions.8 These translations and revisions share (in  varying degrees of consistency) a number of equivalences, especially the rendering of ‫וגם‬/‫“( גם‬also, even”) with καί γε (“indeed”),9 and a similar approach to translation technique (Gentry 2016b: 213). Barthélemy dubbed the members of the Kaige group predecessors of Aquila, “for many of the characteristics initiated and sporadically applied in these texts came to fruition, often in painful literalness, in the second century CE recension that is identified with Aquila” (Greenspoon 1998: 105). In any case, the Minor Prophets scroll from Naḥal Ḥever illustrates that the adaptation of the Greek versions of early Jewish writings to make them agree more closely with details in Hebrew texts was at least one of the activities some Jewish scribes engaged in early in the transmission history of Greek manuscripts.

Tov (1990a) and Rahlfs and Fraenkel (2004: 156–60). The arrangement of the books follows the sequence in the Masoretic text, not that of the Septuagint (Tov 1990a: 8). 7 Barthélemy (1963: 167–8) prefers a mid-first-century CE date for the scroll, while Kahle (1959: 226) quotes the view of C. H. Roberts that the manuscript was written between 50 BCE and 50 CE. 8 Noteworthy members of the Kaige group that Barthélemy (1963: 47) mentions, in addition to 8ḤevXII gr, include LXX Lamentations, LXX Canticles, LXX Ruth, some textual representatives of Judges, sections in 1–4 Kingdoms (2 Kgdms 11:2-3 Kgdms 2:11; 3 Kgdms 22:1-4 Kgdms 25:30), the Theodotion version of Daniel, the asterisked additions to LXX Job attributed to Theodotion, the Theodotion column of the Hexapla, and the Quinta version of the Psalms. Some scholars (e.g., Tov 2012: 143) refer to the revision as Kaige-Th, because of its association with Theodotion, but Gentry (2016b: 213) warns that this designation is more confusing than enlightening. Subsequent studies have built on the results of Barthélemy’s research and it now seems that the translations and revisions of the Kaige group are not homogenous in character and not all of the texts that Barthélemy mentions are members of this group. Cf., e.g., Gentry (2016b: 215–27). 9 Barthélemy (1963: 31–80) discusses nine characteristics that members of the Kaige group have in common and endeavors to connect these characteristics to exegetical principles of first-century Palestinian rabbis. 5 6

152

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Among the great many Hebrew and Aramaic scrolls of writings that were discovered at Qumran, there are manuscripts that grant us a window into the probable wordings of source texts of some Greek translations, as well as copies of compositions that were previously only known in translation.10 For example, five fragmentary manuscripts of the book of Tobit—four in Aramaic and one in Hebrew—from Cave 4 at Qumran have been published by Fitzmyer (1995: 1–76). The Qumran texts lend a degree of certainty to the view that Tobit’s language of composition was Semitic, not Greek (Fitzmyer 2000: 151; Stuckenbruck and Weeks 2015: 239). Indeed, the Greek translation was arguably based on a text that was very close to the version preserved by 4Q196–200 (Stuckenbruck and Weeks 2015: 239). The Aramaic and Hebrew wordings of the Qumran manuscripts generally agree with the longer Greek version of Tobit, GII in the Göttingen edition (Hanhart 1983b), attested by Codex Sinaiticus, MS 319, and the Old Latin manuscripts. These manuscripts affirm the priority of the longer version as a more likely reflection of the original translation over the shorter version, GI, found in the majority of textual representatives, and the third Greek version, GIII (Fitzmyer 2000: 151–2; Stuckenbruck and Weeks 2015: 246–50). Regarding the early Jewish writings that form part of both the Septuagint and Hebrew Bible corpora, their wordings in the Qumran manuscripts agree in a myriad of details with the Greek translations, where those translations differ from readings in the Masoretic text. These agreements over against the Masoretic text include pluses and minuses of sentence constituents, as well as whole sentences, the formulations of phrases, arrangements of sentence constituents and order of verses, parts of speech, the forms of individual words (e.g., person, number, conjugation), and other details, such as the choice of divine designation.11 In some instances of agreements between the Greek texts and the Qumran scrolls, it is possible that the Septuagint translators and the scribes who copied Qumran manuscripts independently reformulated the wordings of the texts they transmitted in strikingly similar ways. They were either influenced by an interpretation with which they both were already familiar, or they independently hit upon the same Table 10.2 Details of the Tobit Manuscripts from Qumran Sigla

Language

Material

Date

Content

4Q196/4QpapTob ar

Aramaic

Papyrus

c. 50 BCE

Parts of Tob. 1-7; 12-1412

4Q197/4QTobb ar

Aramaic

Leather

c. 25 BCE–25 CE

Parts of Tob. 3-9

4Q198/4QTob ar

Aramaic

Leather

c. 50 BCE

Tob. 14:2-6, 10

a

c

d

4Q199/4QTob ar

Aramaic

Leather

c. 100 BCE

Tob. 7:11; 14:10

4Q200/4QTobe

Hebrew

Leather

c. 30 BCE–20 CE

Parts of Tob. 3-5; 10-14

10 The texts of Ben Sira that were discovered at Qumran (2QSir [2Q18] and Sir 51:13-20, 30 in 11QPsa col. XXIXXII) and Masada (MasSir) are important representatives of the Hebrew version of this book, along with the manuscripts that were found in the Cairo Geniza in the late nineteenth century. 11 Martone (2005, 2007, 2012) provides helpful tools for the consultation of the passages where Qumran manuscripts agree with the Septuagint against the Masoretic text. 12 Schøyen MS 5234, which contains words from Tob. 14:3-4, can be added to the fragments of 4Q196 (Hallermayer and Elgvin 2006: 451–61).

THE SEPTUAGINT AND QUMRAN 153

understanding of a passage. Evidently, ancient scribes, at different times and in various contexts, changed the wordings of their texts in common ways during the processes of transmission. The Greek translations and the Qumran manuscripts are prize exemplars of this kind of scribal activity, which was not always systematic or exhaustive. In many other cases, however, the most likely explanation for the agreements is that the wording of the source text of a Greek translation was the same as the reading in a Qumran manuscript. The latter then contributes to a better understanding of the former. It has even been claimed that specific scrolls are closely related (in different ways) to the purported source texts of certain Septuagint translations. The identification of Qumran scrolls as closely related to the source texts of Septuagint writings potentially has far-reaching implications for the characterization of the Greek texts as translations, the explanation of variant readings in the Greek texts, and the conceptualization of the relationship(s) of the translations with other textual representatives, especially the Masoretic text. This version is often the only complete Hebrew text (apart from the Samaritan Pentateuch in the books of the Torah) with which the Septuagint text can be compared. Furthermore, by virtue of their agreements with the wordings of the Greek translations in many types of readings, the Qumran scrolls give external support and credibility to the procedure of retroversion, that is, the reconstructing of details in the source texts of the Greek translations (Tov 1992: 12– 22, 2015a: 87). Scholars in the past retroverted readings intuitively, but the agreements between the Greek translations and the Qumran manuscripts have justified their efforts: “In spite of known trends of exegesis in the translation, of inner-translational corruptions and of our own ability to get back to the Hebrew text underlying the translation, much of what has been done so far in the area of retroverting the Vorlage of the LXX is now supported by the Qumran finds” (Tov 1992: 21). Interestingly, in a number of passages where the wordings of Septuagint writings vary from those in the Masoretic text, the readings in the Greek translations agree with the Table 10.3 Sample Lists of Qumran Scrolls that are closely related to the Source Texts of the Septuagint13 Lange

Tov

4QLev 4QDeutq 4QSamb 4QJerb d

Ulrich

4QNum 4QDeutq 4QSama 4QSamb 4QJerb 4QJerd 11QPsa col. XXVII b

4QDeut 4QSama 4QSamb 4QJerb

q

García Martínez

Cross

4QExod 4QLevd 4QNumb 4QDeutq 4QSama 4QSamb 4QJerb 4QJerd

4QExodb 4QNumb 4QDeutq 4QSama 4QSamb 4QJerb

b

Cf. Lange (2016a: 124), Tov (2015a: 208–12; 2015b: 353–67), Ulrich (2015: 158–60), García Martínez (2005: 258–9), and Cross (1995: 132–8). Scholars use different criteria, such as statistics of agreements and disagreements in wording or shared readings (common errors or characteristic deviances from the Masoretic text) to determine the relationship of a Qumran scroll to the purported Vorlage of a Greek translation. It is therefore to be expected that they do not always agree on the characterization of the text of a Qumran scroll in relation to that of the Greek translation. The case of 4QSama, for example, is heavily debated, because it shows agreements with the representatives of the Old Greek text against the Masoretic text in significant readings, but it also disagrees with the Old Greek representatives in important instances. Moreover, the scroll exhibits many unique readings (Tov 2015a: 209–11; Lange 2017a: 320–3). 13

154

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

wordings of Dead Sea Scrolls. Even if these agreements are in seemingly small details, such as pluses and minuses of prepositional phrases or single words, and changes of a conjunction or the person of a verb, they can have a significant impact on the subject matter of passages. For example, Schiffman (1992: 277–97) discusses passages in the Temple Scroll that share halakhic variants with Septuagint manuscripts. This means that the “readings in the scroll and the LXX either represent a different legal ruling than that of MT, or seek to clarify a legal question left undetermined in MT” (Schiffman 1992: 279). In such instances of similarity in wording (and subject matter) between the Septuagint texts and the Qumran scrolls, it should be determined whether the readings were part of the source texts of the Greek translations or whether the Greek translators independently introduced interpretations into the wordings that are comparable to what is found in the Qumran scrolls. In any event, the textual data illuminate how ancient scribes interpreted and adapted the wordings of passages during the transmission of the texts. They thereby highlight “the intimate links between the scribal process of passing on texts, and the exegetical process of interpreting them” (Schiffman 1992: 293). Finally, the Greek translations in the Septuagint corpora provide evidence of the scribes’ knowledge of Hebrew and Aramaic and, at times, this evidence can contribute to a better understanding of linguistic features that appear in the Dead Sea Scrolls. By the same token, scholars have identified cases where the language of the Qumran scrolls sheds light on seemingly peculiar renderings in the Greek translations.

VIEWS AND DEBATES Although the great value of the Qumran scrolls for understanding the Septuagint texts, their history and subject matter, the source texts of the translations, and the translators’ command of the source languages can hardly be disputed, scholars do debate some matters of detail. For instance, the scholars who have analyzed the readings in the Greek manuscripts from Qumran, especially those in 4QLXXLeva, 4QpapLXXLevb, and 4QLXXNum, have failed to reach a consensus on how best to explain the passages where these readings differ from the ones that have been included in the Göttingen editions of Leviticus and Numbers. The extent of the disagreement among scholars on this issue can be illustrated with a few selected examples.14 Leviticus 26:9 4QLXXLeva

[και εσται μο]υ η διαθ. ηκη εν. υ. μ. ι̥ ν̥ [and m]y covenant [will be] with you

LXXGö

καὶ στήσω τὴν διαθήκην μου μεθʼ ὑμῶν and I will establish my covenant with you (NETS)

MT SP

‫והק(י)מתי את בריתי אתכם‬ and I will maintain my covenant with you (NRSV)

In these examples, the abbreviation LXXGö refers to the Göttingen critical editions. The editions of Leviticus and Numbers were prepared by Wevers (1982a, 1986). Dots under the text (e.g., α̣δ̣ε̣) indicate a damaged letter that may be restored with a high degree of certainty. Circlets under the text (e.g., α̥δ̥ε̥) indicate a damaged letter that is more tentatively restored.

14

THE SEPTUAGINT AND QUMRAN 155

Ulrich (1992: 58) suggests that the reading in 4QLXXLeva reflects a variant Hebrew wording such as ‫“( והי)ת(ה בריתי בתוככם‬and my covenant will be in your midst”) or ‫“( ובריתי אתכם‬and my covenant is with you”). Van der Louw (2008c: 389), however, thinks that this is unlikely. He interprets the reading in 4QLXXLeva as an example where the translator avoided the misunderstanding that a literal translation such as “I will erect my covenant” might have caused. It is conceivable that this rendering could create the wrong impression that there had not been a covenant before. In the opinion of Tov (2008: 351), 4QLXXLeva represents the free rendering of the Old Greek text and the main manuscript tradition adapted the wording towards the Masoretic text. Himbaza (2016: 300) and Fabry (2001: 140) agree with Tov, but Wevers (1997: 441) does not. In his view, the wording in the Qumran scroll was probably influenced by v. 11, where the texts of Codex Vaticanus and Codex Alexandrinus, together with other manuscripts, contain the reading τήν διαθήκην μου ἐν ὑμῖν (“my covenant with you”). Conversely, Petersen (2009: 503) claims that the scribe of 4QLXXLeva was faced with “the difficulties of translating the Hebrew covenantal language” and chose to reshape the wording of the received text in order to present a more intelligible Greek reading. Leviticus 26:11 4QLXXLeva

κ̥αι ου βδ̥ελυξομαι υμας and I shall not abhor you

LXXGö

καὶ οὐ βδελύξεται ἡ ψυχή μου ὑμᾶς and my soul shall not abhor you (NETS)

MT SP

‫ולא תגעל נפׁשי אתכם‬ and my soul shall not abhor you (ESV)

Petersen (2009: 504) proposes that the Qumran scroll smoothed over the “exclusively Semitic expression” in the Old Greek by adopting a more acceptable rendering, but Wevers (1997: 442) maintains that the bilingual copyist of 4QLXXLeva was “unconsciously influenced by his knowledge of the parent text, and changed βδελύξεται ἡ ψυχή μου to βδελυξομαι, which means the same.” Yet, Himbaza (2016: 301) notes that this is one of the cases where the Qumran scroll cannot reflect a revision of the rendering represented by LXXGö, seeing as the latter is more literal and closer to the Masoretic text. Ulrich (1992: 76) and Tov (2008: 353) regard βδελυξομαι as the Old Greek reading, which might ̥ have been based on a Hebrew variant, ‫“( אגעל‬I will abhor”). If ‫ אגעל‬is the original reading, so Tov argues, the wording represented by the Masoretic text, Samaritan Pentateuch, and LXX could be a euphemistic anti-anthropomorphic correction. The correction adds an intermediary entity (‫ נפׁש‬and ψυχή) to avoid having God himself as the subject of the verb. In Skehan’s view (1975a: 222), the Qumran reading “introduces an anthropomorphic turn which is not in the original text.” Leviticus 3:12 and 4:27 In the texts of these verses that have been preserved in 4QpapLXXLevb, the divine name is written as Ιαω. Some scholars argue that this writing of the divine name is more original than κύριος (Skehan 1980: 29; Tov 2008: 356–7; Ulrich 2015: 153–4). Wevers

156

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

(2005: 23) demurs and accepts the view that the latter is the Old Greek equivalent for the tetragrammaton (cf. Pietersma 1984: 85–101). Numbers 3:40 4QLXXNum

αριθμησον ̥ [παν πρωτοτοκον αρσεν των υι]ων Ισραηλ απ[ο μηνιαιου και . . επανω] Count [every firstborn male of the so]ns of Israel fro[m a month old and above]

LXXGö

ἐπίσκεψαι πᾶν πρωτότοκον ἄρσεν τῶν υἱῶν Ἰσραὴλ ἀπὸ μηνιαίου καὶ ἐπάνω Enroll every firstborn male of the sons of Israel from a month old and above (NETS)

MT SP

‫פקד כל בכר זכר לבני יׂשראל מבן חדׁש ומעלה‬ Enroll all the firstborn males of the Israelites, from a month old and upward (NRSV)

Ulrich (1992: 70–1) and Tov (2008: 359) argue that the Qumran reading preserves an earlier stage of the text of the Greek translation than LXXGö, before ἐπισκέπτομαι became the standard equivalent for ‫פקד‬. Wevers (1998: 50–1) and Rösel and Schlund (2011: 445) agree with Quast (1990: 249) that ἀρίθμησον is not the original rendering, but a secondary change that attempted to represent the meaning of the Hebrew more exactly. Skehan (1977: 43) is also of the opinion that ἀρίθμησον is a pre-Christian revision of the Old Greek text. In another study, Wevers (1982b: 238*) calls the use of ἐπισκέπτομαι to translate ‫ פקד‬in its meaning “to number” a Hebraism, because the Greek verb does not fully match the broad semantic range of the Hebrew one. Petersen (2009: 488) approvingly quotes Wevers’ conclusion that the variant in 4QLXXNum clarifies “a Hebraic kind of Greek by a more idiomatic text” (Wevers 1982b: 238*).

OTHER DEBATED TOPICS A number of debates revolve around text-historical issues. Scholarly opinions are divided on which textual representatives have priority in the textual history of a composition and how Qumran scrolls relate to these textual representatives. For the Septuagint this means, first, that scholars differ in their evaluations of the status of Greek Qumran scrolls as representatives of the Old Greek text. Second, they also disagree on whether those Hebrew manuscripts from Qumran that appear to be close to the Vorlage of a Greek translation substantiate the view that the Septuagint text reflects a different Hebrew version to the one in the Masoretic text. The cases of 4QLXXNum and 4QJerb discussed below may serve as illustrations of these issues. Although he admits that the evidence is inconclusive, Tov (2008: 360–1) agrees with Ulrich (1992: 76) that 4QLXXNum is a superior representative of the Old Greek text than LXXGö. Skehan (1977: 39), however, describes 4QLXXNum as “a considerable reworking of the original LXX [i.e. the Old Greek] to make it conform both in quantity and in diction to a Hebrew consonantal text nearly indistinguishable, within the limited scope of our evidence, from that of MT. The manner of this reworking is therefore that

THE SEPTUAGINT AND QUMRAN 157

of the καίγε, or proto-Theodotionic, recension.” Lange (2016a: 150) affirms Skehan’s assessment of the Qumran scroll. Fabry (2001: 142) also thinks that 4QLXXNum is somewhat closer to the Masoretic text than other textual representatives of LXX Numbers. Wevers (1982b: 235*–9*), however, indicates that the revision supposedly exhibited by 4QLXXNum is not specifically Hebraizing in nature, at least not in the same way as the revisions of the Kaige group are. Rather, the scroll shows signs of presenting a clearer and more exact Greek text than LXXGö. Petersen (2009: 494–5, 509) argues that the Qumran manuscript adheres closely to the Old Greek translation and to a Hebrew text that was almost identical to the Masoretic text. The revision was primarily concerned with clarification of the Hebrew and Greek texts. Based on the words that have survived in the fragmentary manuscript of 4QJerb and calculations of the length of lines in the column, some scholars argue that the wording of Jeremiah 10 in the scroll did not contain what are vv. 6–8 and v. 10 in the version of the passage represented by the Masoretic text. Also, v. 5b is positioned after v. 9. Old Greek Jeremiah reflects the same minus and difference in verse order as 4QJerb, when compared to the Masoretic text. These agreements between the Qumran scroll and the Old Greek text in characteristic readings that deviate from the Masoretic text are the grounds for scholars to see a close relationship between 4QJerb and the Hebrew source text of Old Greek Jeremiah (Saley 2010: 2; Ulrich 2015: 158; Tov 2015a: 209; Lange 2017b: 516, 537). Moreover, the verses that are lacking in 4QJerb and the Greek translation have a common theme. They extol the incomparability, greatness, and fearsomeness of YHWH, who is praised as the true and living God and king, while the other verses deride the idols of the nations. According to Tov (2015a: 209), it is likely that this doxology was added later to the version represented by the Masoretic text. 4QJerb and Old Greek Jeremiah would then reflect an earlier, shorter version of the passage. Other scholars disagree and maintain that the Greek translator was responsible for major differences between Old Greek Jeremiah and the Hebrew text. In their opinion, 4QJerb does not provide sufficient evidence for the assumption that the Greek translation was based on a Vorlage that was earlier than the version in the Masoretic text (cf. Fischer 2017: 543–53). The Dead Sea Scrolls are important for any assessment of the language situation in Judea during the period of early Judaism (Gzella 2019: 192). “The occurrence of scrolls in Hebrew, Aramaic and Greek in the caves at and near Qumran and variously at other sites is evidence of some degree of multilingualism amongst Jews in the region at the time. What such multilingualism indicates about the Jews of Judea in the late Second Temple Period has yet to be fully worked out” (Brooke 2019: 121). There is some disagreement over the question whether the presence of Greek manuscripts from Qumran implies that the sectarian movement used the Greek language to communicate and inculcate religious ideas.15 Tov (2008: 342) and Fabry (2001: 133) are skeptical, but Lim (2000: 70) believes that some members of the Qumran sect must have been able to read Greek, and Ulrich (2015: 165–6) is open to the possibility that the leaders at Qumran, or other members of the sect, may have studied their scriptures also in Greek.16 Leaney (1976: 283–300) argues in favor of the idea that the Greek Qumran manuscripts were written

While Aramaic was the dominant language of daily life in early Judaism (Brooke 2019: 121; Gzella 2019: 197, 201), there is evidence to suggest that the Qumran sect used Greek in the context of economic transactions (Richey 2012: 177–97). 16 Cf. VanderKam (2001: 175–81). 15

158

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

for members of the movement and, therefore, that Greek was the first language of some of the sectarians.

RESEARCH METHODS AND THEORIES Most of the research on the topic of the Septuagint and Qumran has been conducted with text-critical concerns in mind. Text-critics of early Jewish writings examine the available representatives of the wordings and subject matter of the compositions, explain the readings in these textual representatives on the basis of evidence for how ancient scribes copied and translated their source texts, and establish the likely course of the textual history of the compositions. Concerning the textual criticism of the Septuagint, the data in the Greek manuscripts from Qumran and Naḥal Ḥever have, with almost no exceptions, been incorporated into a text-historical model that is associated especially with the name of Paul de Lagarde. The model postulates the existence of a common ancestor to which all the wordings in the available textual representatives are genealogically related. In other words, there was one original translation that underwent various revisions in its subsequent transmission history. Many scholars express the view that the readings in the Greek Torah manuscripts from Qumran fit within the text tradition represented by the uncial codices and take this as confirmation that Lagarde’s theory of Septuagint origins is correct: “If de Lagarde’s [Urtext] theory on the history of the LXX needed any further support, it is provided by the texts from the Judean Desert” (Tov 2008: 363). Fernández Marcos (2000b: 71) notes that the evidence from the Greek Qumran scrolls “tips the balance, we think conclusively, in favour of Lagarde’s theory rather than Kahle’s.” Ulrich (2015: 166) also concludes that the Qumran texts “generally confirm the approach of Paul de Lagarde, as opposed to that of Paul Kahle,” while Skehan (1975b: 269) states that the Greek manuscripts from Qumran “counter the sweeping theory of Kahle with tangible facts.”17 With regard to the textual criticism of the Hebrew Bible, the agreements between the Qumran scrolls and the Greek translations confirm the value of the Septuagint as a tool for this discipline (Tov 1992: 21). For their explanation of cases where the Septuagint version differs in wording from the Masoretic text, text-critics have to determine whether a variant in the Greek translation was based on the wording of its source text, created by the translator, or introduced in transmission. By justifying the practice of retroversion, the agreements between the Qumran scrolls and the Septuagint version strengthen the possibility that many readings in the Greek translations reflect a source text that differed from the version in the Masoretic text. Text-historically speaking, the readings in the source texts of the Greek translations may be earlier than their counterparts in the Masoretic text, or the source texts may contain changes to an earlier Hebrew wording that were incorporated at a stage of transmission before the translations were made from them. This also implies that not all the variants in the Septuagint are “free” translations of readings that were identical to the ones in the Masoretic text or are the result of the translators’ misunderstanding of their source texts.

17 Faulkenberry Miller (2007: 1–28), however, has suggested that not all the readings in 4QLXXLeva fit well within the model of a single translation tradition and, therefore, this manuscript does not consistently support Lagarde’s Urtext theory.

THE SEPTUAGINT AND QUMRAN 159

ONGOING RESEARCH QUESTIONS Septuagint scholarship has benefited a great deal from the manuscript discoveries in the Judean Desert and more studies in areas where the Qumran scrolls have an impact on the understanding of the Greek texts are needed. The following are some issues and questions that merit the attention of scholars in future research. First, judging from the differences in opinion regarding the explanation of readings in the Greek manuscripts from Qumran, it is clear that the arguments scholars have put forward in support of their views have not been equally convincing. There is a need for fresh examinations of 4QLXXLeva, 4QpapLXXLevb, and 4QLXXNum that formulate more cogent arguments in favor of the proposed interpretations than the existing ones. Second, the view that the readings in the Greek manuscripts from Qumran do not only support the idea of an original translation, but are, in some instances, closer to its wording  than the readings in the main codices, may require a revaluation of the translation profile of LXX Leviticus and other books (cf. Himbaza 2016: 306–7, Van der Louw 2008c: 395–6). Third, the agreements and disagreements in wording between the Septuagint texts and Qumran manuscripts are important for issues pertaining to the textual history of early Jewish writings, as well as the wordings and source texts of the Greek translations. These data will have to feature prominently in future attempts to better conceptualize the development of the literary writings during the early Jewish period than text-historical theories have done to this point. Fourth, the Qumran scrolls can contribute to the study of the subject matter of Septuagint writings. For example, the ways in which the variant readings in the Greek manuscripts from Qumran affect the subject matter of LXX Leviticus and LXX Numbers must still be determined. This also applies to many passages where the wordings of the Greek translations and Qumran scrolls agree against the Masoretic text. Moreover, the Septuagint texts partook of the “exegetical background” of the Second Temple era, which is also reflected by the “non-biblical” compositions from Qumran (García Martínez 2005: 262–8). Indeed, these Qumran scrolls are a rich store of information about interpretations and interpretive practices that were current during the period of early Judaism. Seeing as the Greek texts were most probably penned by learned Jewish scribes who well knew the source materials and existing interpretations thereof (cf. van der Kooij 2002: 171), it remains to be seen how relevant the information that may be culled from the Qumran scrolls is for understanding the Septuagint. Fifth, the diverse practices ancient scribes employed in transmitting literary writings raise the interesting question to what extent the varied “techniques” scribes used in translating and revising early Jewish writings in Greek correspond to or overlap with the different ways scribes copied manuscripts of such texts in their languages of composition (cf. Screnock 2017). Sixth, the questions surrounding the use of Greek for literary purposes in Judea during the time when early Jewish writings were translated into Greek deserve more examination, especially the possibility that the Greek manuscripts from Qumran were used and perhaps produced by members of the sectarian movement. Furthermore, the possibility that the renderings in the Septuagint and the language of the Qumran scrolls can be mutually illuminating should not be overlooked in future research.

160

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

RECOMMENDED RESOURCES 1. Discoveries in the Judean Desert (1951–2011). The transcriptions of the (reconstructed) texts and the photographs in the plates of the DJD volumes are useful starting points for analyzing the wordings that have been preserved in the Qumran and Naḥal Ḥever manuscripts. They include introductory remarks on the manuscripts, notes on the transcriptions and reconstructions, as well as lists of variant readings. 2. Barthélemy (1963). For the study of the Greek Minor Prophets scroll from Naḥal Ḥever and the Kaige group of texts and revisions, this classic study remains essential reading. 3. Brooke and Lindars (1992). This volume contains fine essays that address many of the topics covered by this chapter. 4. Fabry (2001). Fabry provides a helpful overview of the Greek manuscripts among the Dead Sea Scrolls, as well as brief analyses of the scrolls from Qumran and the Minor Prophets scroll from Naḥal Ḥever. 5. García Martínez (2005). The volumes in the La Bible d’Alexandrie series are important resources for the study of the Septuagint and, in this article, García Martínez indicates how the Qumran “biblical” and “non-biblical” scrolls are relevant to the topics that are treated in the text-comparative and exegetical notes of these volumes. 6. Greenspoon (1998). This is a very accessible introduction to the ways the manuscripts from the Judean Desert have shed light on the Septuagint. 7. Tov (2008). Tov’s article covers much of the same ground as that of Fabry, but he expresses different opinions on some of the issues. It would be a good exercise to read these studies together and to compare their results. 8. Tov (2015b). In this essay, Tov gives an up-to-date summary of how the Hebrew manuscripts from Qumran that show agreements with the Septuagint texts are relevant to the text-critical study of the Hebrew and Greek Bibles. 9. Ulrich (2015). Chapter 10 on the Septuagint scrolls in Ulrich’s book presents a valuable overview of the issues surrounding the topic of the Greek translations and Qumran. Scholars and students beginning to study this subject will undoubtedly benefit from its clear and concise presentation of Ulrich’s views on these issues. 10. Wevers (2005). In addition to a collation of the Greek Torah manuscripts from Qumran, Wevers presents his evaluations of the readings in these manuscripts against the backdrop of his work as the editor of the Göttingen Septuaginta volumes of the Pentateuch.

CHAPTER 11

The Septuagint and the Major Recensions VILLE MÄKIPELTO

INTRODUCTION One of the biggest challenges in Septuagint research is that the earliest form of the translation, the so-called Old Greek text (= OG), is not directly available to us.1 Most Septuagint research follows the paradigm of Paul de Lagarde, according to which, in the case of most Septuagint books, the diversity of readings visible in the Greek manuscripts is a result of secondary revisions to an original Old Greek translation made from the Hebrew (Fernandez Marcos 2000b: 67–84). In other words, what we find in the hundreds of Greek manuscripts are several layers of ancient recensions and revisions, undertaken at different times, by different people in various geographical locations.2 To make the situation more complex, individual manuscripts do not contain merely one textual tradition, but layers of Old Greek and different recensions, to varying degrees, depending on the manuscript. These recensions are an important topic of research as such, because, among many other things, they illuminate the early reception of biblical texts in Judaism and Christianity. Moreover, to put it bluntly: without the study of recensions, one could not utilize the Septuagint in any kind of text-critical research, since an approximate reconstruction of the Old Greek is dependent upon the study of its later recensions. Insights into the recension history of the Septuagint are necessary in order to evaluate the source of variant readings in the Greek manuscript evidence. This chapter provides an overview of the major Jewish and Christian recensions and their research. These include the early Jewish recensions and translations, that is, the Hebraizing Kaige revision and “the Three”—Aquila (αˊ), Symmachus (σˊ), and Theodotion (θˊ)—as well as the major Christian recensions: the Hexaplaric (O) and the Lucianic (L). Since these phenomena are tightly linked with textual criticism of the Septuagint, text-critical matters will receive consistent attention throughout. In this discussion, I will

I wish to thank Dr. Tuukka Kauhanen, as well as the members of the text-critical study circle lead by Timo Tekoniemi at the University of Helsinki, for their helpful comments on this chapter. 2 The terms “recension” and “revision” refer to different extents of reworking. While “recension” conveys the idea of a more thorough and systematic reworking, “revision” refers to more sporadic reworking. Furthermore, both may involve several layers and scribal hands. In this chapter, I will refer to the Kaige reworking as “revision” (sometimes even “tradition” may be more accurate), and the Lucianic and Hexaplaric reworkings as “recension.” Moreover, “revision” can be taken as meaning action, while “recension” as referring to the result of a systematic revision. 1

162

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

demonstrate the recensions mainly with examples from the Historical Books (Joshua, Judges, and Samuel-Kings or the book of 1–4 Kingdoms), which are where most research on Septuagint recensions is focused.3 Nevertheless, the list of key resources for further reading points to plenty of examples beyond the Historical Books. While my aim is to outline the basic features of the ancient revisions, I also wish to demonstrate how they should be considered when embarking upon Septuagint research in practice. Furthermore, one should keep in mind that when dealing with these issues we are operating in an area of active research, where many debates are still underway. Therefore, I will note areas where current debates are especially active, and where further research might eventually pave the way for new insights. Ultimately, the goal is to equip the readers to engage in their own critical research in this complex but invigorating area of Septuagint research. Before discussing the various views and debates related to the major recensions, one preliminary note is in order. Writing a general introduction to any topic in Septuagint research is complicated by the fact that every book in the Septuagint is different. This applies not only to the translation profiles of individual books, but also to the existence, nature, and extent of the various recensions. For instance, certain single early manuscripts are more influenced by recensional texts in some books than others. For some passages we may possess no evidence of a particular recension, while for others we may only possess evidence affected by the recensions. Therefore, after reading this chapter, one is advised to turn to the research literature dealing with specific Septuagint books, in order to successfully apply what has been discussed here to specific material. The T&T Clark Companion to the Septuagint (Aitken 2015b), containing introductory articles on every book in the Septuagint, is helpful in this task.

VIEWS AND DEBATES Early Jewish Revisions and Translations Beginning already in early times—probably in the first century BCE (Aejmelaeus 2017b: 183–4)—learned people started noticing that the text of the Greek Scriptures did not always match the Hebrew version, which later became the authoritative text in rabbinic Judaism (the proto-Masoretic text). This observation was the impetus for most of the major recensions, and even new translations: the Greek text needed to be reworked to make it conform with the proto-Masoretic text. These recensional activities were first undertaken by learned Jewish scholars, and later by Christian church fathers. The Kaige Revision The French scholar Dominique Barthélemy discovered the earliest known Septuagint recension. In 1963, he published the famous first-century CE Greek Minor Prophets scroll (8ḤevXII gr), which was discovered at the site of Naḥal Ḥever. His work Les Devanciers d’Aquila (1963) has been described by many as perhaps the most important study published in the field of Septuagint studies during the twentieth century. Before Barthélemy, Henry St. John Thackeray had also noticed that a distinct translation technique, employing literal word renderings, was used in two specific sections within the book of Kingdoms (see below). He concluded that these sections could not have come from the hands of the original translator (Thackeray 1907: 263). Barthélemy noticed that These are also the Septuagint books for which a critical Göttingen edition is not yet complete, which is why extra care is needed when working with them.

3

THE SEPTUAGINT AND THE MAJOR RECENSIONS 163

the translation technique of the newfound scroll was strikingly similar to those sections in Kingdoms, and surmised that both represented, not a new translation, but a revision of the original translation. According to him, this revision arose from proto-rabbinic Palestinian circles, and was characterized by the motivation to make the Greek translation match more closely with the Hebrew proto-Masoretic text. To be more accurate, the aim was to achieve formal equivalence at the word level between the Greek and the authoritative Hebrew versions. One common example illustrating this pursuit is the translation of the Hebrew particle ‫“( גם‬also”) or ‫“( וגם‬and also”) with the concordant Greek equivalent καί γε; this feature also gave rise to the name given to the revision (Kaige) by modern scholars. Such a combination of conjunctions is striking because classical Greek usually requires an intervening word. This usage of καί γε, and the tendency to even combine the particles to produce καίγε, became more common only later, toward the first century BCE (Aitken 2015a: 30). The extent of the influence of the Kaige revision is still a disputed and open issue. It can be clearly observed in the B text (Codex Vaticanus and related Greek manuscripts) of the books of Judges, Samuel, and Kings. In the B text, the whole of Judges is heavily influenced by this revision, while in Samuel-Kings (1-4 Kingdoms) it is mainly visible in two sections: 2 Samuel 10-1 Kings 1 and 1 Kings 22-2 Kings 25. These are designated as the Kaige sections and referred to using the sigla βγ and γδ, respectively, which in the B text are characterized by a remarkable formal equivalence. For this reason, the Old Greek is often found in the earlier layers of the so-called Lucianic manuscripts. There are different positions concerning where exactly the Kaige section begins in 2 Samuel, with the latest theory positing 2 Sam. 10:6 (Wirth 2016). Moreover, sporadic influence of the Kaige revision has been found in Samuel-Kings even outside the Kaige sections (Aejmelaeus 2017b). As an exception among the historical books, the B text of Joshua does not display a distinctive Kaige layer of revision, but perhaps only sporadic Kaige readings (Mäkipelto 2018: 19–20). The influence of Kaige has also been posited in books such as Ezekiel, Lamentations, Job, Psalms, and Daniel. However, when one moves to the books that were translated closer to first century BCE or even later, the usage of the term Kaige becomes more problematic, since the line between revision and new translation is not clear. For instance, even the initial Greek translation of the Song of Songs demonstrates a close resemblance to the Kaige tradition of revision. How does one distinguish the presence of Kaige revision? There are no simple formulas for recognizing the source of any readings, but text-critical evaluation of all the available variants usually results in an acceptable probabilistic conclusion that a specific reading represents Kaige. There are certain characteristic features of this revision, and recognizing these helps in the evaluation. Barthélemy (1963: 48–80) has listed several helpful examples of the most common features. The latest list of Kaige features has been published by Tim McLay (1998: 131–4), updating earlier work by Greenspoon and others. After McLay’s work, several later studies have also identified more features. Most characteristics in the list are the result of implementing a lexical concordance principle; in other words, certain Hebrew words were always rendered with a certain Greek equivalent. For example, Kaige routinely replaced dynamic translation equivalents of the Hebrew ‫“( איׁש‬man”), such as ἕκαστος (“each”), with the literal Greek ἀνήρ (“man”). In Judg. 2:6, for instance, the majority text states that, after being dismissed by Joshua, the Israelites went “each to his house” (ἕκαστος εἰς τὸν οἶκον αὐτοῦ). Kaige brings this into lexical conformity with the Hebrew by replacing “each” (ἕκαστος) with “man” (ἀνήρ). Another prominent example is the representation of the Hebrew first-person pronoun ‫“( אנכי‬I”) with the expression ἐγώ

164

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

εἰμί (“I am”), even when a finite verb follows; this was necessary since the simple ἐγώ (“I”) was already reserved for the more common Hebrew ‫“( אני‬I”). There are several examples of this phenomenon in the B text of Judges. For example, in Judg. 8:5 the more natural Greek expression ἐγὼ δὲ διώκω (“for I pursue”) was revised by Kaige to ἐγώ εἰμι διώκων (“I am pursuing”). Furthermore, a well-known feature is the preference for the Greek aorist. While the original translators often employed the Greek present and imperfect forms in their translations of the Hebrew consecutive imperfect (wayyiqṭōl), Kaige routinely substituted these renderings with the Greek aorist. This phenomenon is nicely illustrated by, for example, 2 Sam. 11:14 where two original historical presents are preserved by the Lucianic manuscripts, while the majority of manuscripts contain aorist forms as the consequence of Kaige revision: καὶ ἔγραψεν (L: γραφει) Δαυιδ βιβλίον πρὸς Ιωαβ καὶ ἀπέστειλεν (L: αποστελλει) ἐν χειρὶ Ουριου (“David wrote a document to Joab and sent if by the hand of Ourias”4). In this case, the OG should be reconstructed on the basis of the L readings. The presence of these and other equivalents in a text does not automatically mean that the text should be considered part of the Kaige tradition. However, they are helpful tools when analyzing such texts in relation to all of the other evidence. The Three Before the discovery of the Kaige revision, the work of three Jewish translators and revisers was well known: Aquila (αˊ), Symmachus (σˊ), and Theodotion (θˊ). Indeed, the title of Barthélemy’s seminal monograph (Eng. “The Precursors of Aquila”) refers to the fact that the isomorphic translation style already represented by the Kaige revision eventually culminated in the Greek translation undertaken by Aquila.5 Aquila produced his translation of the Hebrew Scriptures at the beginning of the second century CE, following a systematic word-for-word translation principle that was characterized by many stereotyped renderings. Aquila’s work has been characterized as slavishly literal, and even “un-Greek” (Thackeray 1909: 5). This approach extended to the creation of neologisms, emerging from the desire to translate Hebrew words from the same root using Greek words from the same root. While Aquila produced a new translation, another Jewish textual scholar, Theodotion was at work, likely as a reviser of the Old Greek text. He lived in the middle of the second century CE, and is connected with revisional activities in patristic sources.6 The Theodotionic revisional approach can also be described as having a tendency to modify the Greek text toward the Hebrew (proto-)MT. Consequently, there are several similarities between Theodotion and the Kaige revision, which is why some scholars have started using the term “Kaige-Theodotion.” However, this assimilation of terms is potentially misleading and may be unfounded. It is probably best to regard Kaige and Theodotion as belonging to distinct efforts at a similar approach to translation (Law 2008: 7–8). Theodotion’s revision can be situated in an intermediary position between Old Greek and Aquila: while Aquila and Theodotion often agree in revising the Greek text toward the Hebrew, generally Theodotion agrees more often with the Old Greek. An illuminating example of these revisional distinctions can be found in Joshua 5: All translations of the Septuagint are from NETS. However, I have given my own translation for variant readings and other witnesses deviating from the base text of NETS. Translations from Hebrew are my own. 5 The real historical figures behind “the Three” remain obscure and debated. However, it is not necessary to identify the personal identities of the translators in order to use this material in Septuagint studies. Gentry (2016b) offers helpful insights into the possible identities of these figures. 6 Irenaeus, for instance, in his Adversus haereses accused Theodotion of misinterpreting Isa. 7:14 by reading “young woman” instead of “virgin.” See Harvey (1857: 110). 4

THE SEPTUAGINT AND THE MAJOR RECENSIONS 165

Josh. 5:12b MT

‫ויאכלו מתבואת ארץ כנען‬ And they ate from the harvest of the land of Canaan

OG

ἐκαρπίσαντο δὲ τὴν χώραν τῶν Φοινίκων And they enjoyed the fruit of the country of the Phoenicians

Aquila

καὶ ἔφαγον ἀπὸ γενήματος τῆς γῆς χαναάν And they ate from the harvest of the land of Canaan

Theodotion

καὶ ἔφαγον ἀπὸ τῶν καρπῶν τῆς γῆς χαναάν And they ate from the fruit of the land of Canaan

The Old Greek contains the most dynamic translation of the Hebrew sentence. It assimilates the Hebrew words “to eat” and “harvest” by employing a single Greek verb καρπίζομαι (“to enjoy the fruits of”). The Old Greek also translates ‫“( ארץ‬land”) with the free equivalent χώρα (“country”), and interprets ‫“( כנען‬Canaan”) with Φοινίκων (“Phoenicians”). Aquila gives the most literal translation, selecting a lexical equivalent for every Hebrew word and particle, including separate words for eating and the harvest (καὶ ἔφαγον ἀπὸ γενήματος; “and they ate from the harvest”), as well as the most literal equivalents for the Hebrew “land” and “Canaan.” Theodotion, interestingly, has revised the text so that, while every Hebrew word has an exact equivalent, he seems to have been conditioned by the OG καρπίζομαι in using the Greek καρπός (“fruit”). This example nicely illustrates the similar translation approaches taken by Aquila and Theodotion, as well as their differences in terms of the degree of freedom they exhibit in relation to the Hebrew. The third early Jewish reviser followed a somewhat different translation principle when compared to his known predecessors. Symmachus is dated to the late second century CE, and is usually identified either as an Ebionite or a Samaritan who converted to Judaism. The revision by Symmachus was more stylistic, aiming at elegant and idiomatic Greek language, while at the same time accurately reflecting the sense of the Hebrew source text. On the one hand, he avoided the repetitive equivalents of the earlier revisers, while on the other hand in many cases he leaned toward word-for-word renderings more often than the OG. The text of Symmachus was particularly popular among Christians, and its style appealed to Jerome, who was greatly dependent on it when producing the Vulgate. Thus far, the most influential research on characterizing and analyzing the work of Symmachus has been published by Salvesen (1991). Beyond the linguistic features of the early Jewish revisions, some exegetical and/or theological motivations have been discerned behind their revisional activities. These pertain most notably to a concern for the portrayal of God himself. A prominent example is the tendency of Symmachus to eliminate anthropomorphic depictions of God. For instance, in the text of Gen. 1:27, as translated by Symmachus, God does not create humans “according to the image of God” (κατ’ εἰκόνα θεοῦ), but rather “in a different image” (ἐν εἰκόνι διαφόρῳ). Furthermore, Aejmelaeus (2017a: 41–53) has demonstrated that the Kaige revision was at times motivated by theological concerns related to the conception of God. The reviser, for example, apparently wanted to avoid the suggestion

166

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

that God regrets his decisions, so rather than use the standard Greek equivalent μεταμέλομαι (“to regret”) of the Hebrew word ‫נחם‬, he chose the verb παρακαλέω (“to comfort”) (e.g., 1 Sam. 15:11 and Jon. 3:9, 10 in the Naḥal Ḥever scroll). This choice was possible because “to comfort” was already one of the meanings of the Hebrew verb ‫נחם‬. In practical terms, the readings from the Three are not directly available to us. The translations of Aquila, Symmachus, and Theodotion are known only thanks to the work of the third-century CE church father Origen, who included in his famous Hexapla the texts of the Three.7 Since the Hexapla has not been well preserved, its readings must be reconstructed from different sources, most notably from marginal readings in various manuscripts, the ancient Syrian translation (Syro-Hexapla), and quotations from church fathers. Currently, besides the second textual apparatus in the finished Göttingen editions, there are at least three tools that the scholar can utilize to access Hexaplaric readings. First, the second apparatus of the so-called Larger Cambridge Edition of the Septuagint also identifies readings attributed to the Three from various sources. Second, Nigel Turner completed the work started by Joseph Reider by compiling An Index to Aquila (Reider 1966), which lists all of the equivalents used by Aquila for certain Hebrew words, together with their attestations and identifications in manuscripts. Third, the two-volume collection of Hexaplaric material by the late nineteenth-century scholar Frederick Field can still today be quite helpful (Field 1875), although it is certainly outdated and contains errors. The important work by Field will eventually be updated once The Hexapla Institute has finished producing its new critical edition of the Hexaplaric material (see www.hexapla.org).

Christian Recensions The Hexapla The production of the Hexapla in third-century CE Palestine was a remarkable achievement. Origen of Alexandria (185–254 CE) can arguably be called the first Christian scholar dedicated to a careful study of the text of the Old Testament. The vision that guided Origen in compiling his six-column edition was essentially the same that guided the earlier Jewish revisers: differences in the Greek text vis-à-vis the Hebrew text needed to be accounted for. Such work was also perceived as useful in defending Christian beliefs in the contemporary debates with Jewish believers. Being trained by the Alexandrian grammarians, Origen carried out his work through meticulous interpretation, comparison, and annotation of the extant texts. Although there are considerable uncertainties pertaining to the nature of the Hexapla, it has traditionally been assumed that this parallel-version of the Old Testament consisted of six columns: 1. The Hebrew text written with the Hebrew script 2. A transliteration of the Hebrew text with the Greek alphabet 3. The text of Aquila 4. The text of Symmachus 5. The received text of the Septuagint, annotated by Origen with Aristarchian symbols 6. The text of Theodotion

For a fuller discussion of the Hexapla, see the chapter by Gentry in the present volume.

7

THE SEPTUAGINT AND THE MAJOR RECENSIONS 167

In his commentary on the Gospel of Matthew, Origen writes that his aim was “to heal” (ἰάσασθαι) the Greek text. In practice, in the fifth column he marked with Aristarchian symbols those elements in his received Septuagint that were missing from the Hebrew using an obelos (÷), and likewise those elements in the Hebrew that were missing from the Septuagint using an asteriskos (※). To signal the end of such an element, he wrote a metobelus ( ̷. or ↙). These markings have survived to us in the Syro-Hexapla and some Greek manuscripts, however only sporadically and at times incorrectly. The result of Origen’s text-critical efforts is known as the Hexaplaric recension, which was derived from the fifth column of the Hexapla. The text of the fifth column was made popular especially due to the influence of Eusebius and Pamphilus of Caesarea. The Hexaplaric text was copied and used extensively, which is why we now possess several manuscripts of the Septuagint that are mainly Hexaplaric in nature (often labeled O in critical editions), as well as Hexaplaric influences in manuscripts that are classified to other text groups, such as Codex Vaticanus and the Lucianic manuscripts. One should note that almost everything we assume that we know about the Hexapla and Origen’s work is quite uncertain. There are many questions pertaining to it; for example, Origen’s purpose in creating the Hexapla and the extent and nature of the Hexapla are currently being debated and scrutinized. Therefore, one is advised to consult the chapter by Gentry in this volume to gain a fuller understanding of the ancient sources and modern debates related to Origen and the Hexapla. Significant quantitative differences exist between the Masoretic text and the Septuagint version of several books. Such differences were noticed already by Origen, as is attested by traces of the Hexapla in various sources. An example appears in Joshua 24, which contains additional material both in the Masoretic text and the Septuagint: Josh. 24:17

MT

OG

O

‫כי יהוה אלהינו הוא המעלה אתנו ואת אבותינו מארץ מצרים‬ ‫מבית עבדים ואשר עשה לעינינו את האתות הגדלות האלה‬ For YHWH our God, he brought us and our fathers out from the land of Egypt, from the house of slavery and performed those great miracles before our own eyes κύριος ὁ θεὸς ἡμῶν, αὐτὸς θεός ἐστιν· αὐτὸς ἀνήγαγεν ἡμᾶς καὶ τοὺς πατέρας ἡμῶν ἐξ Αἰγύπτου The Lord is our God, he is God, he brought us and our fathers up out of Egypt κύριος ὁ θεὸς ἡμῶν ÷ ἐστιν ̷. αὐτὸς ἀνήγαγεν ἡμᾶς καὶ τοὺς πατέρας ἡμῶν ἐξ Αἰγύπτου ※ ἐξ οἴκου δουλείας καὶ ὅσα ἐποίησεν ἡμῖν τὰ σημεῖα τὰ μεγάλα ταῦτα .̷ The Lord ÷ is ̷. our God, he brought us and our fathers up out of Egypt ※ from the house of slavery and did for us all those great miracles ̷.

There are two differences between the MT and the OG that sparked Hexaplaric recensional activities. First, the OG contains an additional affirmation after the mention of “Lord our God,” reading αὐτὸς θεός ἐστιν (“he is God”). According to the witness

168

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

of the Syro-Hexapla and some other manuscripts, the fifth column of the Hexapla likely read in its text only the additional verb ἐστιν, which was marked with an obelos, as attested by the Syro-Hexapla. Furthermore, one Hexaplaric manuscript (19) and a few manuscripts identified with the so-called Lucianic text in Joshua (54, 75, 314) omit the sentence altogether, likely due to later copying of the Hexaplaric text, during which the passages marked with an obelos were eventually left out altogether. Second, the Masoretic text contains a long addition describing Egypt as a house of slavery and referring to the miracles performed by God in Egypt. This sentence, missing from the OG, which likely represents an earlier textual form of the verse, was added to the Greek text by Origen and marked with an asteriskos (※). The Syro-Hexapla has preserved this marking. Manuscripts 85 and 344 are witnesses to an intermediate phase in the importation of the Hexaplaric material into the running text, with the phrase added to these manuscripts as a marginal reading. Furthermore, a considerable amount of manuscripts read the whole phrase as part of their main text due to the later copying of the Hexaplaric tradition. As illustrated by the above example, the Hexaplaric material in manuscripts can be spotted relatively easily in textual criticism of the Septuagint. In the best cases, this material is identified with the help of external signs: Aristarchian symbols or marginal readings in specific manuscripts. Moreover, the tools mentioned in conjunction with the Three, especially Field’s edition and the future edition by The Hexapla Institute, greatly help in reconstructing the Hexaplaric material. If no external signs are present in the manuscript material, the agreement of a variant Greek reading with the Masoretic text is enough to raise suspicion. However, since we now know that such revisional activity was undertaken even before Origen, identifying Hebraizing readings with Origen should not be automatic. Again, Origen was not the first person to revise the Greek toward the Hebrew. An overall text-critical assessment of all the textual witnesses pertaining to a specific passage should always form the basis of such conclusions. The Lucianic Recension The last major recension to be discussed is perhaps the recension whose nature, extent, and text-critical significance are currently most debated in Septuagint research. The city of Antioch in Syria was one of the major centers of Christianity during Roman times, well known for its vibrant tradition of biblical exegesis. Early sources describe Antioch as having its own textual tradition of the Greek Bible. For instance, in his preface to the translation of Chronicles, Jerome lists three recensions, among which one is connected to the Christian presbyter Lucian of Antioch (c. 240–312 CE).8 Thus, the second major Christian recension of the Greek Old Testament is now usually called the Lucianic text, following the association of this textual tradition with Lucian himself. This textual tradition is found in manuscripts that are usually marked in critical editions with the siglum L. However, these manuscripts are different in various books, and one should therefore pay close attention to the work done pertaining to specific books. For example, while manuscripts 19 and 108 belong to the L group in Samuel-Kings, in Joshua-Judges these manuscripts are designated as O (Hexaplaric). This once again highlights the

Constantinopolis usque Antiochiam Luciani martyris exemplaria probat (“In Constantinople and as far as Antioch copies by the martyr Lucian are commended”); Würthwein (1995: 60).

8

THE SEPTUAGINT AND THE MAJOR RECENSIONS 169

eclectic nature of manuscripts. Notably, some scholars prefer the term “Antiochene” (or “Antiochian”) text for this group, which emphasizes less the assumed reviser and more the geographical locale.9 This term is appropriate, since the earliest church fathers who cite the Septuagint according to this textual tradition, most notably John Chrysostom and Theodoret of Cyrrhus, were indeed Antiochene. It can now be considered scholarly consensus that the Lucianic or Antiochene manuscripts contain a text with two phases: the recensional layer by Lucian, and the proto-Lucianic text that he received. The later recensional layer most probably derives from Lucian, from about 300 CE (Kauhanen 2012: 13–14). However, the earlier history of this textual form and its significance for the textual criticism of the Septuagint is a matter of current debate, as will be illustrated below. The identification of the features characteristic of the Lucianic recension proper, that is, the later layer of the L manuscripts, is an important question that has been taken up by many scholars. While there is no single and systematic revisional principle behind the recension that would apply to all texts (Würthwein 1995: 60), some features typical for the reviser include (Fernández Marcos 2000b: 230): 1) Grammatical and stylistic changes 2) Interpolations. E.g., articles, proper names, making implicit subjects and assumptions explicit 3) Harmonization with parallel texts and near context 4) Revision toward the proto-MT text derived from Hexaplaric material Some examples of these phenomena are in order. In the Septuagint, the Hebrew beginning of a sentence ‫“( ויהי‬and it happened/was”) is often translated with καὶ ἐγενήθη (“and it came about”), which utilizes the passive form of the verb. An example of the first feature of the Lucianic revision is the somewhat systematic substitution of the middle verbal form ἐγένετο, which is more in line with Attic style. Such grammatical and stylistic changes are not systematic and are affected by the context of each passage. For example, while the reviser often adds articles to nouns in line with a better Greek style, in some passages where one would expect such additions they do not occur. A good example of a passage where such additions do occur is 2 Sam. 15:2, where the desirable answer to Absalom’s call is Ἐκ μιᾶς φυλῶν Ἰσραήλ ὁ δοῦλός σου (“your servant is from one of the tribes of Israel”), which the Lucianic reviser complemented with the requisite definite articles: Ἐκ μιᾶς τῶν φυλῶν του Ἰσραήλ ὁ δοῦλός σου.10 Perhaps the most easily noticeable characteristic of the recension by Lucian is the frequent modification of the Greek text toward the Hebrew proto-Masoretic text. It should, however, be noted that Lucian probably did not know the Hebrew text as such, and these Hebraizing modifications were taken from the Hexaplaric textual tradition (Fernández Marcos 2016: 230–1). An example of such Hexaplaric influence in the Lucianic text can be found in the transposition in 2 Samuel 5:

A third option was used by Max Margolis in his critical edition of Joshua, that is, the designation S (“Syrian”). Before the Göttingen editions for these books are finished, the Lucianic readings can be found from the apparatus of the Larger Cambridge Edition (Brooke and McLean 1927) in which the L manuscripts are marked with the sigla boc2e2. An edition of the Lucianic text of the books of Samuel has also been published by Fernández Marcos and Busto Sáiz (1989).

9

10

170

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

2 Sam. 5:5a MT

‫בחברון מלך על יהודה שבע שנים וששה חדשים‬ At Hebron he reigned over Judah for seven years and six months …

OG

ἑπτὰ ἔτη καὶ ἓξ μῆνας ἐβασίλευσεν ἐν Χεβρών ἐπὶ τὸν Ἰούδαν Seven years and six months he reigned at Chebron over Iouda …

L

καὶ ἐβασίλευσεν ἐν Χεβρών ἐπὶ Ἰούδαν ἔτη ἑπτὰ καὶ μῆνας ἓξ And he reigned at Chebron over Iouda for seven years and six months …

O

ἐν Χεβρών καὶ ἐπὶ τὸν Ἰούδαν ἐβασίλευσεν ἑπτὰ ἔτη καὶ μῆνας ἓξ At Chebron over Iouda he reigned for seven years and six months …

When counting the reign of David, the OG reads first the regnal time (“seven years and six months”), contrary to the MT which places it at the end. The Lucianic text contains a transposed version of the sentence, reading the regnal time last as in the MT. Interestingly, although following the Hexaplaric transposition toward the MT with the regnal time placed at the end, the Lucianic text does not render the text as literally as the Hexaplaric manuscripts, which also swap the positions of the place-names and the regnal time but retain the verb “to reign” in the middle. Thus, it seems that the Lucianic text has received the transposition of the regnal time toward the MT from the Hexaplaric tradition, but retained the verb “to reign” at the beginning for stylistic reasons. Later in the verse, the Lucianic reviser further employed his unifying touch by changing τριάκοντα τρία ἔτη (“thirty-three years”) to τριάκοντα και δύο ἔτη καὶ μῆνας ἓξ (“thirty-two years and six months”), so that the exact sum of regnal years mentioned in 2 Sam. 5:5 would fit the total of forty mentioned in 2 Sam. 5:4. Finally, to return to the question of the earlier, proto-Lucianic phase of the Lucianic recension, scholarly opinion is still divided. More specifically, scholars disagree over the character of the base text that was later revised by Lucian, and how it relates to other Septuagint witnesses. Julius Wellhausen proposed long ago that some distinctive variants in Lucianic manuscripts must derive from a proto-Lucianic textual tradition, rather than from Lucian himself (Wellhausen 1871: 221–4). This conclusion is necessary, because many of these variants are supported by witnesses that are much older than the 300 CE Lucianic recension itself. These older witnesses most notably include the Old Latin (OL) secondary version of the Septuagint (Vetus Latina) and the biblical text that was likely used by Josephus, Hippolytus, Irenaeus, Tertullian, and Cyprian (Kauhanen 2012: 13–14). Some scholars have even suggested that there was a full-fledged protoLucianic recension (or series of revisions) that, according to them, is revealed by shared readings between the L manuscripts and Qumran scrolls, most notably 4QSama (Cross 1964; Ulrich 1978: 258–9). This theoretical proto-Lucianic recension would have been the result of revising the OG toward an existent non-Masoretic Hebrew text, as partly attested by Qumran. However, the assumption of a proto-Lucianic recension has come under severe criticism, and most have abandoned it, as the evidence provided to back it up is quite tenuous (Kauhanen 2012: 20–3). Nevertheless, the question of the nature of the actual proto-Lucianic text remains important, and is completely separate from the assumption of a proto-Lucianic recension.

THE SEPTUAGINT AND THE MAJOR RECENSIONS 171

Readings preserved in the proto-Lucianic text are especially helpful for textual criticism of those books in the Septuagint where the most important manuscripts are affected by the Kaige revision. Especially in the case of Judges and the βγ and γδ sections of Samuel-Kings, the OG text cannot be reconstructed on the basis of the B text and related manuscripts. Fortunately, at least in the case of these books, after the OG text was translated the textual tradition split in two directions: while the proto-B text was transmitted in Palestine and was affected by the Kaige revision, the proto-Lucianic text was likely copied in a different geographical location where it was not affected by Kaige (Brock 1996: 306). Therefore, Lucianic manuscripts often preserve OG readings in passages where all the other manuscripts have lost them due to Kaige influence. However, before one can identify proto-Lucianic readings that represent OG, one must make sure that these distinct readings are not derived from three other phenomena that affected the L manuscripts: 1) the later Lucianic recension itself, or its own inner-Greek developments; 2) the Hexaplaric readings that were incorporated into these manuscripts; or 3) typical scribal mistakes or possibly free copying that may have taken place in the earlier transmission of the proto-Lucianic text. When the Old Latin witnesses agree with the distinct readings in the L witnesses, it strengthens the argument that the reading could go back to the Old Greek, since these witnesses have origins independent of the later recensions. One simple example of this particularly complex text-critical issue may be illuminating. As mentioned before, the B text of Judges is thoroughly affected by the Kaige revision, while the text of Codex Alexandrinus (A text) and related manuscripts is a text closer to the OG, albeit also infused with Hexaplaric material. Rahlfs avoided the problem of reconciling the two differing textual traditions by printing both the B and A texts, one on top of the other, in his edition. Besides the A text, a few Lucianic manuscripts (54, 59, 75, 314) of Judges contain a pre-Hexaplaric text untouched by Kaige, which is why the OG text is often sought there. In Judges, this group is called AII (Satterthwaite 2015: 103). The following example is from Judges 6, which is situated after Gideon has been called to help the Israelites and has destroyed the altar of Baal: Judg. 6:32a MT

‫ויקרא לו ביום ההוא ירבעל לאמר ירב בו הבעל‬ And he was called on that day Jerubbaal, saying: “Let Baal contend against him”

B text

καὶ ἐκάλεσεν αὐτὸ ἐν τῇ ἡμέρᾳ ἐκείνῃ Ιαρβααλ λέγων δικασάσθω ἐν αὐτῷ ὁ Βααλ And he called him on that day Jerubbaal, saying: “Let Baal contend against him”

L & A text

καὶ ἐκάλεσεν αὐτὸ ἐν τῇ ἡμέρᾳ ἐκείνῃ Δικαστήριον τοῦ Βααλ And he called it on that day Baal’s Court of Justice

OL

Et vocavit illud in illa die iudicium Baal And he called it on that day Baal’s Court of Justice

172

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

The text-critical principle set forth by Paul de Lagarde states that the reading that corresponds least with the MT is likely earlier (see, e.g., Aejmelaeus 2017c: 162–163). Moreover, a freer rendering is usually earlier than a slavishly literal one (Kauhanen 2012: 24). In Judg. 6:32 it is easy to notice that the Greek translation in the B text corresponds perfectly with the Masoretic text. The version shared by the Masoretic and B texts is longer, and contains the name Jerubbaal together with a brief etiological explanation for the name. One possible interpretation for this version is that the name Jerubbaal was given to Gideon on account of his deeds. Based on our survey concerning the Kaige tradition, the exact Greek rendering of the Hebrew in the B text is enough to raise suspicion of belonging to the Kaige tradition. Indeed, the Lucianic tradition together with the Old Latin witness Codex Lugdunensis contains a tradition further away from the MT. Lugdunensis is also corroborated by the A text. In this textual tradition, Gideon is not renamed, but apparently the opposition to destroying the altar of Baal is given the name Δικαστήριον τοῦ Βααλ (“Baal’s Court of Justice”). The reading does not carry any typical features of the Lucianic reviser, and since it is further away from the MT it cannot be Hexaplaric. This reading also has the support of the Old Latin witness.11 Therefore, we have an excellent example of a proto-Lucianic reading that is the most probable representative of the Old Greek text in Judg. 6:32a (Trebolle Barrera 2005: 405). Lastly, to make things more complicated, one should note that the text-critical value of the Lucianic manuscript in relation to the B text in the historical books is a very debated issue. A case in point is the debate between Law and Kauhanen (2010) and Kreuzer (2010) in the Bulletin of the International Organization for Septuagint and Cognate Studies. Whereas Kreuzer prefers a more general notion that the Lucianic text indeed almost always represents the OG, Law and Kauhanen argue that this should be evaluated on a case-by-case basis. Sometimes the OG is even found outside of the L or the B text (Law and Kauhanen 2010: 78). This conversation has long roots in the history of Septuagint studies. In sum, research on the Lucianic recension is especially challenging, since all the Lucianic manuscripts contain at least two layers: the proto-Lucianic text and the Lucianic recension. Furthermore, these manuscripts are affected by Hexaplaric influences and the typical changes that take place in the copying of texts by hand. Any critical study of the Lucianic recension must also take into account the text-critical problems involved especially in the historical books, as well as information about the translation technique of the Old Greek translator, in contrast to the later revisers. Therefore, this discussion also illuminates how in Septuagint research all questions are connected, and how conclusions in one area of research may have major effects on another area.

Ongoing Research Questions The study of recensions is arguably a fundamental aspect of Septuagint research, where open questions, debates, and contrasting schools of thought will likely persist far into the future. Although much agreement has been achieved and meticulous research has been carried out, one could argue that this area of Septuagint research is still in a state of flux and development. At least five current challenges could be highlighted. First,

Although the Codex Lugdunensis also contains the B reading as a doublet.

11

THE SEPTUAGINT AND THE MAJOR RECENSIONS 173

the critical Göttingen editions, which would give us a good critical approximation of both the Old Greek text and the recensions, are still yet to be completed for the most textually complex books in the Septuagint. As a result, many continue to rely solely on the edition by Alfred Rahlfs which may result in bypassing important manuscript data or working with a faulty approximation of the Old Greek text. The publication of critical editions, furthermore, is only a first step, which should be followed by critical scholarly evaluation and re-editing. Second, the ancient recensions are known only through fragmentary evidence, and this evidence has not yet been compiled into comprehensive and critical up-to-date editions. Third, many of the rarer secondary translations, which serve as witnesses to the plurality of the Greek text and its recensions—such as the Old Latin, as well as the Georgian and the Armenian Bibles—are only starting to be researched critically.12 Fourth, many historical details pertaining to the revisers themselves are blurry. This is where interdisciplinary collaboration with other historical fields of study will be especially necessary as research progresses. Fifth, there is much disagreement as to what should be emphasized when reconstructing the Old Greek text that underlies the recensions. For example, to what degree can the Old Latin witnesses be used in reconstructing Old Greek or even Hebrew readings? Furthermore, what should we do when the Old Greek reading seems to be buried under recensions in every known manuscript? This fifth challenge in particular is where very much disagreement among scholars still persist. All scholarly work on these and other challenges will have implications for biblical studies at large. For example, if the earliest text of the Hebrew Bible is found only after careful text-critical analysis of the Septuagint, then other methods such as literary, redaction, and form criticism would do well to pay heed to the findings in this field. This may even have implications for historiography, which requires that the biblical texts are used critically. The study of the recensions will also have implications for the critical study of the New Testament and, for example, a correct understanding of the quotations of the Hebrew Scriptures in the New Testament. One should be aware of the whole recensional plurality of the Septuagint in order to make secure conclusions about how the writers of the New Testament handled their source text (Kujanpää 2018: 4–8). Furthermore, the study of recensions should be both informed by and taken into account in the wider study of the interpretation of scriptural texts in early Jewish and Christian thought. These challenges and implications—and there are certainly even more than those mentioned here—are hopefully enough to attract plenty of future research and new researchers into the exciting and rewarding field of Septuagint studies.

RECOMMENDED RESOURCES 1. Fernándz Marcos (2000b). A comprehensive discussion of the transmission and revision of the Septuagint by early Jewish and Christian revisers. Recommended reading if one wishes to deepen one’s knowledge about the issues discussed in this chapter. 2. Aitken (2015b). Since questions pertaining to recensions vary across different Septuagint books, the reader is advised to check out articles on specific books and find more references in this excellent volume.

On these important witnesses, see the chapter by Cox in the present volume.

12

174

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

3. Gentry (2016b). A helpful introduction to research on the early Jewish and Christian recensions, containing many helpful technical details and quotations of primary sources. 4. Aejmelaeus and Kauhanen (2017). A collection of articles illuminating the current research situation and debates on the Kaige revision and the Lucianic text in the historical books of the Old Testament. 5. McLay (1998). This is a helpful tool to be used in the text-critical analysis of Greek variants.

CHAPTER 12

The Septuagint and the Secondary Versions CLAUDE COX

INTRODUCTION For most of us, our first encounter with the “secondary versions”—sometimes known as “daughter” or “sub-versions”—comes when we use an edition of the Greek text that provides the sources for its study and establishment, such as Brooke-McLean or the Göttingen eclectic editions of the Septuagint. In the latter, the evidence for variants appears in an order that begins with the papyri, then uncials, text groups identified by sigla, individual Greek manuscripts whose textual affiliation is mixed, patristic evidence, and then the secondary versions—beginning with the Old Latin and continuing in alphabetical order. The experience can be overwhelming. For example, the first time a string of secondary versions appears in Wevers’ edition of Genesis (1974a) is at 1:14: τοῦ 2°] και A M C´–16 78 129 75 46s y–121 527 31 55 508 509 730 Ath II 205 La Aeth Arm Bo Pal These sigla are specific to the critical edition and represent the Old Latin (La), Ethiopic (Aeth), Armenian (Arm); Bohairic (Bo), one of the Coptic dialects; and Palestinian Syriac (Pal). Missing from this string is the Syro-Hexapla (Syh), a primary witness to Origen’s Hexapla, as the siglum suggests. It is extant only fragmentarily in Genesis. The Georgian version has not been collated in Göttingen editions thus far since there has been a lack of editions of texts, a consequent inadequate understanding of its textual history, and a dearth of specialists with training in Septuagint studies and in the Georgian Bible, but those realities are changing for the better. How did all these various versions come into being and what role do they play in the textual criticism of the Septuagint? To these questions we now turn.

OVERVIEW OF PRODUCTION AND TRANSMISSION OF THE SECONDARY VERSIONS By the second century, Jews and Christians realized that their sacred texts of the earlier Scriptures diverged in particulars, as evidenced in the argumentation of Justin’s Dialogue with Trypho. Jewish scholars had returned to the Hebrew text. Christian teachers, not knowing Hebrew, depended upon and defended the Septuagint, itself a work of Hellenistic Judaism. Of course, both textual traditions had experienced corruptions, the

176

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

former less than the latter because there a definite and rigorous scribal tradition was in play to preserve and protect the received Hebrew text. It was not until Origen that a serious attempt was made to compare the texts of the two traditions, Hebrew and Greek (Swete 1900: 483). If the Hexapla project was fraught with difficulties from its inception, high marks must be given to Origen for conceiving such an ambitious undertaking: a multi-columned book consisting of the Hebrew, the Hebrew in Greek transliteration, the translation of Aquila, the translation of Symmachus, Origen’s Septuagint text and, finally, the translation bearing the name of Theodotion. Six columns, hence “Hexapla”—the text presented word by word or phrase by phrase down its column, the five columns to the right of the Hebrew calibrated according to the first. At a glance a person could see how the Hebrew and its Greek translations related to one another. Words like “bold” and “brilliant” easily come to mind. This is the traditional portrait of the Hexapla, but today there are important suggestions that see Origen’s work first as a Tetrapla, that is, four columns, but the principle of comparison remains.1 At the end of the next century, the fourth, Jerome, in another bold endeavor, anticipated the scholarship of twelve hundred years later when he based a new Latin translation of the earlier Scriptures not upon the Septuagint, as previous translations had done, but upon the Hebrew. However, his work was not popular and did not immediately displace all the various Old Latin translations made before Jerome (see Trebolle Barrera 2016: 319–23). No one-volume copies of Jerome’s work appeared until much later and neither his work nor other Latin translations remained static. Like other text traditions at the time, and particularly so long as copies of the parts of the Bible were manuscripts, that is, hand-copied, changes and corruptions inevitably occurred in them all. Christians appropriated the Old Greek corpus. To this they attached their own writings, which eventually came to be called the New Testament. It was these two collections, expanded by the so-called Apocrypha, more or less, that constituted the Christian scriptures. They were all in Greek, a language that Alexander the Great had spread right across the ancient world, from Europe to India. This common language proved to be of inestimable help in the spread of the Christian church. And it was not long before Christian missionaries and scholars were translating the Bible from Greek into the languages of those among whom the Christian message took root: aside from Latin, into Syriac, Coptic, Ethiopic, Armenian, Georgian, and Arabic. This is how those secondary versions cited in Wevers’ apparatus (above) came into being. It almost goes without saying that the translation into these various languages initiated a new textual tradition in each instance. In most cases, details about the precise origin of the translation remains unknown, that is, where, when, by whom, and on the basis of what kind of source texts? As with the case of the Septuagint, these “daughter versions” were sometimes subject to revisions intended to improve the text by resorting to comparison with some other source text. In the case of the Armenian, there is, uniquely, a historical tradition that provides a date for the initial translation (c. 406 CE). It also explains that this initial work (from Syriac? Greek?) was “established” (revised?) on the basis of “accurate copies” brought back to Armenia after the Council of Ephesus in 431 CE

On the Hexapla, see Gentry in this volume.

1

THE SEPTUAGINT AND THE SECONDARY VERSIONS 177

(Koriwn 1941: 75–7). This information is immensely helpful, but many questions remain, including what the words mean that are enclosed within quotation marks!

Polyglots The Hexapla seems never to have been recopied in its entirety, though its fifth column, with the Old Greek text mechanically supplemented so that it was equal to the Hebrew in length, was popular to such an extent that it corrupted most of the textual tradition. It is presumed to have perished at Caesarea when Islamic armies overran Palestine in the early seventh century. But the idea of a multi-columned, or at least multi-translation, Bible reappeared with the emergence of polyglot Bibles in the sixteenth century. Setting forth as they did the possibility of comparing the text of the Hebrew Bible with later translations, these editions were harbingers of subsequent, more sophisticated attempts at setting forth the evidence of source texts (Hebrew; Greek) and these texts as translated into other languages, for example, Latin. The polyglots are one of the fruits of the Renaissance in so far as it led Europeans to try to recover the roots of western civilization. For biblical studies this produced a keen interest in the text of the Bible as it existed in various ancient translations. The first of the polyglot Bibles was the Complutensian Polyglot, a work of Spanish scholarship, 1502–17, and published in 1520. It sets forth in three columns primary texts of the Old Testament, Hebrew (with Aramaic Targums for the Pentateuch at the foot of the page), Latin (Vulgate), and Septuagint with an interlinear Latin translation. The text of the New Testament was presented in Greek and Latin. Other polyglots followed, published in Antwerp, Heidelberg, Hamburg, and Paris. The London Polyglot (1654–7), the last of the great polyglots, improved upon its predecessors. The texts are arranged side by side or one above the other, as many as nine across two facing pages: Hebrew, Samaritan Pentateuch, Septuagint, Latin Vulgate, Syriac, Arabic, Ethiopic (Psalter), and Persian (Gospels) (Jellicoe 1968: 350–2; Swete 1900: 171–85; Nestle 1903: 440–1; Reilly 1911). None of these texts or translations was critically established, but the principle of comparison was explicitly adhered to, and the value of the versions or secondary versions clearly promoted. A page of the Complutensian Polyglot is reproduced in Würthwein (2014: 300, Plate 47). A straight line can be drawn from these sixteenth- and seventeenthcentury attempts to provide the tools for comparative textual studies, with a view to recovering original readings, and the diplomatic and critical editions of the Bible represented by the larger Cambridge Septuagint, the Göttingen editions of the Old Greek, as well as the various editions of the Nestle NT and the United Bible Societies’ editions that overlap with Nestle.

From Multi-Columns to Text and Critical Apparatus Holmes-Parsons’ edition of the Old Greek (1798–1827) followed upon the polyglots. In this massive undertaking that saw some 297 Greek manuscripts collated, there was an attempt to present a text with a critical apparatus. The base manuscript for collation is that of the Paris Polyglot, that is, Vaticanus (fourth century), with variant readings cited in a large apparatus below the text. Notably, the secondary versions are well represented: Old Latin versions, Coptic (Bohairic and Sahidic), Arabic, Slavonic, Armenian, and Georgian, “obtained partly from MSS., partly from printed texts” (Swete 1900: 186). For a photo of a page of Holmes-Parsons, see Jobes and Silva (2015: 67).

178

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

At the end of the nineteenth century, Cambridge University Press announced a major edition of the Old Greek, “intended to provide material for a critical determination of the text” (Swete 1900: 189). This so-called larger Cambridge Septuagint began to appear in 1906, with the edition of Genesis (Brooke and McLean 1906). The last volume to see the light of day was Esther, Judith, Tobit in 1940, leaving the series incomplete. Elegant in appearance, this diplomatic edition prints Vaticanus as its base text. Once again the versions are represented by the Armenian (A), Bohairic (B), Coptic (C, i.e., Sahidic), Ethiopic (E), Old Latin (L), Palestinian Aramaic (P), and Syro-Hexapla (S). Jobes and Silva reproduce a page of Genesis, with explanatory notes on the facing page (2015: 148). The advances over Holmes-Parsons include greater accuracy, an attempt to group manuscripts, and the addition of other witnesses, including readings from “the Three” (i.e., Aquila, Symmachus, and Theodotion) placed in a second apparatus (Swete 1900: 190). The Armenian version was collated from the edition of Zohrapian ([1805] 1984). Nestle regarded the larger Cambridge Septuagint as “but a step towards the ideal of a truly critical edition. For the text is that of a single MS with all its faults” (1903: 441). The Göttingen edition presents a critically established eclectic text of the Old Greek, the kind of edition that Nestle envisaged in 1903. The resources necessary for establishing the original text are all there in the apparatus: the Greek papyri and manuscripts, arranged in groups; quotations of the text in Jewish and Christian authors and commentators; and the secondary versions. Not all secondary versions appear in earlier volumes (e.g., the Armenian was not collated for Isaiah [Ziegler 1939]), but for Genesis (Wevers 1974a) they are well represented: Old Latin, Ethiopic, Arabic, Armenian, Coptic (Achmimic, Bohairic, Fayumic, and Sahidic dialects), Palestinian Syriac, and Syro-Hexapla. The various secondary versions were collated by the respective editors of the Göttingen volumes, a truly daunting task. Zohrapian’s edition ([1805] 1984) represents the Armenian, as in the larger Cambridge Septuagint.

The Demise of the Secondary Versions The importance of the secondary versions of the Septuagint was dramatically reduced by two developments, namely, by archeological finds that offered texts in the original languages of the biblical texts (Hebrew, Greek, Aramaic) and by advances in the field of textual criticism. In the nineteenth century the Egyptian sands began to give up papyrological treasures that offered Greek biblical texts hundreds of years older than medieval Greek minuscules or even the earliest of the great uncials, Vaticanus (fourth century). Names like Oxyrhynchus, Chester Beatty, Petrie, Bodmer, Zenon, Grenfell and Hunt, Deissmann, and the names of individual papyri designated by number—for example, P13 or P46— became familiar to everyone involved in the text criticism of the Bible. To cite a strikingly important example, MSS 847 and 848 (both part of P.Fouad. 266) preserve a sizeable part of the last half of the book of Deuteronomy (Wevers 2006: 14). P.Fouad. 266 dates from the middle of the first century BCE! Interest rightfully shifted away from the secondary versions to these early resources, of which there were increasingly many. Such papyri not only predated the earliest great uncials, but also the earliest of the secondary versions, including the Old Latin. Furthermore, discovery of clay tablets in vast numbers in Mesopotamia, for example, at Nineveh, and at Ugarit (1929) provided lexicographers and researchers with languages and literatures cognate to Hebrew and the Scriptures in Hebrew. A vast amount of

THE SEPTUAGINT AND THE SECONDARY VERSIONS 179

material of every kind was found, as can be determined by a glance at the size of the third edition of Ancient Near Eastern Texts relating to the Old Testament (Pritchard 1969), 710 large pages! If all these resources were significant, they were more than equaled by the Dead Sea Scrolls (1947): previously the earliest Hebrew texts of the Bible were medieval, but now there were original manuscripts dating from the second century BCE. If the Septuagint corpus has its greatest importance as a resource for “the textual criticism of the Hebrew Bible” (Nestle 1903: 449)2 what happens when manuscripts in Hebrew now predate witnesses to the Old Greek, often by hundreds of years? And what then of the secondary versions? Developments in the science of textual criticism also served to relegate the secondary versions to a lower status. In the days of the polyglot Bibles, the secondary versions were regarded as significant textual witnesses. The lack of critical editions was recognized as a problem by Nestle more than a hundred years ago. But even when the text critic has the advantage of a critical edition of a versional text, there is an awareness that the secondary version has its own textual history, which might be quite complicated in its own right. Furthermore, a translation is a translation: no two languages are identical and there may be severe limitations placed upon the use of a secondary version for this reason. Epp (1992: 428) provides a thumbnail summary of some characteristics of the versions and more extensive remarks on the use of several secondary versions for the textual criticism of the Septuagint can be found in Fernández Marcos (1985: 346–62), namely, the Armenian, Coptic, and Old Latin, and more extensively in Aejmelaeus (2000), albeit limited to the Psalter. Two more considerations may be added to these developments. Firstly, when all the work has been done to produce a critical edition of a secondary version of a biblical text, that text, still a translation, represents only one Greek manuscript. For example, the critical edition of Armenian Job (Cox 2006) is based on an analysis of more than 130 manuscripts, but the original text, represented in the critically established text, was translated from only one Greek manuscript. “The Armenian version” or “the Sahidic” may sound weightier, more significant, than, say, “MS 341,” but in reality the evidence of a secondary version has no more weight than one manuscript of the Septuagint. Is all this work worth it? Secondly, there is an increasing awareness that the collation of versional evidence must be more than mechanical. If the evidence is collated somewhat mechanically, as is the case in some editions of the Greek, then it requires a specialist to determine its worth. The day has passed when one editor could be expected to be proficient in more than a handful of languages, because each language has its own way of doing things. It requires years of acquaintance with and study of a language to be able of handle it proficiently. All this may seem discouraging to the person who comes upon the string of versional evidence in Gen. 1:14 cited at the beginning of this contribution. Does this evidence have any importance? If so, how does one assess its importance among the other textual evidence that is available? Is it going to be worthwhile to learn one or more of these languages and to become proficient enough to use a secondary version in one or more of those languages? We now turn to these questions.

Nestle does not say that the textual criticism of the Hebrew Bible is the most important purpose of the Septuagint, but his remark is easily read that way. His precise words are, “But, even before these minutiae (i.e., the lack of critical editions of versions, quotations in the Greek Fathers) be settled, G can and must be used for that purpose for which it is of the greatest importance, namely the textual criticism of the Hebrew Bible.”

2

180

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

The Revival of the Secondary Versions in Recent Septuagint Research It is clear that there will not be a return to the time when the secondary versions had an exaggerated importance for the textual criticism of the Septuagint. That said, the secondary versions retain a role in the textual criticism of the Septuagint because: 1) they represent additional witnesses; 2) their textual witness may be a relatively early one; and 3) they provide information about the development of the text history of the Septuagint. Each of these points is examined below. In order for a secondary version to realize this potential it is essential that there be critical editions of the texts. Only then can a particular version be properly understood and its potential assessed. The translations reflect different translation techniques, perhaps even between books of the same secondary version, and each has gone through a complex textual history. As discussed above, the text critic using a secondary version is really bringing to bear the witness of one manuscript, namely, the Greek source text of that translation of a particular book. Just as one manuscript can change its textual character from one book to another because the manuscript as a whole represents a collection of books that derive from various exemplars and scribes, so too the witness of a secondary version can change from book to book. The novitiate may be put off by the complexity of all this, but then the textual criticism of the Septuagint is nothing if not complex. The Secondary Versions Represent Additional Witnesses The evidence that the text critic brings to bear in the establishment and study of the text of the Septuagint is threefold: the Greek manuscripts, patristic citations, and the secondary versions. It should go without saying that the Hebrew (and Aramaic) source, to the extent that is extant (e.g., the MT) or recoverable, is uniquely vital. As Orlinsky once wrote (1962: 121), it is necessary to examine every available manuscript of the Septuagint or of a secondary version to determine textual affiliations. If the combined witness of all the manuscripts of a secondary version carries the weight of but one Greek manuscript, that witness is still important, perhaps significantly important if the witness is early and attests a type of text that is of special interest, as in the case of the Armenian translation of Job. No more needs to be said on this point, whose intention is simply to underline the conviction that, in matters of textual criticism, all the evidence requires investigation. The Evidence of the Secondary Versions Is Important Because It May Be Early The last century and a half has given us many earlier witnesses to the biblical text than were available before. Even so, in the case of papyri, much of this is fragmentary and does not predate the fourth century CE. The same holds true for uncial manuscripts, and most minuscule manuscripts are medieval. We can use the book of Genesis as an example. Only one uncial, Vaticanus, is fourth century or earlier and in this case the text begins only at Gen. 46:28. Manuscript G, a Leiden manuscript, is from the fourth or fifth century but attests only Gen. 31:53–36:18. Wevers has a substantial list of papyri—twenty-nine to be exact—but only fifteen of these are from the fourth century or earlier. Of particular note are 814 (c. 90 CE) and 942 (c. 50 BCE), but all are invariably fragmentary, often consisting of only a few verses (e.g., 814 attests part of 14:5-8, 12-15). The list of minuscule manuscripts stretches to sixty-seven items, but only about a dozen of these

THE SEPTUAGINT AND THE SECONDARY VERSIONS 181

Table 12.1 The Main Secondary Versions and Their Dates Old Latin Aethiopic Armenian Coptic Georgian Syro-hexapla

As early as second century Fourth to sixth centuries Early fifth century (c. 406) Third and fourth centuries Fifth to eighth century Seventh century

manuscripts are from the tenth century or earlier. It is true that late manuscripts can preserve an early form of text but, all things considered, early is better than late. Some of these minuscule manuscripts are fragmentary too. Seen in that light, the witness of the secondary versions is relatively early. Table 12.1 presents a list of the main secondary versions and their dates. These dates are derived from the respective contributions in ABD (6:794–813). In addition, readers are referred to the excellent articles on the various secondary versions in THB, the benchmark resource for the textual history of the Hebrew Bible and the textual traditions that flow from it, including the Septuagint and its secondary versions. The simple list provided above belies the complexities of the secondary version evidence: sometimes the evidence is fragmentary. The manuscripts are late and difficult to access, there is a lack of critical editions, a secondary version may evince different textual affiliations, the Georgian was not collated in the Göttingen Septuagint thus far, and the Syro-hexapla is primarily important for its preservation of hexaplaric readings and materials. Nevertheless, the textual evidence of the secondary versions is still early or relatively so if it can be retrieved, assessed for its importance, and brought to bear.3 The Secondary Versions Help Us Understand How the Source Text Was Read at the Time of Its Translation There will always be a need for specialists who can the explain the process by which original texts came into being and then were translated into other languages in the early period of their transmission. Judaism, Christianity, and Islam are text-based, book-based religions. One of the tasks of textual criticism is to understand and explain this textual development. The secondary versions are part of the story. Each had a community, perhaps still in existence, whose self-understanding is rooted in Scripture in translation. For example, the Armenian version, though a secondary version of the Septuagint, is significant in and of itself, and can be studied and commented upon within the particular context of its community. It has surprised me how often I have turned to the Armenian translation as I have tried to understand how the Old Greek translation of Job was read by the translator. Scarcely anyone is fully capable of working with all the languages and text traditions represented by the secondary versions, though editors of the Göttingen editions have been expected to collate them all until recently. Given this state of affairs, I am going to confine myself in the following discussion to the Armenian version, with which I have worked extensively, and use the book of Job as a point of reference, because there is for

The same situation obtains with respect to the versional evidence for the textual criticism of the NT: “The early versions are important witnesses for the Greek texts of the New Testament because they derive from a relatively early stage of the tradition.” See Aland et al. (2001: 23*).

3

182

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Job a critical edition of the Armenian (Cox 2006). I am going to set forth the importance of this one secondary version according to the three points made above, in the belief that the other secondary versions can be of similar benefit.

RECENT RESEARCH ON THE ARMENIAN VERSION The date and circumstances of the earliest translation of the Bible into Armenian are known, even the names of some of its translators (Koriwn 1941: 75–7; Cox 2016: 370–1). The date of c. 406 places it as early as Codex Alexandrinus (fifth century) and almost as early as Vaticanus (fourth century). Just a few years earlier would place the translation in the fourth century. In all likelihood the source text was a fourth-century manuscript. The Armenian translation is somewhat more than a century, or less than two centuries, this side of the famous Chester Beatty Papyri—mostly third century. For example, P45 (c. 200) contains fragments of the Gospels and Acts. As for Job, there are four papyri that date to the fourth century or earlier (MSS 854, 857, 955, 974) but, taken together, they preserve only a handful of verses. The presence of many hexaplaric signs in Armenian manuscripts of Job indicates that the translation is related to the Greek tradition. There are no remnants of a Syriacbased translation, if such ever existed. Armenian Job offers a striking example of what a secondary version can provide. The manuscripts are medieval—the earliest manuscripts are from the thirteenth century, but there is a fragment from Sinai that is from the eighth century that contains about twenty verses of chapters 37–38. Its text deviates little from the critically established Armenian text. With the advantage now of a critical edition of the Armenian (Cox 2006) and a critical edition of the Old Greek (Ziegler 1982), some dramatic results emerge. Armenian manuscripts and Zohrapian’s edition ([1805] 1984) preserve hexaplaric asterisks and also attest hexaplaric readings. Perhaps it was the presence of the signs of hexaplaric activity in Zohrapian’s edition that led Ziegler to place the Armenian secondary version with “O” (the Origenian or Hexaplaric group) as a representative of the hexaplaric recension. However, a re-collation of the Armenian on the basis of a critical edition of its text determines that it is an excellent witness to the type of text Lucian produced at Antioch. Not only that, the Armenian stands with Alexandrinus (fifth century) as the earliest witness to the Lucianic text in Job. The main L (= Lucianic) group in Job consists of MSS A V 575 637, Julian and Chrysostom, and now the Armenian. Venetus (eighth century) preserves the text only from 30:8. The commentaries of Julian the Arian and Chrysostom are relatively early (fourth century), but the manuscripts that preserve their work are medieval, and using ancient commentaries has its own set of challenges. MSS 575 and 637 date from the thirteenth and eleventh centuries, respectively. It is noteworthy that the Armenian shares uniquely some readings with MS 637, which means that we now know these readings in 637 existed some seven hundred years before the Greek minuscule itself was copied. For a list of readings shared between MS 637 and Arm, see Cox (2006: 398). The place of Armenian Job among Lucianic witnesses can be illustrated by citing its collation in Ziegler (1982). The collations are Ziegler’s but revised—Zeigler does not cite Arm at 1:12d, 16a, 20c (see Table 12.2).

THE SEPTUAGINT AND THE SECONDARY VERSIONS 183

Table 12.2 Ziegler’s Collations for Job 1 with Revisions 1:4b ἐποιοῦσαν B-S*] εποιουν rel 1:5c  θυσίαν B-S* 55 534 542 644 Or II 381 Iulcit Chr I 384 PsChr VI 566 Syn Amb (sacr 5, 4, 25) Hi (Ez 14) An (partim)] θυσιας rel (Test Or XVII 24 Lo. Did Cyr IX 1425 Chr III 329 VII 455 XI 204 Cyp Hi (Am 2,5: ep 73,2) IulE Cass Zeno) = MT 1:5f οὖν > L´-637-406-534 68 795 La Aeth Arm An = MT 1:5f fin] + αυτων A´-Iul 68 248c 251 Arm 1:6c  fin] + περιελων την γην και εμπεριπατησας εν αυτη (αυτην pro εν αυτη Eus) lI Co Arm Or XVII 24 Lo. Eus 1:12d παρὰ τοῦ κυρίου] απο προσωπου κ. La Syhmg L´’-534´-728 C1 (139txt)- 251 d 55 68 542 795 Arab: cf 2:7a et Gen 4:16 = MT 1:13c  init] pr (※La Syhmg) ησθιον (εσθιον 55* 637) και La Syhmg L´-Sc- 534´ 55* 68 542 Aeth Arab Arm An = MT 1:16a ἄγγελος] + προς ιωβ: cf. L´-575 68 1:17b κεφαλάς] αρχας L`-Sc 68 705 Aeth (vid) Arm 1:20b κεφαλῆς B-S*] + αυτου rel = MT 1:20c προσεκύνησεν] + τω κυριω L´’-Sc-613-728 C´ 55 68 157mg 543 795 797 Bo Dam An 1:21d  ἐγένετο B-S* La 680* Co Arab Did Cyp Orlat VI 86.234 VIII 366 An Aug (ci 1,10) Quod Cass Caes] pr και rel (Test Chr Tht Ol) 1:22a init] pr et

In these thirteen variants, Arm reads with the clear majority of witnesses, designated by “rel” (i.e., reliqui “the rest”) four times (1:4b, 5c, 20b, 21d). In eight instances it reads with Lucianic witnesses, either many of the witnesses of that tradition (1:5f 1º, 12d, 13c, 16a, 17b, 20c) or few (1:5f 2º: L, i.e., main L group witnesses; 1:6c lI, i.e., sub-group I of the L group). Once Arm uniquely adds the conjunction et (1:22a). In seven instances the Armenian, as part of the Lucianic text tradition, represents a longer text (1:5f 2º, 6c, 13c, 16a, 20b, 20c, 21d). Only once is its text shorter than the OG (1:5f 1º). In at least one case the plus is hexaplaric (1:13c); the words ησθιον και are part of the Lucianic text since Lucian employed a hexaplaric text for his recension. As this brief exploration shows, the Armenian version is one witness among the Lucianic witnesses. There are no instances in chapter one or in Job as a whole where the Armenian alone witnesses to the original Old Greek. The secondary version offers the attestation equivalent to one Greek manuscript, an individual Greek manuscript that is part of a group of manuscripts but carries the unique imprint of one copyist. The source text of Armenian Job had a corruption at 36:38, where verses 28g-k have been transposed from 37:5b-7. No other witness has this precise corruption. Surely Armenian copyists noticed this later, perhaps the translator was even aware of it, but the text was not “fixed up” by the removal of one of the parallel sets of lines. The collation of the Armenian manuscript tradition against Ziegler’s text and apparatus reveals that the Armenian translation underwent a later revision on the basis of an even fuller Greek text than L represents. Two of the four Armenian text groups—including the base manuscript for Zohrapian’s edition, Venice 1508—contain about two dozen variant readings, usually plusses, that are scattered over the first nineteen chapters of Job. Ziegler worked from Zohrapian’s edition, so these secondary readings are represented in the collation of the Armenian in his apparatus. In the critical edition of Armenian Job, such

184

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

readings are indicated by “= G” in the apparatus. For example, here is one such reading, in Armenian and in transliteration: 3:12b fin] + մոր իմոյ M2587´ Zoh´ = + μητρος μου L’`-575-406-534´ The transliteration of մոր իմոյ (“of my mother”) is mor imoy. In this citation “M2587´” refers to two Armenian manuscripts and “Zoh´” refers to the edition of Zohrapian (1984) and a manuscript that shares its text type. This plus (“of my mother”) at the end of 3:12 has been added from a yet fuller type of Lucianic text than that which served as the source text for the original Armenian translation. As is so often the case in textual corruption, the inclination is toward a fuller, more complete text as the most desirable.

The Importance of the Armenian Version for Textual Criticism and Analysis of Old Greek Job In this section the three points made in the last one are taken up with respect to the book of Job. The Armenian Version of Job Represents One Witness to the Old Greek Text It is one more textual witness, alongside Greek papyri, Greek manuscripts (uncial and minuscule), patristic evidence, and the other secondary versions such as Old Latin, Ethiopic, Coptic, Syriac (Syro-Hexapla), and Arabic (a fragmentary, ninth-century witness).4 The text critic and researcher is really interested in the Greek manuscript from which the Armenian was translated. As is true of the secondary versions generally, the recovery of that manuscript is limited by the ability of the Armenian language to represent fully and exactly the grammar and syntax of Greek. The Armenian Version Is of an Early Date If in fact Lucian’s work on the text of Job was done c. 300 CE (Jobes and Silva 2015: 47), that Lucianic text tradition was only about 100 years old when the source text of Armenian Job was brought to Armenia. The Armenian version is thus a significantly early witness to the Lucianic text recension. It belongs to the main Lucianic group (L) in Job, which otherwise consists of only four Greek manuscripts (A V [extant from 30:8] 575 637) and two patristic commentators (Julian and Chrysostom). See Cox (2006: 409). None of these witnesses attests uniquely an original reading for the text of Old Greek Job; nor do they as a text group. However, one can hardly overestimate their collective importance for the text-criticism and the text-history of the Greek translation. The Armenian Version Helps Us Understand How the Old Greek Was Understood at the Time It Was Translated Three examples are examined below in some detail, but there are others that I observed in working closely on the OG Job and Arm Job: see 9:4b; 15:14b; and 37:18a in “Notes on the Armenian Critical Text and Translation” (Cox 2006: 275–374). In addition, it is noteworthy that the translator treated ἀδικίας (“injustices”) as an accusative plural (not genitive singular) in Job 15:16b, so “a man drinking injustices like water” (NETS; compare Brenton 1844: “drinking unrighteousness as a draught”; and parsing in Accordance Bible Software version 8.4). Note that in Job the siglum for OL (Vetus Latina) is “La” and that this secondary version represents Jerome’s translation of a hexaplaric Greek text.

4

THE SEPTUAGINT AND THE SECONDARY VERSIONS 185

The first example is found at 31:16. The Old Greek and its translation in NETS are as follows: Job 31:16

ἀδύνατοι δὲ χρείαν, ἣν ποτ᾽εἶχον, οὐκ ἀπέτυχον, χήρας δὲ τὸν ὀφθαλμὸν οὐκ ἐξέτηξα. But as for the powerless, what need they had—I did not miss it, and the eye of the widow I did not let waste away.

The question involves the aorist verb ἀπέτυχον. Is it first singular—so NETS—or third plural, “they did not miss out on”? The first singular and third plural forms are identical. Taylor (2009: 58) parses ἀπέτυχον as third plural and this parsing is reflected in Brenton’s (1844) translation (“But the helpless missed not whatever need they had”), in GE (“the needy were not unheard in their necessity” [ad loc. ἀποτυγχάνω]), and LXX.D (“Und Machtlose haben nicht ermisst, was sie jemals brauchten,” that is, “And the needy have not missed whatever they needed”). Treating ἀπέτυχον as third plural like this is a possible reading but there are several reasons why a first singular is more likely. First, the underlying Hebrew is first singular though, given the Greek translator’s approach to the text, this in itself is not conclusive. However, second, typically if the translator changes the person of a verb like that in v. 16a, the change is toward continuity with what precedes, and in this case that would be the first singular (v. 15a; see v. 17b, following v. 17a). This makes a change to the third plural unlikely. Third, the parallel line (v. 16b) focuses upon Job as actor, in the first person. Finally, and decisively, Ziegler (1982) cites no variant readings for ἀπέτυχον. Significantly, the Armenian attests the first singular (… ոչ արդեաւք երբէք վճարեցի; “I never ever failed to fill”). This leads one to conclude that Ziegler believed ἀπέτυχον to be first singular, since he does not cite the Armenian (or other secondary versions—Coptic too is first singular) as attesting a reading different from the lemma. In the light of these considerations, namely, agreement with the source text, the general translation strategy, context, and early verification, a much stronger case can be made for a first singular than for a third plural in v. 16a. Job thus asserts that he did not fail to address the needs of the powerless. Second, there is no more fascinating instance of the ingenuity of the Armenian translator of Job than that found at 37:12b. It involves Theodotion’s use of a transliteration, whose translation Origen used here to fill in the shorter Old Greek text so as to bring it to the same length as the Hebrew. This conflated, ecclesiastical text is the source text for the Armenian. How is a translator to deal with such an unexpected word as θεεβουλαθωθ (theeboulathōth)? It is not a Greek word. The line of text is as follows, along with the line before it to provide context. The accompanying English translation is from NETS. The siglum ※ is an asterisk, with which Origen marked his additions from Theodotion. ※ καὶ αὐτὸς κυκλώματα διαστρέψει And he will twist round the discs ※ ἐν θεεβουλαθὼ είς ἔργα αύτῶν· by theebulatho to their works A footnote explains theeboulatho: “Heb[rew] = his guidance.” The underlying Hebrew word is ‫בתחבולתו‬, vocalized as ‫ ּבְ תַ חְ ּבּוֺלתָ ו‬in BHS. The Ketib ‫ לָתֹו‬is singular (“his guidance”) and the Qere ‫ ֺלתָ יו‬is plural (“his guidances”). The Hebrew is not our concern, only its

186

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

transliteration by Theodotion. This is Ziegler’s entry in his apparatus for θεεβουλαθώ, the reading in his critical text: ἐν θεεβουλαθώ Rahlfs = Complutensian Polyglot] εν -θων lII; εν -θωμ 339; εθετο βουλαθ S*; ενθα εβουλετο Sc1; εν θεεβουλαθ Sc2; εν θεεβουλαθωθ (θεβ. Bohairic) rel (i.e., the remainder of witnesses, including Arm). The Greek text tradition is more or less uniform: the majority text adds a θ (theta). The Armenian attests that θ. The original reading is preserved only in the manuscript(s) underlying the Complutensian Polyglot. More interesting are the corrections preserved in manuscript Sinaiticus (S): both S* and Sc1 try to make sense of this strange word by dividing it and finding a verb either at the beginning, so εθετο βουλαθ (“he placed boulath”), or at the end, so ενθα εβουλετο (“there he wished”). The second corrector returns to the transliteration. The second Lucianic sub-group (that is, lII) reads ν (nu) at the end of θεεβουλαθω, but neither this corruption nor the related one, εν –θωμ in O (= Origenic = hexaplaric) group MS 339, makes an attempt at translation. The parent manuscript of the Armenian was an uncial manuscript, hexaplaric and Lucianic, that probably looked similar to fourth- or fifth-century manuscript G, a page of which is reproduced by Würthwein (2014: 277). This manuscript is contemporary with the source text of the Armenian translation of Job. In that case what the translator saw was something like this, beginning with διαστρέψει at the end of v.12a. ΔΙΑΣΤΡΕΨΕΙ ※ ΕΝΘΕΕΒΥΛΑ ΘΩΘΕΙΣΕΡΓΑΑΥΤΩΝ ※ ΠΑΝ The Armenian translator tries to translate εν θεεβουλαθωθ, like the copyist and first corrector of Sinaiticus, another manuscript contemporary with the Greek text the Armenian translator used. The Armenian translation of v. 12ab appears as follows, with its English rendering: Job 37:12

: Եւ ինքն դարձուսցէ զչրջանակս· Ew ink‘n darjusc‘ē zšrǰanaks; And he will twist round the discs : ուր կամեցաւ՝ եդ զգործ նոցա, ur kamec‘aw ed zgorc noc‘a, where he wished, he put their work

The translation of ἐν θεεβουλαθωθ εἰς (“in theeboulathōth into”) is ուր կամեցաւ՝ եդ “where he wished, he put”). How did the Armenian translator arrive at this? First, the translator took the preposition ἐν (“in”) as referring to place, and rendered it with ուր (“where”). Next, the translator noticed that the word ΘΕΕΒΟΥΛΑΘΩΘ contains the root βουλ-, as in βούλομαι (“to wish, want, desire”), which leads to կամեցաւ (“he wished”). The final -Θ of ΘΕΕΒΟΥΛΑΘΩΘ the translator reads with the following preposition, ΕΙΣ (“into”), to produce ΘΕΙΣ, which is a participle form (θείς) of the verb τίθημι (“to put, place”). This participle the translator represents with the finite verb դնեմ (dnem) in its aorist third singular form եդ (“he put”). The result is, “Where he wished, he put their work.” This even makes some sense in the context! It can be compared with S* and Sc1 where scribe and corrector attempt a similar approach.

THE SEPTUAGINT AND THE SECONDARY VERSIONS 187

Ziegler’s apparatus identifies the likely source text of the Armenian version at 37:12b among “rel[iqui],” that is, “the remainder (of the witnesses).” The fascinating treatment of Theodotion’s transliteration provides an insight into the way that the text was read by a translator who was attempting to make the Greek text intelligible to readers of a different language: the text must make some sort of sense! By a new division of the words and some ingenuity the translator makes sense of Theodotion’s transliteration. There is a translation strategy at work. Finally, the Armenian is also instructive in its handling of his Old Greek text at Job 39:29a, where the OG attests ἐκεῖσε ὢν ζητεῖ τὰ σῖτα (“From there it [that is, the vulture] searches for food”; NETS 2007). The Greek translator represented ‫“( מׁשם‬from there”) with the adverb ἐκεῖσε (“to there”), to which he adds ὤν, the present masculine participle of εἰμί (“to be”). What does ἐκεῖσε ὤν mean? Brenton (1844) renders it with “Thence (he seeks food),” that is, “from there.” In Hellenistic Greek the distinction between the adverbs ἐκεῖσε and ἐκεῖ has been lost. The former is “more elegant” (see BDF §103). The Armenian version translates ἐκεῖσε ὤν ζητεῖ with the adverb “there” and two finite verbs, անդ կայ և խնդրէ (“there it is and searches,” or perhaps more colloquially, “there it is perched and searches”). The translator renders ἐκεῖσε simply with անդ (“there”). The common Greek collocation that consists of a participle plus a finite verb the translator renders with two finite verbs joined by the conjunction “and.” This rendering anticipates that of LXX.D, which translates ἐκεῖσε ὤν as “Dort ist er (und sucht seine Nahrung)” (“There it is [and seeks its food]”). This rendering appears preferable to Brenton and NETS, which are indebted to the Hebrew and NRSV, respectively. The vulture is diligent: it searches for food even while sitting on its brood (v. 27b). A revised NETS should read, “While it is there it searches for food.”

General Conclusions Concerning the Importance of the Secondary Versions The use of the secondary versions for the purposes of text criticism begins, if not with Origen, then with the great polyglot Bibles of the sixteenth century, where they are presented side by side with the Hebrew and Greek texts. Today they are relegated largely to occasional citation in the compact editions of the Greek NT, represented by Nestle-Aland’s twenty-sixth edition or the fourth revised edition of UBS, The Greek New Testament. The Göttingen edition of the Old Greek continues to collate the versions fully; these editions offer complete collations of evidence, from papyri, Greek manuscripts (uncial and minuscule), the secondary versions and patristic sources (commentaries and such like). Here too, however, users should be aware that the evidence of the secondary versions has not in every instance been handled well or that it has been dealt with mechanically, and that it is not fully trustworthy apart from critical editions of this evidence. This is not a matter of competence, but a matter of the overwhelming requirements placed upon the editors in the past. The situation is improving with newly available critical editions of versional evidence and the delegation of the collation of the secondary versions to specialists in those languages. The twin discoveries of a sizeable number of early Greek papyri and of the Dead Sea Scrolls, together with advances in the science of textual criticism, have served to relegate secondary version evidence to a level of lesser importance. It might be possible to argue that much of the evidence of the secondary versions is now of very little significance at all and is, perhaps, not worth collating because it is not decisive for determining original readings in source texts. That would be short-sighted. The evidence of the secondary

188

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

versions contributes to the understanding of the textual tradition of the Bible. Part of the role of the text critic is to explain this unfolding story in all of its developments. In this new situation, with more and better resources and developments in the science of textual criticism, the place of the secondary versions requires a reassessment. The recent appearance of Brill’s THB gives generous attention to the secondary versions. For example, the Armenian version has never received such extensive, ground-breaking recognition. The continuing importance of the secondary versions resides in at least three contributions this evidence makes: first, the evidence of the versions is important because it represents one more tool by which to practice textual criticism upon the Old Greek text. True, it is evidence of a particular kind, but it offers one more point of leverage. Second, the secondary versions are important because their witness is relatively early, certainly earlier than that of the medieval Greek manuscripts. For example, the translation of the Bible into Armenian took place in the very early fifth century. The Armenian text is as old as or older than codex Alexandrinus. For the textual criticism of the Old Greek, it is the Greek manuscripts that lie behind the secondary versions that are the text-critic’s interest: therefore, retroversion is essential. The limitations of a secondary version in this respect must also be recognized. When several secondary versions alone share a variant, perhaps the language of the source text could not be duplicated in another language. Third, the witness of the secondary versions offers insight into the history of the biblical text and how it was understood, because each secondary version represents not only a translation but also something of a commentary on the source text.

ONGOING RESEARCH QUESTIONS Whatever usefulness the secondary versions might have rests on the availability of critical editions of their texts. To cite again the example of the Armenian version, there is a critical edition of only a single book of the New Testament (Alexanian 2012). For the Old Testament and Apocrypha we are in an only slightly better situation: critical editions that present a critically established text exist for only two books, Job (Cox 2006) and 4 Ezra (Stone 1979). Further, the number of scholars who possess the multi-disciplinary skills to use the secondary version in relation to its source text(s) is small. The required skills include a specialized competence in the language of the secondary version, in the languages of the source texts, in the science of textual criticism, and in the history of the biblical texts generally. Rewarding opportunities await those who possess these tools.

RECOMMENDED RESOURCES 1. Alexanian (1992), Birdsall (1992), Bogaert (1992), Brock (1992), Mills (1992), Zuurmond (1992). These entries on the various subversions in ABD provide a good source for introduction with helpful bibliography. 2. Fernándos Marcos (2000b: 346–62). A generally useful overview, with brief introductions to the secondary versions, as well as a bibliography. 3. Lange and Tov (2016, 2017a, 2017b). The first volume (2016) of THB provides excellent introductions to the various versions (Greek, Targums, Peshitta, Latin Vulgate) and secondary versions of the Hebrew Bible, while the two other volumes (2017a, b) provide information on the various primary versions and secondary versions with respect to individual books.

THE SEPTUAGINT AND THE SECONDARY VERSIONS 189

The contributions are written by specialists in the various versions and, as such, are far more reliable than short introductions by those who must rely on secondary sources. An essential resource. 4. Würthwein (2014). A uniquely useful introduction to the transmission of the biblical text from Hebrew, to Greek, and to the versions. However, the treatment of the secondary versions like the Armenian is extremely brief and inadequate. 5. Aejmelaeus and Quast (2000). Includes helpful contributions on the Latin, Coptic, Ethiopic, “Syro-Hexaplaric,” Armenian, and Georgian secondary versions.

190

CHAPTER 13

The Septuagint and Origen’s Hexapla PETER J. GENTRY

INTRODUCTION AND OVERVIEW Sometime between 234 and 250 CE in Caesarea of Palestine, the Christian scholar Origen (d. 253/4 CE) produced an edition of parallel-versions of texts of the Old Testament that enabled him to compare the Septuagint with several later Jewish revisions of the Septuagint, as well as with the Hebrew text of his time. This production of a synoptic text of the OT was labeled the Hexapla by later writers because it was sixfold, that is, contained six editions of the text. It seems that the various editions were arranged in columns (see below). The first provided the text of the OT in Hebrew, in the square, “Assyrian” script used by the Jews of Origen’s day. The second contained a full vocalization of the reading tradition of the Hebrew text represented by means of transliteration into Greek, supplying the oral reading tradition and allowing Origen to pronounce or read the Hebrew text aloud.1 The fifth column contained the text of the Septuagint as received by Origen before 230 CE. Wevers (1992: 40) argued that the text received by Origen as the Septuagint already evinced some casual and sporadic revision toward the Hebrew. (Sorting out whether the differences are translational or textual is problematic.) In columns three, four, and six, Origen placed the Jewish revisions of “the Three”: Aquila, Symmachus, and Theodotion, respectively.2 In addition to the Greek versions of the Septuagint and the well-known Jewish recentiores (chronologically: Theodotion, Aquila, and Symmachus), Eusebius mentions other recensions that are simply called “Fifth” (Quinta), “Sixth” (Sexta), and “Seventh” (Septima) since their origins were unknown (see Eusebius, Hist. Eccl. 6.6, discussed below.) Fragmentary manuscript evidence indicates that these others only occurred in some books, such as Psalms, where the sixth column was probably occupied by Quinta (Caloz 1978), while the texts of versions such as Theodotion and Sexta were indicated by

Here we must remember that, before the scribal activity of the Masoretes (c. 700–900 CE), copies of the Hebrew Scriptures did not represent vowels fully in the system of writing. Beginning sometime in the ninth century BCE, the letters hê, wāw and yôd (and later also sometimes ʾālep) were used to represent long vowels and, in the Qumran Scrolls, even short vowels. However, this system was not consistent or systematic and, moreover, certainly did not represent all the vowels. 2 On the major recensions, see Mäkipelto in the present volume. 1

192

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

marking the variations in a few marginal notes in the columns of the versions to which they were most similar. Hence, the names Heptapla and Octapla occur in several sources (Field 1875: I:xi). In order to both understand and undertake research in the Hexapla, it is necessary to put Origen into his historical and social setting. Origen grew up and lived in Alexandria until at least 220 CE. Certainly he would have been instructed in the tradition of the Alexandrian grammarians, which goes back to Ptolemy I who brought scholars from Athens and founded the famous library in Alexandria. There, five generations of scholars collected and edited the canonical texts of ancient Greece (Fraser 1972), as shown in Table 13.1.3 Hellenistic grammarians combined both literary scholarship and textual criticism in their work. Dionysius Thrax provides a good definition (Ars Grammatica §1): Γραμματικὴ ἐστιν ἐμπειρία τῶν παρὰ ποιηταῖς τε καὶ συγγραφεῦσιν ὡς ἐπὶ τὸ πολύ λεγομένων. Μέρη δὲ αὐτῆς ἐστιν ἕξ· πρῶτον ἐνάγνωσις ἐντριβὴς κατὰ προσῳδίαν, δεύτερον ἐξήγησις κατὰ τοῦς ἐνυπάρχοντας ποιητικούς τρόπους, τρίτον γλωσσῶν καὶ ἱστοριῶν πρόχειρος ἀπόδοσις, τέταρτον ἐτυμολογίας εὕρεσις, πέμπτον ἀναλογίας ἐκλογισμός, ἕκτον κρίσις ποιημάτων, ὅ δὴ καλλιστόν ἐστι πάντων τῶν ἐν τῇ τέχνῇ. Grammar is an experimental knowledge of the usages of language as generally current among poets and prose writers. It is divided into six parts: 1. Trained reading with due regard to prosody; 2. Explanation according to poetical figures; 3. Ready statement of dialectical peculiarities and allusions; 4. Discovery of etymology; 5. An accurate account of analogies; 6. Criticism of poetical productions, which is the noblest part of grammatical art.4 Key terms from this passage to be considered for our study are as follows: ἔκδοσις, σημεῖα, ὑπομνήματα, and διόρθωσις. The term ἔκδοσις means “edition” and refers to the Table 13.1 Librarian Tenures in Ptolemaic Alexandria Head Librarian

Ruler

Zenodotus of Ephesus (c. 285–270)

Ptolemy I Soter (306–282) Ptolemy II Philadelphus (282–246)

Apollonius Rhodius (c. 270–245)

Ptolemy II Philadelphus (282–246)

Eratosthenes of Cyrene (c. 245–204/201)

Ptolemy III Euergetes (246–222) Ptolemy IV Philopator (222–204) Ptolemy V Epiphanes (204–180)

Aristophanes of Byzantium (c. 204/201– 189/186)

Ptolemy V Epiphanes (204–180)

Apollonius Eidographos (c. 189/186–175)

Ptolemy V Epiphanes (204–180) Ptolemy VI Philometor (180–145)

Aristarchus of Samothrace (c. 175–145)

Ptolemy VI Philometor (180–145)

Chronological chart is from Francesca Schironi, personal communication, November 2017. It is based on Fraser (1972: 1:447-9). See Schironi (2018). All dates BCE. 4 Translation based on Davidson (1874: 3–4).

3

THE SEPTUAGINT AND ORIGEN’S HEXAPLA 193

grammarian’s own personal or privately corrected copy of a text. The grammarian’s ἔκδοσις was not a published work. Further, the earliest grammarians employed σημεῖα or “signs” to indicate problems in the text. Aristarchus improved a system of signs invented by Zenodotus and Aristophanes. It is sufficient to mention just four of them at this point: (1) From Zenodotus the obelos (—) for athetesis or rejected readings (2) From Aristophanes the asteriskos (※) for repeated lines (3) Innovative use of antisigma (Ɔ) for transposed lines (4) The (dotted) antisigma periestigmenon (·Ɔ·) for tautologies Eventually the marginal notes required detailed explanation, so ὑπομνήματα or “commentaries” were produced in addition to the ἔκδοσις to comment on and explain further the problems indicated by the σημεῖα. Finally, the term διόρθωσις refers to types of “correction” made by these scholars to their own private text. Related developments are worth mentioning at the outset. Recently, Anna Kharanauli (2018) used the research of Neuschäfer (1983) and Sgherri (1977) to locate passages in the commentaries of Origen where he refers to issues in the text that correspond precisely to the range of matters covered by διόρθωσις in the Alexandrian tradition. Origen groups examples of diaphony (διαφωνία) in the text and discusses their probable origin. Some differences are due to variations in the Hebrew manuscripts. Others are unintentional errors made by copyists of the Septuagint such as dittography, haplography, or orthography. According to Origen, still others are due to the impiety of wicked scribes. Furthermore, Dickey (2015) has demonstrated that columnar translation of the works of Cicero and Virgil created a model for Origen: employment of columns as an interpretive tool for bilingual texts in both original and translation.

RESEARCH METHODS, PRIMARY SOURCES, AND VIEWS With this historical background we may now discuss how to interpret the evidence left to us of Origen’s Hexapla. One of the problems in the history of Hexapla scholarship is failure to recognize the multidisciplinary nature of research necessary for accurate conclusions. Relevant fields include: accurate knowledge of ancient book copying procedures, the codicology of the manuscript sources, geography, ancient history, and the textual relations of the various witnesses and the history of the textual transmission (Gentry 2016a). Students of the Septuagint are also wise to collaborate with Classicists in their work on the history of the Alexandrian scholars.5 Primary evidence for Origen’s Hexapla survives in four kinds of sources, the first two of which may be treated together: (1) manuscripts containing copies of the actual synopsis or columnar arrangement of texts; (2) marginal notes in manuscripts, especially Catena manuscripts, of the Septuagint that provide readings from the Hexapla;

Grafton and Williams (2006) have also demonstrated the importance of viewing Origen and Eusebius from the perspective of social historians, but that approach alone overlooks the background of Origen as an Alexandrian grammarian.

5

194

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

(3) colophons in manuscripts, some in Greek manuscripts, but the majority of them in manuscripts of the Syro-Hexapla (see below); and (4) brief statements in early patristic texts made, for example, by Origen, Eusebius of Caesarea, Epiphanius, Jerome, Rufinus, and Theodoret of Cyrrhus.

Manuscript Evidence and Marginal Notes There are only four fragmentary manuscripts constituting copies of the actual Hexapla synopsis, which are presented in Table 13.2.6 Additionally, Olivier Munnich (1995: 172–4) has argued that traces of the columnar system used by Origen may be found in the annotation system of the marginalia of RA 344 (Athos, Παντοκρατορος 24, tenth century), an important manuscript belonging to the “s” group in the Göttingen editions of the Pentateuch edited by Wevers. This claim is questionable. Regardless, all copies of the actual synopsis are fragmentary and fairly late. Nonetheless, the evidence of these four preserved fragments, along with the descriptions by Jerome (Ep. ad Titum 3.9, see below) and Epiphanius (De Mens. Et Pond. §7), support the idea that the Hexapla was laid out in columns. We also have manuscripts that are derived from the Fifth Column. Some of them are not only early, but also preserve the Aristarchian signs. These are presented in Table 13.3.

Table 13.2 Known Manuscripts of Hexapla Synopsis 86

Rome, Biblioteca Vaticana, Barberiniani gr. 549

Ninth/tenth century (Cat.)

Contents: One verse of the Hexapla in a Twelve Prophets MS (Hos. 11:1) Edition: J. Ziegler (2014) with F. Albrecht 113

Milan, Biblioteca Ambrosiana, B. 106 sup.

c. 966 CE (Cat.)

Contents: A twelfth-century hand cites two lines of the heading of the book of Psalms in a note attached to the flyleaf Edition: Mercati (1895–6: 663–7) 1098

Milan, Biblioteca Ambrosiana, O. 39

Tenth century (Cat.)

Contents: Approximately 148 verses from the Psalms Edition: Mercati (1958) 2005

Cambridge, University Library T-S 12. 182

Seventh century (Cat.)

Contents: Thirteen verses from the Psalms Edition: Taylor (1900)

Listed according to Rahlfs’s number, physical location, library signature, date and manuscript type, contents, and edition. Also below.

6

THE SEPTUAGINT AND ORIGEN’S HEXAPLA 195

Table 13.3 Hexapla Fifth Column Manuscripts 922

Oxford, Bodleian Library

250–350 CE

Gr. bib. d. 4 = SC 31708 = P.Grenf. 1.5 928

Oxford, Sackler Library

250–325 CE

P.Ant. 1.8 and 3.210 G

Leiden, Bibliotheek der Rijksuniversiteit

Fourth/fifth century

Voss. graec. in qu. 8 “Codex Sarravianus-Colbertinus” V

Venice, Biblioteca Marciana

Eighth century

Gr. 1 = “Codex Venetus”

Of these, MS 922 contains fragments of a hexaplaric text of Ezek. 5:12-6:3 and clearly attests asterisks. Schironi (2015: 24) prefers to date this papyrus codex to 250–350 CE. Also, Cuppi (2012: 24) concludes that MS 928 (which also constitutes fragmentary remains of a papyrus codex of Proverbs, Wisdom, and Sirach) has an asterisk as well. In G, 153 out of 454 original folios of a codex of the Pentateuch are preserved, in which lines are also marked by asterisks and obeli. Although G is probably a century later, MSS 922 and 928 are likely to date within fifty years of Origen’s death, so these manuscripts bring us very close to Origen himself.

Colophons in Manuscripts Colophons are brief notes, normally at the end of a main text in a manuscript, written by the copyist himself and providing various information concerning the copyist and his work, the date of copying, and sometimes the sources used. They are like scribal signatures. In majuscule manuscripts colophons are in the scribe’s own handwriting, at that point permitted to revert to his own idiosyncratic style of script and not required to follow the formal uncial style to write the colophon. Hiebert (2001: 182–5) provides the most extensive list of colophons to date, though it is nonetheless incomplete. Eight are preserved in three Greek manuscripts and some twenty-two in eleven different manuscripts of the Syro-Hexapla as shown in Tables 13.4 and 13.5.7 The last serious study of colophons relating to the Hexapla was by Giovanni, Cardinal Mercati (1941: 1–48). Texts of the colophons given in later works, such as Devreesse (1954: 122–6), Nautin (1977: 322–5), and Grafton and Williams (2006: 340–2), are dependent upon Mercati (1941) and Middeldorpf (1835). The body of texts given by Mercati by no means represents all the available colophons, and in the case of those in the Syro-Hexapla he generally presents less than half of the text. With the further abbreviation of these texts by later scholars, the information is not adequately represented. A forthcoming monograph by Gentry and Meade will present the texts in

The sigla in the left-hand columns are from Rahlfs and Wright. See Wright (2004).

7

196

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Table 13.4 Greek Colophon Manuscripts Related to the Hexapla Q S 88

Rome, Biblioteca Apostolica Vaticana

Fol. 171–2

Isaiah

Vat. gr. 2125 = “Codex Marchalianus”

Fol. 568

Ezekiel

London, British Library

Quire 36 Fol. 5r

2 Esdras

Add. 43725 = “Codex Sinaiticus”

Quire 37 Fol. 3r

Esther

Rome, Biblioteca Apostolica Vaticana

Fol. 121r

Jer./Baruch

Chigiani R VII 45

Fol. 130r

Jer./Lamentations

Fol. 167r

Daniel

Fol. 315v

Ezekiel

Table 13.5 Syro-Hexapla Colophon Manuscripts London, British Library XLVIII

Add. 14,442

Fol. 46b

Genesis

XLIX

Add. 12,134

Fol. 132b

Exodus

LI

Add. 12,133

Fol. 169b

Joshua

LII

Add. 17,103

Fol. 70b

Judges, Ruth

LIII

Add. 14,437

Fol. 122a

3 Kingdoms

LV

Add. 14,434

Fol. 128b

Psalms

LVIII

Add. 14,668

Fol. 29b

Ezekiel

3: xi

Or 8732

Fol. 136b

Isaiah

Fol. 90a

4 Kingdoms

Princeton, University Library

Fol. 18a

Genesis

Scheide Library M150

Fol. 62b

Exodus

Fol. 138b

Numbers

Milan, Biblioteca Ambrosiana

Fol. 38v, 52r

Job

c. 313. Inf.

Fol. 66r

Proverbs

Fol. 70r

Ecclesiastes

Fol. 72r

Canticles

Fol. 80r

Wisdom of Solomon

Fol. 114r

Twelve Prophets

Fol. 142r

Lamentations

Fol. 150v

Daniel

Fol. 173r

Ezekiel

Fol. 193r

Isaiah

Paris, Bibliothèque nationale Syr. 027

THE SEPTUAGINT AND ORIGEN’S HEXAPLA 197

full (with translations) of all the known colophons. Five colophons are presented below to demonstrate the significance of these sources. The “Esther” Colophon London, British Library, Codex Sinaiticus, fourth century (Quire 37, Fol. 3r) Critical Edition (ed. Gentry)8 Ἀντεβλήθη πρὸς παλαιώτατον λίαν ἀντίγραφον δεδιορθωμένον χειρὶ τοῦ ἁγίου μάρτυρος Παμφίλου. Πρὸς δὲ τῷ τέλει τοῦ αὐτοῦ παλαιωτάτου βιβλίου ὅπερ ἀρχὴν μὲν εἶχεν ἀπὸ τῆς πρώτης τῶν βασιλείων· εἰς δὲ τὴν Εσθηρ ἔληγεν, τοιαύτη τις ἐν πλάτει ἰδιόχειρος ὑποσημείωσις τοῦ αὐτοῦ μάρτυρος ὑπέχειτο ἔχουσα οὕτως: Μετελήμφθη καὶ διορθώθη πρὸς τὰ ἑξαπλᾶ ᾽Ωριγένους ὑπ’ αὐτοῦ διορθώμενα. Ἀντωνῖνος ὁμολογητὴς ἀντέβαλε, Πάμφιλος διώρθωσα τὸ τεῦχος ἐν τῇ φυλακῇ διὰ τὴν τοῦ θεοῦ πολλὴν καὶ χάριν καὶ πλατυσμόν. [καὶ εἴγε μὴ βαρὺ εἰπεῖν, τούτῳ τῷ ἀντιγράφῳ παραπλήσιον εὑρεῖν ἀντίγραφον οὐ ῥᾴδιον.] ⋙⋙ διεφώνει δὲ τὸ αὐτὸ παλαιώτατον βιβλίον πρὸς τόδε τὸ τεῦχος εἰς τὰ κυρία ὀνόματα ⋙⋙

[1] [The present manuscript was] collated against a very old copy corrected by the hand of the holy martyr Pamphilus. At the end of this same very old book, which began at First Kingdoms and ended at Esther, such a signature, broadly speaking, in the same martyr’s own hand is appended as follows: [2] “Copied and corrected by the Hexapla of Origen, as corrected by his own hand. Antoninus the confessor collated, and I, Pamphilus, corrected the volume in prison, by the great and wide favour of God. [And if it not be presumptuous to say so, it would not be easy to find a copy equal to this copy.]” Now the old book disagrees with this volume in respect to certain proper names.9

The note is divided in two by indentation. The first part ([1]) is by a corrector or scribe of Codex Sinaiticus, who tells us that he used an extremely old manuscript containing the books of Kingdoms to Esther to correct the codex. This ancient copy was corrected by Pamphilus, a great admirer of Origen’s work. The corrector or scribe cites the colophon in his source text to prove this. The second part of the note ([2]) is a copy of the colophon in the copyist’s source text which was written in Pamphilus’s own handwriting. It states that the source text was first copied from the Hexapla and then corrected against the Hexapla with Antoninus as collator and Pamphilus as corrector. Also noted is the fact that Pamphilus did the work in prison. Skeat (2004) suggests 309 CE as a date since Pamphilus was martyred the following year. The sentence about the manuscript being “a best copy” may have been written by Pamphilus or perhaps even by the scribe who corrected Codex Sinaiticus using his manuscript.

Indentation and sigla as in manuscript. Unless otherwise noted, all translations are the author’s.

8 9

198

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Colophon to Proverbs in Codex Syro-Hexaplaris Milan, Biblioteca Ambrosiana, c. 313. Inf., ninth century (Fol. 66r) ‫ܪܫܝܡ ܗܘܐ ܒܟܬܒܐ ݂ܝܘܢܝܐ‬ ‫ܕܡܢܗ ܐܬܦܫܩ ܠܣܘܪܝܝܐ ܟܬܒܐ‬ ܵ ‫ܗܢܐ ܕܦܐܠܬܐ܃ܒܬܪ ܫܘܠܡܐ‬ ܵ ‫ܕܝܠܗܝܢ ܗܟܢܐ ܀ ܐܬ ܼܢܣܒܝܢ‬ ܵ ܵ ‫ܬܐ܂ ܡܢ ܨܚܚܐ‬ ݂ ‫ܘܐܬܦܚܡܝܢ ܦܐܠ‬ ݁ ‫ܘܐܬܟ ܼܬܒܘ‬ ‫ܚܬܝܬܐ ܕܐܬܣܝܡ‬ ݂ ܵ ‫ܒܐܝܕܐ‬ ‫܂‬ ‫ܘܠܝܐ‬ ݂ ‫ܒܗ ܡܢ ܠܒܪ ܣܟ‬ ܼ ‫ܕܦܡܦܝܠܘܣ ܘܕܐܘܣܒܝܘܣ܂ ܕܒܗ‬ ܵ ‫ܪܫܝܡܢ ܵܗ ܼܘܝ ܘܗܠܝܢ ܀ ܐܬ ܼܢ ܵܣܒܝܢ‬ ܵ ‫ܡܢ ܫܬܝܬܝ‬ ‫ܦܨܐ ܕܐܘܪܓܢܝܣ‬ ‫ܗܠܝܢ ܕܐܫܟܚܢܢ ܀ ܀ ܀‬ ‫ܘܬܘܒ ܀ ܒܟܝܪܐ ܕܝܠܗ܂ ܦܡܦܝܠܘܣ‬ ‫ܘܐܘܣܒܝܘܣ ݁ܬܪܨܘ ܀‬ ܼ [1] It was noted in the Greek book from which this book of Proverbs was translated into Syriac, after the end of them, as follows: [2] “The Proverbs were copied and collated from an accurate copy that was made in which scholia were written in the margins, by the hand of Pamphilus and Eusebius. In it these words were also inscribed: [3] ‘These things that we found were taken from the Hexapla Version of Origen’. “And again, [4] in his own handwriting, ‘Pamphilus and Eusebius corrected.’” Major punctuation signs (‫ )܀‬divide this colophon into four sections, with one repeated three times after text that does not fill out the line. The first part ([1]) indicates that the last three parts constitute a colophon or colophons in the Greek Vorlage of the Syriac translation. The second part ([2]) mentions a colophon in the Greek Vorlage of the Syriac translation that describes how it was produced. It claims that the source text was physically similar with marginal notes and that these marginal notes or scholia were copied by no less than Pamphilus and Eusebius. The third and fourth sections constitute the colophon in the source text from which the Greek Vorlage was copied and belong together, as the word “and again” (wtwb) indicates. The third section ([3]) employs the first person plural to indicate that Pamphilus and Eusebius found the materials that they provided in the marginal notes in Origen’s Hexapla. The fourth section ([4]) declares that Pamphilus and Eusebius corrected the text produced in this way. This final statement is difficult to interpret. In fact, much of the scholarly commentary on these colophons is speculative. Probably this final statement refers to the copying process. A pandect bible manuscript was copied by a group or team of scribes, normally numbering three. Then experts corrected the manuscript. Often one read the exemplar text aloud while another made corrections in the copy (Skeat 2004). This would mean that the source text for the Vorlage of the Greek copy used by the Syriac translators was produced by scribes under the supervision of Pamphilus and Eusebius (Gentry 2016a). Colophon to 3 Kingdoms London, British Library, LIII, Add. MS 14,437 (Fol. 122a) The manuscript, that in our possession (?) was translated from Greek into Syriac, was taken from the Hexapla, that is, from the “Six Columns” that was among the manuscripts

THE SEPTUAGINT AND ORIGEN’S HEXAPLA 199

of the “Six Columns” of the Library of Caesarea, Palestine, and was collated to a copy in which there was noted at the end as follows: “I, Eusebius, corrected as accurately as possible.” It was translated, then, from the Greek language into Syriac in the month Shebat, of the year 927 according to the numbering of Alexander, Fourth Indiction at Enaton of Alexandria in the Holy Monastery of Antonine Monks. Colophon to Isaiah London, British Library, 3: xi, Or 8732, 734 CE (Fol. 136b) The Prophecy of Isaiah was completed according to the Seventy. It was taken and copied from a book of Eusebius and Pamphilus that also they corrected from the library of Origen. Praise to the Father and praise to … Colophon to 4 Kingdoms Paris, Bibliothèque nationale, Syr. 027, 15 December 719–18 January 720 CE (Fol. 90a) This book was translated from the Greek language into Syriac, from the Version of the Seventy-Two. The Holy Abbot Mar Paul, faithful Bishop in Alexandria, the great city, at the command and exhortation of the holy and pious Mar Athanasius, faithful Patriarch. In the monastery of Saint Mar Zacchaei Callinicensis while he was staying in Alexandria, in the days of the beloved of God, Mar Theodora, Archimandrite of the Cloister. In the year 928, Fifth Indiction. Whoever reads, let him pray for the beloved of God Mar Thoma, deacon and syncellus of the holy and pious Mar Athanasius, Patriarch, who labored and provided for the rest of them who labored and toiled with him, that God may repay them on account of their efforts and labors the salvation of their souls by the prayers of his children and of all his saints. Colophon to Job in Codex Syro-Hexaplaris Milan, Biblioteca Ambrosiana, c. 313. Inf., ninth century (Fol. 52r) The book of Job, the righteous, has been completed according to the version of the Seventy. Job was taken from the old Tetrapla. To return to the final statement in the Colophon to Proverbs in Codex SyroHexaplaris, an alternative interpretation of the fourth section is also possible that can be held in conjunction with the first interpretation and which may be derived from other statements in the colophon. The proposal here has been anticipated by G. Jenkins (1991) and Schironi (2012 and 2015) and is as follows. Origen made his own ἔκδοσις from the Hexapla. The format of this document was exactly what we see in the Syro-Hexapla and was, in fact, the Tetrapla.10 According to the colophons, Pamphilus and Eusebius continued to develop the ἔκδοσις created by Origen from the Hexapla. They added marginal notes or readings from the Hexapla. Thus, an edition was produced from the Fifth Column of the Hexapla that eventually became textually distant from that of the Fifth Column. This edition was the Tetrapla and had the signs and readings from the Three where they differed from the Septuagint. Widely varying views exist on the relation between the Hexapla and the Tetrapla, as detailed by Hiebert (2001). The most common one is that the Tetrapla preceded the Hexapla. This article argues the reverse. As Schironi (2012) argues, Origen simplified the Aristarchian signs, using almost exclusively the obelos (÷) for passages not in the Hebrew and the asteriskos (※) for passages inserted and hence repeated from the Three where the Septuagint lacked material corresponding to the Hebrew. The antisigma (Ɔ) was employed as well, although rarely, For an explanation of another edition produced by Origen called the Tetrapla, see below.

10

200

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

to mark transpositions.11 Furthermore, Schironi shows that P.Grenf. 1.5, a fragmentary papyrus from Oxyrhynchos dated only fifty years after Origen, exemplifies the beginnings of this ἔκδοσις. Codices G, M, and V also have passages marked by asteriskoi and obeloi. On the other hand, the copies of the Hexapla that have survived have no signs and no additional material added from the Three where text is extant in Hebrew but not in the Septuagint. Therefore, the codices with the signs are from the Tetrapla, the manuscripts without them are copies of the Hexapla. In sum, the colophons cast important light on the early history of the Hexapla and yet remain largely unexplored.

Patristic Testimony Patristic testimony helps shed light upon both the nature of the Hexapla itself, along with Origen’s working method. As for the former, Origen’s own statements do not provide a description of the Hexapla as such, but we do learn from later sources such as Eusebius, one of his successors in Caesarea, and from the fragmentary copies that survive of the Hexapla Psalter (noted above). The description of Eusebius, bishop of Caesarea from 314 to 339 CE, is as follows (Hist. eccl. 6.16; cf. Rufinus, Hist. 6.16.4): So accurate was the investigation that Origen brought to bear upon the divine words, that he learned the Hebrew language thoroughly, and acquired his own copies—in actual Hebrew script—of the original Scriptures transmitted by the Jews. He tracked down the other versions of those who translated the holy writings besides the Seventy [translators]. In addition to the well-known translations of Aquila and Symmachus and Theodotion, he discovered a few others, which having escaped notice for a long time, he tracked down and brought to light (from what hidden nooks I do not know). On account of their obscurity and not knowing the translator, he merely indicated this: that the one he found at Nicopolis, near Actium, and the other in another place such as this. At any rate, in the Hexapla of the Psalms, after the four well-known editions, he placed beside them not only a fifth, but also a sixth and a seventh translation; in the case of one of these he indicated that it was found at Jericho in a jar in the time of Antoninus the son of Severus [211–217 CE]. Bringing all of these together into the same [copy], he separated them by cola [phrases] and correlated them one to another, [placing them] after the actual Hebrew text, and so he left us the copies of the Hexapla, as they are called. He made a further separate arrangement of the editions of Aquila, Symmachus, and Theodotion, together with the version of the Seventy, in the Tetrapla. In a few places Origen does discuss his own textual work (Comm. Matt. 15.14): Now it is clear that the difference among the copies is great, either from the carelessness of certain scribes, or from the knavish audacity of some, or from some neglecting to correct what is written, or from some adding to or taking away in the correction the things that seemed good to themselves. With the help of God’s grace, I have tried to repair the disagreements in the copies of the Old Testament on the basis of the other versions. When I was uncertain of the Septuagint reading because the various copies did not tally, I settled the issue by consulting the other versions and retaining what was in agreement with them. Some passages did not appear in the Hebrew; these I marked

Schironi does not seem to be aware that the antisigma was used as well, although rarely.

11

THE SEPTUAGINT AND ORIGEN’S HEXAPLA 201

with an obelos as I did not dare to leave them out altogether. Other passages I marked with an asterisk to show that they were not in the Septuagint but that I had added them from the other versions in agreement with the Hebrew text. Elsewhere, Origen also discusses why he accepts passages in the Greek Daniel that are not in the Hebrew (Ep. Afr. 3–7): Know then, in respect to these things, what we must do not only in the case of the History of Susanna that is in use in every church of Christ in that Greek copy which the Greeks use, but is not in the Hebrew … but also of thousands of other passages which I found in many places when, with my modest effort, I was comparing the Hebrew copies with ours … For so Aquila, following the Hebrew reading, gives it, who has obtained credit among the Jews of having interpreted the Scriptures with no ordinary care, and whose version is most commonly used by those who do not know Hebrew, as the one which has been most successful. Of the copies in my possession whose readings I gave, one follows the Seventy, and the other Theodotion; and just as the History of Susanna which you call a forgery is found in both, together with the passages at the end of Daniel, so they give also these passages, amounting, to make a rough guess, to more than two hundred verses … in many other of the sacred books I found sometimes more in our copies than in the Hebrew, sometimes less. I adduce a few examples … in the whole of Job there are many passages in the Hebrew which are wanting in our copies, generally four or five verses, but sometimes, however, even fourteen, and nineteen, and sixteen. But why should I enumerate all the instances I collected with so much labor, to prove that the difference between our copies and those of the Jews did not escape me? Again, in Genesis, the words, “God saw that it was good,” when the firmament was made are missing in the Hebrew, … and other instances are to be found in Genesis, which I marked, for the sake of distinction, with the sign the Greeks call an obelos, as on the other hand I marked with an asterisk those passages not found in our copies which are in the Hebrew. Origen’s statement in the Epistle appears to be diametrically opposed to that in his Commentary. In the former Origen advocates the Septuagint as the Bible of the Church; in the latter he accepts the Hebrew Bible and the recentiores as the instruments to heal the διαφωνία, or “disagreement,” among manuscripts of the Septuagint. The debate over Origen’s assessment of the different witnesses to the biblical text began early: Rufinus and Jerome took up opposing interpretations of his position. But we must remember that Origen was a versatile man with apologetic, educational, and exegetical as well as textual interests. In some of his endeavors these interests coincided, but not necessarily in all (see Kamesar 1993 and esp. Romeny 1997: 113–15). But there is another intriguing statement about his working method in the newly published Homilies on the Psalms by Origen (Hom. Ps. 77.1.1): Καὶ ὅσα μὲν διὰ τὸν θεὸν καὶ τὴν χάριν αὐτοῦ ἐκάμομεν, συνεξετάζοντες καὶ τὰ Ἑβραϊκὰ καὶ τὰς ἐκδόσεις ὑπὲρ τοῦ ἰδεῖν τὴν διόρθωσιν τῶν σφαλμάτων, οἶδεν· ὄσα δὲ θέλομεν καὶ περὶ τὰ λείποντα ποιῆσαι, αὐτὸς εὐοδώσει. He knows all that we have labored over for God and for his grace, in examining together the Hebrew text and the other editions to ascertain the proper correction of these mistakes.

202

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Origen hardly intended his remarks in the Commentary on Matthew or in the Epistle to Africanus to be taken as a complete description of his work. In fact, there is no explicit reason to think that he is describing the Hexapla in these passages. More likely, he is describing his work on the Tetrapla, the edition produced from the Hexapla. On this view, the Hexapla was only a tool to heal the differences between the LXX versus the Hebrew and the Three. This possibility is corroborated by the fact that the Commentary on Matthew and the Epistle to Africanus were written late in Origen’s life: the Hexapla was ancient history by that time, even from Origen’s own point of view. In light of the preceding primary sources, the following facts about Origen’s general procedure in producing the Hexapla are clear and must inform research methodology in this complex area of Septuagint scholarship: (1) The copies of the Old Greek (Septuagint) known to Origen differed from the Hebrew text he knew at various places and for a variety of reasons; (2) The aim of Origen’s work was to bring the Old Greek into quantitative alignment with the Hebrew; (3) Origen marked the passages in his copies of the Greek Old Testament which were lacking in the Hebrew with a sign called an obelos (—); (4) Origen added from other Greek versions available to him passages extant in the Hebrew but absent in the Septuagint and marked these with an asterisk (※). As mentioned above, the asterisk and obelos were text-critical signs developed and used by the librarians in Alexandria and perfected by Aristarchus of Samothrace (c. 217–145 BCE) in the course of editing the texts of Homer. Origen also used a metobelos ( ̷. or ↙) to mark the end of the phrase or word pre-marked by an asteriskos or obelos. Another sign called an antisigma (Ɔ) or lemniscus (∽) was used to mark displaced passages (only found in the Syro-Hexapla). It is probably correct that these critical signs were first placed, not in the original Hexapla synopsis, but in an edition derived from the synopsis begun by Origen and developed further by Pamphilus and Eusebius (Fernández Marcos 2000b: 213–15; Dines 2004: 101; Schironi 2015). Almost every other detail concerning the Hexapla is still debated, including Origen’s purpose in creating it. Was he interested in apologetics, or exegesis, or critical problems in the textual transmission? Whatever the purpose, his work did provide a helpful synopsis allowing one to compare the earliest translation, the Septuagint, with the later Jewish recensions and also with the Hebrew parent text, at least that of c. 200 CE in Caesarea. In terms of the details of its contents, contrary to Nautin’s denial (1977), Jenkins (1998b) has shown that the Mercati Palimpsest did originally contain a column in Hebrew letters. He argues from the codicological measurements of a copy of some folia of Hexapla Psalms preserved in the Cairo Geniza that the creator of the palimpsest text cut off the column in Hebrew. Alternatively, by this time, no Greek scribe would have been able to copy text in Hebrew script and this may be the reason why the first column was left out. Scholars are divided on whether the Tetrapla mentioned by Eusebius ever existed, or, assuming that it existed, whether it was produced perhaps before the Hexapla, or afterward, representing a condensed version of the Hexapla minus the first two columns containing the Hebrew text (Sipilä 1998: 16–38; Hiebert 2001: 184, n. 21). Nonetheless, from the colophons in the Syro-Hexapla there is no reason to doubt the existence of the Tetrapla. In fact, according to the colophon to the Twelve Prophets in the Syro-Hexapla, the Tetrapla had an identical layout on the page as the Syro-Hexapla, with the main text

THE SEPTUAGINT AND ORIGEN’S HEXAPLA 203

a derivative of the Fifth Column of the Hexapla and marginal notes added providing readings of the Three (see also Jenkins 1991 and 1998a). One could assume that where the readings of the Three were not supplied, they were in fact identical to the Septuagint or deviated minimally. Thus, the Tetrapla was “fourfold”; that is, it supplied four editions of the text.12

ONGOING RESEARCH QUESTIONS Aside from the questions raised by the primary evidence discussed above, major ongoing research questions relate to the post-Hexaplaric recension and what Jerome calls the trifaria varietas,13 for which patristic and colophon evidence is also crucial. There is in our manuscript tradition of the Septuagint a recension based on the Fifth Column of Origen’s Hexapla.14 It is difficult to reconstruct the early history of the Hexapla. Origen died in 253. The date given for his successor, Pamphilus, is usually 240–310 CE. We do not know at what point Pamphilus began his labor of preserving the library of Origen (Jerome, Vir. ill. 75). Some of his books were certainly lost during the persecution of Diocletian (Eusebius, Hist. eccl. 8.2), perhaps parts of the Hexapla among them. Many colophons speak of Pamphilus and Eusebius working together to copy either the Hexapla itself or a hexaplaric recension based on it (Mercati 1941; Grafton and Williams 2006). Eusebius was bishop of Caesarea from 312 to 340 and continued to enhance the library at Caesarea greatly, yet his precise role in the transmission of the Hexapla or hexaplaric recensions is uncertain. Euzoius, bishop of Caesarea from 373 to 379, made an effort to copy and preserve papyrus books that were deteriorating on parchment (Jerome, Vir. ill. 113), no doubt the Hexapla among them. We know from Epiphanius (Pan. 64.3.4) that Origen’s books were produced on papyrus and that, by the time of Euzoius, parchment was being used for codices. Patristic writers like Jerome, Epiphanius, Rufinus, and Theodoret of Cyrrhus evince firsthand knowledge of the Hexapla that helps further clarify its characteristics and transmission. Toward the end of the fourth century CE, Jerome provides a clear statement about different editions or versions of the Septuagint at that time in his Praefatio in Paralipomena (PL 28, 1391A–1393A): Now, in fact, when different versions are held by a variety of regions, and this genuine and ancient translation is corrupted and violated, you have considered our opinion, either to judge which of the many is the true one, or to put together new work with old work, and shutting off to the Jews, as it is said, “a horn to pierce the eyes.” The region of Alexandria and Egypt praises in their Seventy the authority of Hesychius; the region from Constantinople to Antioch approves the version of Lucian the Martyr; in the middle, between these provinces, the people of Palestine read the books which, having been laboured over by Origen, Eusebius and Pamphilus published. And all the world contends among them with this threefold variety [trifaria varietas]. And Origen certainly not only put together the texts of four editions, writing the words in The Tetrapla may not have been laid out in columnar format since it did not present original text and translation(s). I am indebted to John D. Meade for this suggestion. 13 Jerome employed the term trifaria varietas to refer to three different revisions of the Septuagint in his time: Hesychian or Egyptian, Origenian or Palestinian, Lucianic or Antiochene. 14 This recension is normally designated by the siglum O in the editions of the Göttingen Septuagint (as opposed to ο΄, the siglum normally used in manuscripts for the text of the Fifth Column of the Hexapla). 12

204

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

a single row so that one regularly differing may be compared to others agreeing among themselves, but what is more audacious, into the edition of the Seventy he mixed the edition of Theodotion, marking with asterisks those things which were missing, and placing obeli by those things which are seen to be superfluous. Some scholars consider Jerome a liar (Nautin 1977: 326–8; Nautin 1985), since he admired Origen’s scholarship to the point of worship and admittedly bragged about his own. His claims echo those of Origen at times (Nautin 1977: 326–8). Nonetheless, Jerome’s testimony on the Hexapla should be believed, since he actually traveled to Caesarea and spent some time consulting the Hexapla in person. If one consults his commentaries on biblical texts (e.g. Ecclesiastes, Jeremiah) they are so full of readings from the Three that either Jerome had copies himself, as he in fact claims, or he made notes in Caesarea to work from, or else traveled very regularly to Caesarea (cf. Jay 1985: 407–17). Jerome also discusses his hexaplaric sources and claims to have transcribed texts from the Hexapla of Origen in the library of Caesarea (Comm. Tit. 3.9): This is also why we have been careful to correct all the books of the old law that the learned man Adamantius [i.e., Origen] had arranged in the Hexapla, having transcribed them from the library of Caesarea, by means of these authentic copies in which the very Hebrew words were written out in their own characters and expressed in the next column in Greek letters. Aquila likewise and Symmachus, the Seventy too, and Theodotion occupy their own column. But some books, especially those that were composed in verse among the Hebrews, have three other added editions, which they call the fifth, sixth, and seventh translation, editions that have followed the authority of the translators without their names. (Scheck 2010: 341–2; cf. Jerome, Comm. Ps. 1.4) Another learned patristic writer was Epiphanius, whose testimony concerning the mechanics of Origen’s work is trustworthy in spite of chronological problems in his report. It also provides important information about the Tetrapla (Pan. 64.3.5-7): First, making a painstaking effort to collect the of the six [Old Testament] versions—Aquila, Symmachus, the Septuagint, Theodotion, (6) and a fifth and a sixth [version] — setting each Hebrew expression next to them, and the actual letters as well. But directly opposite these, in a second column next to the Hebrew, he made still another parallel text, but in Greek letters. (7) Thus this is, and is called a Hexapla, and besides the Greek translations two parallel texts, of the Hebrew actually in letters, and of the Hebrew in Greek letters. It is thus the whole Old Testament in the version called the Hexapla, and in the two Hebrew texts. (translation by Williams 2013: 136) Epiphanius also discusses the Aristarchian signs and the editions in the Hexapla at length (De Mens. et Pond. §§2–21; Syriac version edited and translated by Dean 1935). What is interesting is that when Epiphanius discusses columns he uses the word Hexapla. Yet when he discusses the signs, he does not use the word Hexapla, which indicates that the signs were present in the Tetrapla, but not the Hexapla.15 This observation is based on the research of John D. Meade (personal communication).

15

THE SEPTUAGINT AND ORIGEN’S HEXAPLA 205

Colophons in Codex Marchalianus (Isaiah, Ezekiel), Codex Sinaiticus (II Esdras, Esther), and in the Syro-Hexapla provide important information on the copying of Origen’s Hexapla as already noted. Scholars calculate that the original work consisted of forty codices, each the size of Codex Vaticanus (Grafton and Williams 2006: 96–132). The evidence from the colophons indicates that Antoninus the Confessor, Pamphilus (the successor of Origen), and Eusebius (the successor of Pamphilus at Caesarea) copied the Fifth Column into separate manuscripts. In these copies the biblical books were put together in blocks (e.g., the Wisdom Books of Proverbs, Ecclesiastes, Canticles, Job, Sirach, and Wisdom). Although the prevailing view is that no complete copy of the synopsis was ever made, Ceulemans (2008) has shown that this perspective must be modified. Some of these copies derived from the Hexapla (or the Tetrapla) must have been the parent texts of the Syro-Hexapla. They contained as a main text a copy of the Fifth Column with the Aristarchian signs and readings from Theodotion, Aquila, and Symmachus laid out in the margins just as in the Syro-Hexapla. In later copies, the Aristarchian signs were omitted. Some claim these later copies, in fact, were those sent to Constantine. Constantine ordered fifty “copies” of the Septuagint from Eusebius, which were produced and delivered three or four at a time (Eusebius, Vit. Const. 4.37.1).16 Kraft (2013) has argued that this refers to Eusebius bringing together small-scale codices used earlier for parts of scriptural writings (each roughly equivalent to the contents of a scroll) into a scriptural library from which physical “Bibles” could be produced. Skeat (2004: 20–1) believed it possible that Codex Sinaiticus was produced in Eusebius’s scriptorium at this time but not dispatched because it was not up to standard. Modern scholars suggest that the original Hexapla may have been lost when the Muslims conquered Palestine in 638 (Jellicoe 1968: 124–5). It is possible, however, that it may have survived only in part in copies past the fourth century CE. Overall, it seems that Origen and his successors were interested only in the Tetrapla and not in preserving the original Hexapla. As one can see from the above discussion, our impoverished knowledge of the early history of the Hexapla and the fragmentary remains makes it difficult to distinguish the text of the Fifth Column of the Hexapla, specified in the patristic sources by the siglum ο΄, and the hexaplaric recension of the Septuagint derived from it. Was this due to Origen’s own work in moving from Hexapla to Tetrapla? What changes were introduced by Pamphilus and Eusebius? Was Eusebius responsible for leaving out the special signs when producing the fifty Bibles for Constantine? These are questions that scholars must continue to attempt to answer.

RECOMMENDED RESOURCES 1. Field (1875). To this day, the description in vol. I:iii–xci (tr. Norton 2005) of the Hexapla and his collection of hexaplaric remains is still the best. The Hexapla Institute (http://hexapla.org/). The Hexapla Institute at The Southern Baptist Theological Seminary in Louisville, KY has undertaken the Hexapla Project under

Although the phrase τρισσά και τετρασσά is difficult to interpret.

16

206

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the aegis of the IOSCS and led by Peter J. Gentry of Louisville, KY, Alison Salvesen of the University of Oxford, and Bas ter Haar Romeny of Vrije Universiteit Amsterdam. The goal of the Hexapla Project is to gather all hexaplaric materials published since Field’s Edition and prepare critical editions of all known materials. Thus far the following books have been completed as dissertations: Song of Songs (Ceulemans 2009), Ecclesiastes (Marshall 2007), Numbers (Burris 2009; McClurg 2011), and Job (Woods 2009; Meade 2012, 2020). These works will eventually become available in hard copy by Peeters publishers, if they have not already (e.g., Meade 2020), as well as in the electronic database at The Hexapla Institute website.

CHAPTER 14

The Septuagint and the Biblical Canon JOHN D. MEADE

INTRODUCTION The question of the origins of the Christian Old Testament canon has been a vexing one for students and specialists alike. Two prominent answers have been given. First, that early Christians received a wider Jewish Alexandrian canon containing more books than the twenty-two books of the Palestinian Jewish canon. Second, that Christians formed their own Old Testament canon even after they separated from Jews after 70 CE. Current research on this question focuses on the evidence of the early period, Greek canon lists, Greek citations and commentaries, Greek manuscripts, and early printed editions. Accordingly, views, evidence, and research method are closely intertwined in this area of research and are therefore discussed together as a means of identifying and delimiting the books of the Septuagint canon.

OVERVIEW OF THE PRIMARY EVIDENCE To state the obvious, the best method of research for understanding the Septuagint canon is close examination of the primary sources. This section will therefore discuss the most important ones.

Late Second Temple Period Evidence for a recognized, exclusive list of biblical books or a canon first appeared after 94 CE in Josephus’s Against Apion, where he refers to “only twenty-two books,” which he subsequently grouped into three sections: five of Moses, thirteen of prophets, and four remaining books (A. Ap. 1.37-42). Around 100 CE, 4 Ezra 14:44-47 numbered the books published openly as twenty-four, and Gos. Thom. 52 in the second half of the second century reveals that “twenty-four prophets have spoken in Israel, and they all spoke of you [i.e., Jesus].” The numbering of these books as twenty-two or twenty-four attests the narrow scope of the Jewish canon at that time and may indicate its closure at an earlier time, but the evidence for the earlier period is limited. In the past, many scholars believed that around 90 CE at a council at Yavneh (or Jamnia) the Jews closed the books that filled the Writings, and therefore, the Jewish tripartite canon that soon became the Rabbinic Bible or the Hebrew Bible was considered

208

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

closed from the second century. However, most scholars now view Yavneh as only an assembly or a school, without authority, that discussed the scriptural status of Ecclesiastes and Song of Songs but made no official ruling. After the research of Lewis (1964), it became untenable to hold that this assembly was the occasion for the closing of the Jewish canon; the date of closing the Writings and therefore the close of the canon could be around 164 BCE (Beckwith 1985: 152) or in the second century CE (McDonald 2017: 484). From the earlier period, six passages of the putative, Jewish tripartite canon consisting of Torah, Nevi’im, and Ketuvim have garnered attention (Prol. Sir. 1-2, 8-10, 25-25; 4QMMT C 10-11; 2 Macc. 2:13-15; Philo Vit. Cont. 25; Lk. 24:44; Josephus Ag. Ap. 1.39-40). However, the exact contents of the Nevi’im and its relationship to the third section, which still had no definite title by c. 130 BCE, are still open to debate. John Barton (1986: 39–41) cautions against an entirely clear line separating “Prophets from Writings around the turn of the era.” More than uncertainty of the contents within these sections, scholars now question whether any of these passages refer to the tripartite Jewish canon that would later be known from B. Bat. 14b, and therefore, they call the relative antiquity of the tripartite canon into question (McDonald, 2018). Some scholars still see a tripartite canon referenced in the three relevant places in the Prologue to Sirach, though with less certainty in the other places (Steinberg and Stone 2015: 11–15). Stephen Dempster (2008: 116–17) notes the possible division between Prophets and Writings in 2 Macc. 2:13-14. Other scholars believe that the quadripartite structure of Law, History, Wisdom, and Prophets, known primarily from later Greek manuscripts, is the earliest structure of the Jewish canon, thereby suggesting that the tripartite structure of the canon is the more recent revision (McDonald, 2018). Gilles Dorival (2003: 92) suggests that those church fathers who transmit a quadripartite canon may have been influenced by Jews who still used such a canonical structure, which may also reflect the earlier passages of 2 Macc. 2:13-15, 4QMMT C, and Philo Vit. Cont. 25, which divided religious literature into four parts as well. These six passages, along with the many bipartite references, indicate a corpus of scripture with inviolable authority present within the diversity of Judaism, but they do not identify specific books (but see Hanhart 2002: 3). If the bipartite reference, “the Law and the Prophets,” as Barton (1986: 44–55) posits, refers to Torah and nonTorah scripture, then the corpus of Scripture could still be quite wide. The debate over whether the compositions of the Second Temple period are proof of scripture proliferation (Mroczek 2016) or evidence of “revelatory exegesis” or modified modes of revelation (Jassen 2016: 368) continues unabated. In any case, VanderKam (2012: 55) concludes that there was “a limited set of books that was a functional collection of authoritative texts” on which Jewish groups could agree. Based on Qumran, Philo, NT, and Josephus, there probably was a Jewish canon in the late Second Temple period that consisted of the Torah, Prophets, Psalms, and other books. That is, a strong core canon had already been formed but there probably was some uncertainty over some books at the edges as the varied reception of Esther and as the rabbinic statements on Ecclesiastes, Song of Songs, and Sirach show. Thus, the rabbis were defending a more-or-less defined canon (with some fuzziness around the edges) that they had received from the pre-70 period (Alexander 2007: 65). Researchers must confess that the paucity of evidence means that not all their questions of this early period will be answered.

THE SEPTUAGINT AND THE BIBLICAL CANON 209

This same paucity of evidence also poses a challenge for proving that Jews in Alexandria fixed a canon containing more books than Josephus’s exclusive twentytwo books, which he claims all Jews everywhere regard (for the “Alexandrian Canon Hypothesis,” see Swete 1900: 197). Since the work of Albert Sundberg (1964: 51), most scholars have abandoned this hypothesis: “It is to be remembered that the hypothesis is built upon what is, at best, a tenuous foundation and is … incapable of proof.” Philo of Alexandria (c. 20 BCE–40 CE) cited the Pentateuch numerous times and by comparison made minimal use of many of the other books that Josephus would later include in his twenty-two. Thus, some conclude that Philo esteemed only the Pentateuch as Scripture and the rest of the books as tradition (Lim 2013: 91, 183), while others maintain that Philo essentially held to the “the Law and the Prophets” (Chapman 2000: 271–3). Philo does not cite deuterocanonical literature as scripture. As presented in Table 14.1, Biblindex shows that Philo used, though did not necessarily cite from, the following.1 Although there were forty combined references to Wisdom and Sirach, only one or two of these references had verbal correspondence of two words (e.g., Wis. 4:8 in Heir 290). Most instances shared a similar theme but had no verbal parallels to suggest Philo was dependent on Wisdom or Sirach. We can contrast Philo’s use of these two wisdom books with his use of Prov. 3:4 in On Drunkenness 84, which he introduces, “It seems also good to me to say what is in the Proverbs.” In the same section, he follows this quotation with another citation from Prov. 4:3. In On the Preliminary Studies 177, he cites Prov. 3:11-12 introducing it, “Therefore, I think, did one of Moses’ disciples, who is named a man of peace, which is in our ancestral tongue Solomon, say.” Philo shows great reverence for a book of Solomon’s like Proverbs. In On the Change of Names 48, Philo quotes Job 14:4-5 by introducing it, “as Job says.” This citation aids his interpretation of Gen. 17:1 in the preceding context, thereby indicating Job’s authority. Furthermore, the New Testament offers no evidence for the Alexandrian canon. Thus, if this canon existed, there is no early support for it, and we must wait for the fourth- and fifth-century magisterial codices to which we will return below. Recently, some scholars have attempted to resuscitate and revise the Jewish Alexandrian Canon hypothesis by focusing on a Jewish Greek corpus or the Septuagint Canon. Jan Table 14.1 Philo’s Citation of Old Testament and Deuterocanonical Literature Joshua

4×

Psalms

42×

Judges

2×

Job

5×

1 Samuel

13×

Proverbs

30×

1-2 Kings

6×

Ecclesiastes

2×

Isaiah

23×

Esther

1×

Jeremiah

10×

1 Chronicles

7×

Ezekiel

6×

Wisdom

29×

Hosea

3×

Sirach

11×

Zechariah

1×

Available online: http://www.biblindex.mom.fr/ (accessed March 13, 2020).

1

210

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Joosten (2016b: 690–1, 698) attempts to define a Septuagint Canon according to the following criteria: (1) a Greek composition; (2) a coherent corpus; and (3) a historical backdrop in the Jewish diaspora of Egypt. However, Joosten must finally appeal to the codices, “The Septuagint canon as it appears to us, with some inner diversity, in the earliest biblical codices is distinct from the rabbinic canon attested in the Masoretic text.” Similarly, Philippe Guillaume (2004) argues that the Palestinian Nevi’im (Prophets) was a distinct collection of books formed in response to an earlier Alexandrian “Biblical Chronology” (Jewish history books plus Judith and Tobit), which became the basis for the rest of the historical books in the Septuagint to which the Poetic and Prophetic books were added. The Christians inherited these lists from Alexandrian Jews and formed the Septuagint.

Old Testament Canon Lists The Greek canon lists present the clearest evidence of the situation (Gallagher and Meade 2017: 70–173). From around 100–400 CE, there are twelve OT canon lists (see Table 14.2). In approximate chronological order, they are as follows: ●●

The Bryennios List (c. 100–150)

●●

Melito of Sardis (c. 170)

●●

Origen (c. 220; apud Eusebius)

●●

Cyril of Jerusalem (c. 350)

●●

Athanasius (367)

●●

The Synod of Laodicea (before 380)

●●

Apostolic Canons (c. 380)

●●

Amphilochius of Iconium (c. 380)

●●

Gregory of Nazianzus (381–90)

●●

Epiphanius (three lists; Panarion in c. 375; On Weights and Measures in 392)

After 400, there are at least seven more Greek canon lists of the OT up to c. 850 (see Table 14.3): ●●

The Hypomnestikon or Joseph’s Bible Notes (c. 393–431; Menzies 1996: 86–7)

●●

The Dialogue of Timothy and Aquila (sixth century; Conybeare 1898: 66)

●●

Ps-Chrysostom (two lists; c. 500–550; list one: PG 56:313ff; list two: PG 56:317-86)

●●

Ps-Athanasius (c. 500; PG 28:284-9)

●●

John of Damascus (c. 754; PG 94:1180)

●●

Nicephorus I (c. 850; PG 100:1056-60).

Melito

Genesis–Exoddus

Numbers

Leviticus

Deuteronomy

Joshua

Judges

Ruth

1–4 Kingdoms

1–2 Chronicles

Psalms

Proverbs

Ecclesiastes

Song

Job

Isaiah

Jeremiah

The Twelve

Daniel

Ezekiel

Esdras

Bryennios

Genesis–Leviticus

Joshua

Deuteronomy

Numbers

Ruth

Job

Judges

Psalter

1–4 Kingdoms

1–2 Chronicles

Proverbs

Ecclesiastes

Song

Jeremiah

The Twelve

Isaiah

Ezekiel

Daniel

1–2 Esdras

Esther

Table 14.2 Greek Canon Lists c. 100–400

Maccabees

Outside:

Esther

Job

Ezekiel

Daniel

Jeremiah + Lamentations + Epistle of Jeremiah

Isaiah

[The Twelve]

Song

Ecclesiastes

Proverbs

Psalms

1–2 Esdras

1–2 Chronicles

1–4 Kingdoms

Octateuch

Origen

Daniel

Ezekiel

Jeremiah + Baruch + Lamentations + Epistle of Jeremiah

Isaiah

The Twelve

Song

Ecclesiastes

Proverbs

Psalms

Job

Esther

1–2 Esdras

1–2 Chronicles

1–4 Kingdoms

Octateuch

Cyril of Jerusalem

Tobit

Judith

Esther

Sirach

Wisdom

To be read:

Daniel

Ezekiel

Jeremiah + Baruch + Lamentations + Epistle of Jeremiah

Isaiah

The Twelve

Job

Song

Ecclesiastes

Proverbs

Psalms

1–2 Esdras

1–2 Chronicles

1–4 Kingdoms

Octateuch

Athanasius

Daniel

Ezekiel

Jeremiah + Baruch + Lamentations + Epistle of Jeremiah

Isaiah

The Twelve

Job

Song

Ecclesiastes

Proverbs

Psalms (150)

1–2 Esdras

1–2 Chronicles

1–4 Kingdoms

Esther

Octateuch

Laodicea

Proverbs

The Twelve

Psalms (151)

Five of Solomon

The Twelve

Song

Ecclesiastes

Joel

Song

Job

Proverbs

Psalms

Joel

Ecclesiastes

1–4 Maccabees

Micah

Psalms

Judith

Job

1–2 Esdras

Amos

Job

Esther

Micah

1–2 Esdras

1–2 Esdras

1–2 Chronicles

1–4 Kingdoms

Amos

1–2 Chronicles

1–2 Chronicles

Hosea

1–4 Kingdoms

1–4 Kingdoms

Octateuch

Amphilochius

Hosea

Octateuch

Octateuch

16 Prophets

Gregory of Nazianzus

Apostolic Canons

Table 14.2 Continued

The Twelve

Isaiah Jeremiah + Baruch + Lamentations + Epistle of Jeremiah Ezekiel

Ezekiel Daniel

1–2 Esdras

1–4 Kingdoms

1–2 Chronicles

Judges + Ruth

Joshua

Song

Ecclesiastes

Proverbs

Psalter

Job

Genesis–Deuteronomy

Epiphanius 2

Jeremiah + Baruch + Lamentations + Epistle of Jeremiah

Isaiah

The Twelve

1–2 Chronicles

1–4 Kingdoms

Song

Ecclesiastes

Proverbs

Psalter

Job

Octateuch

Epiphanius 1

Jeremiah + Lamentations

Isaiah

The Twelve

Song

Ecclesiastes

Proverbs

1–4 Kingdoms

1–2 Chronicles

Psalter

Ruth

Judges

Job

Joshua

Gen.–Deut.

Epiphanius 3

Ezekiel Daniel

Ezekiel

Daniel ? Esther ?

Jeremiah

Jeremiah

Malachi Isaiah

Isaiah

Malachi

Wisdom Sirach

Sirach Wisdom

Haggai Zechariah

Haggai

Zephaniah

Zephaniah

Zechariah

Useful + Beneficial:

Disputed:

Habakkuk

Habakkuk

Esther

Esther

1–2 Esdras

Daniel

1–2 Esdras

Nahum

Jonah

Nahum

Ezekiel

Daniel

Esther

Epiphanius 3

Epiphanius 2

Epiphanius 1

Obadiah

Obadiah

Amphilochius

Sirach

Jonah

Gregory of Nazianzus

Outside:

Apostolic Canons

Table 14.2 Continued

Isaiah

Jeremiah

Ezekiel

Daniel

Esdras

Judith

Esther

Twelve

Isaiah

Jeremiah

Ezekiel

Daniel

Job

Sirach

Wisdom

The Twelve

Song of Songs

Maccabees

Nahum

Haggai

Ecclesiastes with the Songs

Ecclesiastes

Tobit

Proverbs of Solomon

Proverbs

Apocrypha:

Zephaniah

Psalter of David

Psalms

Esther

Micah

Habakkuk

Job

Outside:

Jonah

Nahum

1–4 Kingdoms

Esdras

David (= Ps.)

Job

16 Prophets

Song

Ecclesiastes

Sirach

Proverbs

Esdras

Malachi

Zechariah

Micah

Jonah

Obadiah

Amos

Joel

Hosea

Daniel

Ezekiel

Jeremiah

Isaiah

Sirach

1–4 Kingdoms

Isaiah

Malachi

Zechariah

Haggai

Zephaniah

Habakkuk

Obadiah

Amos

Joel

Hosea

Twelve

Job

Song

Ecclesiastes

Proverbs

Psalter (151)

1–2 Esdras

1–2 Chronicles

1–4 Kingdoms

Octateuch

1–2 Chronicles

1–4 Kingdoms

Octateuch

1–2 Chronicles

Octateuch

Octateuch

Ps.-Athanasius

1–4 Kingdoms

Ps.-Chrysostom 2

Octateuch

Ps.-Chrysostom 1

Dialogue of Timothy + Aquila

Hypomnestikon

Table 14.3 Greek Canon Lists c. 401–850

Table 14.3 Continued

Tobit

Judith

Esther

Sirach

Wisdom

To be read:

Daniel

Ezekiel

Jeremiah (+ Baruch + Lamentations + Epistle of Jeremiah)

Ps.-Athanasius

Table 14.3 Continued

1–3 Maccabees Wisdom

Wisdom Sirach

Tobit and Tobias

Susanna

Judith

Esther

Psalms + Odes of Solomon

Sirach

The Twelve Antilegomena:

Esther

Daniel

Virtuous + Beautiful:

Ezekiel

Daniel

Baruch

Ezekiel 1–2 Esdras

Isaiah Jeremiah

Isaiah

Job

Jeremiah

Song

Song Twelve

Proverbs Ecclesiastes

Proverbs Ecclesiastes

1–2 Esdras Psalter (151)

Job Psalter

1–4 Kingdoms 1–2 Chronicles

1–4 Kingdoms

Octateuch

Octateuch 1–2 Chronicles

Nicephorus I

John of Damascus

THE SEPTUAGINT AND THE BIBLICAL CANON 217

The Contents of the Lists The twelve Greek lists from the early period show remarkable unity in their contents, which largely overlap with the traditional Hebrew canon of twenty-two books. That is, in no uniform order, these lists include the five books of the Law of Moses, the Former and Latter Prophets, and the five Poetic books. Esther was normally included, but some lists— such as Melito, Athanasius, and Gregory—still excluded this book probably because of the earlier disputes over it. Jeremiah regularly encompassed not only Lamentations but also Baruch and the Epistle of Jeremiah. The title “Daniel” usually subsumed the works called Susanna and Bel and the Dragon. 1 Esdras is usually joined by the variant literary edition of 2 Esdras, which equals Ezra-Nehemiah. Helpfully, Joosten (2016b: 699) suggests that the relationship between 1 and 2 Esdras is comparable to OG Daniel and “Theodotion” Daniel, as two versions of the same book. Furthermore, most lists do not comment on whether the Psalter contained 150 or 151 Psalms (though see Laodicea [150] and Apostolic Canons [151]) or which forms of Esther and Daniel were intended. The ancients probably viewed these differences with the Jewish canon as matters of textual form and not of canon, which, as we will see, they consistently referred to as the Hebrew canon. The Africanus-Origen correspondence and later Jerome highlight differences of textual form between Jews and Christians, but early Christians believed their canon was equal to the Jewish canon. Of these twelve lists, only the Apostolic Canons, which has a complex textual history, includes the deuterocanonical books, but even some of its Greek witnesses preserve the narrower list of books. The seven later lists from c. 401 to 850 largely reflect the earlier ones (see Table 14.3). John of Damascus, following Epiphanius, includes Esther and does not include the deuterocanonical books (cf. Epiphanius De Mens. et Pond. 4-5). Epiphanius had Law-Poetry-Writings-Prophecy-Remaining. John has Law-Writings-Poetry-ProphecyRemaining, thus transposing the sections of Poetry and Writings. The Hypomnestikon and Ps.-Athanasius omit Esther and do not include deuterocanonical books. Ps.-Athanasius, however, does report an alternative numbering of the canonical books where “some of the ancients” say that Ruth is joined to Judges and Esther is included to maintain the twenty-two books of the Hebrews. The Dialogue of Timothy and Aquila includes Judith. Nicephorus excludes Esther and is the first to include Baruch independently of Jeremiah. Francesca Barone is undertaking fresh textual work on the Synopsis Scripturae Sacrae of Ps.-Chrysostom (Barone 2009). So far, her work has revealed that all the deuterocanonical books in Ps.-Chrystosom 2 included in Migne, except for Sirach, should be removed from the reconstructed archetype of this work. Therefore, this source reveals a list that includes Sirach but also omitted 1 Chronicles–Judith, Job, Wisdom, Proverbs, Ecclesiastes, Song of Songs, and Psalms, probably showing the incompleteness of the archetype. Perhaps, the original work contained these books, and they were omitted accidentally through transmission. The original author may have considered more books canonical than he finally commented on and listed. Barone’s edition of Ps.Chrysostom 1 will have the reading ὁ ἰώβ (“Job”), instead of Ruth, which is the reading in Migne’s edition. In summary, these nineteen Greek canon lists from 100 to 850 provide the most specific information on the contents of the Greek OT, and significantly, none of these lists represents what scholars today would refer to as the Jewish Alexandrian Canon or the Septuagint Canon. There is remarkable consistency across them, but minimal variation does occur over the status of Esther or a deuterocanonical book here or there. The cause of this consistency appears to be the Hebrew canon and to this matter we now turn.

218

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

The Significance of the Hebrew Canon Of these nineteen canon lists from around 100 to 850, thirteen mention explicitly or imply that their twenty-two books correspond to the twenty-two letters of the Hebrew alphabet and consequently the twenty-two books of the Jewish canon. Interestingly, Hilary of Poitiers, Jerome of Stridon, and Rufinus of Aquilaea also preserve the connection between the Hebrew alphabet and the Christian Old Testament in Latin (Gallagher and Meade 2017: 194–222). Those who mention or imply the connection between the Hebrew alphabet and the canon are as follows: ●●

Origen (apud Eusebius)

●●

Cyril of Jerusalem (implied)

●●

Athanasius

●●

Synod of Laodicea (implied)

●●

Gregory of Nazianzus

●●

Epiphanius (three lists)

●●

The Hypomnestikon (implied)

●●

Dialogue of Timothy and Aquila

●●

Ps-Athanasius

●●

John of Damascus

●●

Nicephorus I (implied)

These early Greek Christians believed that their twenty-two-book canon was formed consciously according to the twenty-two-book Hebrew canon, which corresponded to the twenty-two letters of the Hebrew alphabet. They mostly included books that would correspond with the contents of Josephus and B. Bat. 14b. However, mere mention of the Hebrew alphabet did not mean that these canon lists equaled Baba Batra and the later Hebrew codices in every detail as described above. But early patristic canon theory with its focus on the Hebrew alphabet and the twenty-two books of the Hebrews ensured that early Christians would form their OT according to the Jewish canon though with slight variation at the edges of that canon, probably due to the earlier Jewish disputes over the book of Esther (e.g., b. Meg. 7a). Albert Sundberg (1964: 146) supposed that the Eastern Christian fixation with the Hebrew canon developed in the fourth century and not before. However, the Bryennios List and Melito’s list from the second century show that Christians had already accepted the more limited Jewish canon by excluding the deuterocanonical books except for a possible, but improbable, mention of the Wisdom of Solomon in Melito. Although debated, Origen’s canon list probably does not represent his canon list for the church (see Gallagher 2012: 37–8). Therefore, the later fourth-century lists are in continuity with the  earliest Greek Old Testament that was formed consciously after the Jewish canon in the few centuries prior to the fourth century. They included the books of the Jewish canon in Greek dress and they regularly excluded the deuterocanonical books. Greek Usage of Old Testament Books Another important way to determine which works of religious literature had authoritative status is on the basis of the books Greek authors cited and on which ones they wrote commentaries.

THE SEPTUAGINT AND THE BIBLICAL CANON 219

Citations of Old Testament Books A criterion for determining authoritative scripture is the analysis of formal citations within religious literature. But here we must be careful to avoid a potential pitfall, for even formal citation does not necessarily mean an ancient author reckoned the work as part of his canon. To illustrate this point, we should  consider one of Steve Mason’s conclusions regarding Josephus’s twenty-two books in Ag. Ap. 1.37-42. It is worth citing at length: As we have seen, he [Josephus] also continues his narrative to the present, treating books such as Pseudo-Aristeas and 1 Maccabees the same way he treats biblical material. For the biblical period itself, he splices in all sorts of oral and written traditions. He quite thoroughly alters the texts to suit his own needs. And in numerous ways he evokes a prophetic aura (though not the terminology) for his own accounts. In other words, if we lacked the Against Apion, Josephus himself would offer a clear case for an open canon. But we do have the Against Apion, in which this same Josephus emphatically, but also matter-of-factly, insists that the Judean records have long since been completed in twenty-two volumes. Plainly, then, the circumstantial evidence of Josephus’s own use of the Bible in the Antiquities does not mean what it might otherwise have seemed to mean: it does not, after all, imply an open canon. (2002: 126) Without the clear statement of Josephus’s canon theory in Against Apion, one would conclude from the Antiquities that Josephus’s canon was open and fluid because he uses many works outside of his twenty-two books. But because of the Against Apion, we learn that he conceived of the canon as closed from long ago. With this potential pitfall of citation analysis noted, a summary of the evidence of the NT and the second century should be rehearsed here, followed by initial soundings from the later period. The situation presented in the canon lists mentioned above is also reflected in the earliest citations of the OT in the NT. That is, the NT authors only formally cite as Scripture books contained in the Hebrew twenty-two-book canon, not books that belonged to the putative, wider Septuagint canon. Oskar Skarsaune summarizes the situation: There has been much scholarly debate on the question of whether “circumstantial evidence” (i.e., the actual use of authoritative books) in the first century CE supports or contradicts the notion of a “closed” canon in that period. If quotation frequency is regarded as significant circumstantial evidence, the New Testament seems to indicate that its authors (with the one exception mentioned [1 Enoch in Jude]) quoted the Hebrew canon, and its books only, as Scripture. (1996: 445)2 Furthermore, regarding the second century, Skarsaune concludes that if one excludes Clement of Alexandria, only books from the Hebrew canon are cited formally with two exceptions: Irenaeus quotes from the book of Baruch twice (3:29-4:1; 4:36-5:9) but attributes the citations to Jeremiah. In doing so, Irenaeus shows that he cites from a book of Jeremiah that included Baruch along with Lamentations, the default, but not only, position of early Christians up through the fourth century (Skarsaune 1996: 445–6). This means that for the first two centuries of the Church’s history no deuterocanonical book See also Joosten 2016b: 689, 698; pace Sundburg 1964: 53–5, who sees a wider usage of texts in the NT.

2

220

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

was formally cited as scripture. Formal citation, not simple usage or mere acknowledgment of a book or tradition, is what concerns us here. For example, 1 Clement 55 depends on the stories of Judith and Esther as examples of piety for the church, but Clement does not cite these books as scripture. More work should be done to analyze these kinds of uses of the so-called deuterocanonical literature, but Skarsaune’s point about lack of formal citation of these books still holds. But what about the later period when church fathers cite formally from deuterocanonical books that putatively belong to a Septuagint Canon? Regarding how Athanasius cited canonical books in comparison with the Wisdom of Solomon, Johan Leemans (1997: 368) concluded, “We also saw that, at least with regard to the Book of Wisdom, there is no difference between “canonical books” and books to be read to the catechumens … Athanasius clearly valued the writings of both categories on an equal level.” Although Athanasius’s citations of Wisdom and Sirach reveal esteem for these books, he does not use them or all the deuterocanonical and noncanonical literature combined as much as he used the books in his “canonical” category (Gallagher and Meade 2017: 129 n. 274). Athanasius’s OT canon list mirrored the twenty-two-book Jewish canon with an exception or two (particularly Esther), and he categorized such books as Wisdom of Solomon, Sirach, Judith, and Tobit as “books to be read” in a middle tier between canonical and apocryphal books. He considered the books contained in this middle tier as useful scripture. Therefore, the twenty-two books in Athanasius’s canon list in his thirtyninth Festal Letter do not exhaust all of scripture, and therefore, his scope of scriptural books is wider than his canon list. More research comparing how the fathers use the books in their canon lists (when a father has a list) and books outside of the lists is a desideratum. Many preserved theoretical differences between these books, but research into whether they observed these distinctions in exegetical works such as homilies would be helpful to the field. Furthermore, analysis of scripture usage in fathers who did not transmit a canon list would continue to show the varied approaches to religious literature in this period. Greek Commentaries on Old Testament Books As scholars have noted, several church fathers wrote early Greek commentaries on the canonical books of the Old Testament (Kannengiesser 2004). According to the Göttingen Septuaginta, some of these commentators included the additions to Jeremiah and Daniel. For Jeremiah, Chrysostom (fourth century) does not comment on the additions to the book, while Theodoret of Cyrus (fifth century), Olympiodorus of Alexandria (sixth century), and Basil of Neopatrae (modern Ypati; ninth century) do (Ziegler 2013: 11–17). For Daniel, the situation is more complex. Hippolytus (third century) comments on Susanna and Daniel but does not specially treat Bel and the Dragon (Ziegler 1999: 125– 7). Chrysostom’s comments are preserved in catenae. There are no preserved comments on the longer Daniel 3 (Song 3 Childr.) or Susanna—maybe some comment on Bel 1:1-2—but omissions in catenae are not significant evidence (Ziegler 1999: 128). Only scholia from Polychronius of Apamea’s (fifth-century) commentary remains. At the Song 3 Childr. there is the following note in Mai’s edition, “Bless all. Now it is necessary to know that the hymn itself is not in the Hebrew or Syriac books, for it is said to be added from things having been spoken by some after these” (1825: 113). In the same edition, his comments on Bel 1:1-2 are preserved (Mai 1825: 159–60). Theodoret of Cyrus comments on the Song 3 Childr., while he does not treat Susanna or Bel (Ziegler 1999: 129). Basil of Neopatrae also comments on the additions to Daniel.

THE SEPTUAGINT AND THE BIBLICAL CANON 221

Regarding commentaries on the deuterocanonical books, in the fourteenth century a Malachias Monachus (otherwise unknown) wrote commentaries on Wisdom of Solomon and Sirach (Albrecht 2013: 882–4; for text descriptions, see Ziegler 1980: 12–13, 2016: 12–13). Also, Matthew Kantakouzenos (co-emperor of Byzantium 1353–7; d. 1391) in the fourteenth century wrote a commentary on the Wisdom of Solomon (Albrecht 2012: 212–16; for text description see Ziegler 1980: 13–15). Furthermore, Rahlfs (1914: 424) listed an anonymous commentary on Wisdom from a sixteenth-century manuscript. There appear to be no Greek commentaries written on 1 Esdras, Judith, Tobit, or 1–4 Maccabees. The Greek fathers did cite both canonical and deuterocanonical literature similarly. Many fathers commented on the additions to Jeremiah and Daniel since they were presumed to belong to these books, and in rare cases they provided more extended commentary on parts of deuterocanonical books as Origen commented on Wis. 7:17-21 (Gallagher 2012: 41–2). Otherwise, there appears to be no commentarial literature on the deuterocanonical books in the Greek church before the fourteenth century. However, usage of a book may not necessarily indicate its canonical status, and the reverse is also true: limited usage does not necessarily indicate noncanonical status. For example, 2-3 John were rarely cited as scripture but were included in most of the canon lists. Citations and commentaries on books reveal high esteem for scriptural works, but they may not reveal whether a book was canonical or not (Law 2013: 121–2, also problematizes the presence and absence of early Christian commentaries and citations for determining canonical scripture).

Manuscripts of the Greek Old Testament We turn now to the material evidence of the Greek manuscripts of the Old Testament to see what they reveal about the Septuagint and the biblical canon. Yet, we encounter another problem of method, for the scribes of these manuscripts did not leave comments on their estimation of individual books, and this has led to challenges in interpreting their significance. In Brill’s Textual History of the Bible, Armin Lange (2016b: 74–5) says: I have also referred repeatedly to the invention of the mega-codices as a catalyst in the development of the Christian canon. It was only the mega-codex format that made it possible to combine all biblical Scriptures (Old and New Testament) into one book. Consequently, the question of which books should be included in such a mega-codex as the one Bible will have brought the issue of canon to the attention of Christian thinkers and officials … It is all the more interesting that none of the three preserved mega-codices corresponds precisely in its table of contents with any of the canon lists of the fourth century. Lange concludes that the codex equals the biblical canon, and proceeds to put the codices into tension with the extant canon lists. He is not alone. Jan Joosten similarly (2016b: 688) says: The earliest Christian Bibles, of the fourth and fifth century onward, contain in their Old Testament part several books that are absent from the Hebrew canon. The additional books are not in all codices exactly the same ones, but there is an irreducible core: Sirach, Wisdom of Solomon, Tobit, Judith. The outlines of a distinct Septuagint canon are recognized also in Patristic and synodic lists of the early Christian centuries. The lists present a complex image, however: clearly the early Church knows of the

222

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Hebrew canon with its 22 or 24 books, and several prominent figures, from Origen onwards, argue for the need to adhere to it. At the same time, the larger canon of the great codices is also widely known. Joosten notes the tension between the narrower canon lists and the putative wider canon of the codices, and since he maintains that the contents of a codex correspond to a canon, he considers the lists to present a complex image. Joosten continues his argument for a Septuagint canon according to internal criteria, even as he equates the Septuagint canon with the codex and its wider contents. A more consistent way forward would be to interpret the canon lists as reflecting the canon, while the codex includes canonical books alongside other useful religious and spiritual literature. That is: the contents of a codex do not correspond to the canon. With that caveat, we still must investigate the contents of the full Greek bible manuscripts. Full Greek Bibles and Complete Old Testaments Perhaps, surprisingly, there are only seven full Greek bible manuscripts up to the fifteenth century: B (fourth), S (fourth), A (fifth), Ra 68 (fifteenth; Venice), Ra 106 (fourteenth; Ferrara), Ra 122 (fifteenth; Venice), Ra 130 (twelfth/thirteenth; Vienna). Furthermore, there are two full Old Testament manuscripts: Codex Venetus (eighth; without Psalms) and Ra 46 (eighth/ninth; Paris; without Psalms). Table 14.4 contains a table of the contents of these manuscripts. The three magisterial codices of Sinaiticus, Vaticanus, and Alexandrinus agree on the inclusion of Judith, Tobit, Wisdom, and Sirach, though not on their order and placement. They do not all contain the books of Maccabees with only two including 1 and 4 Maccabees and only Alexandrinus containing 2 and 3 Maccabees. Therefore, the material evidence of the codices is conflicted over the inclusion of the Maccabean literature. But what about the other four books that are included in all three? It is possible the compilers considered these books canonical because they chose to include them. However, we must avoid anachronism and not foist the concreteness of our modern Bible on top of the ancient codex. In the absence of explicit statements from the scribes of these codices, it is more consistent to conclude with most of the canon lists that these books were useful scriptures and were not considered to be authoritative in matters of doctrine. The later minuscule manuscripts continue to exhibit flux on the books of Maccabees. Furthermore, Ra 68 and Ra 122 reveal a tendency to differentiate the canonical books from the deuterocanonical books by placing the latter between the Old and New Testament. That is, the deuterocanonical books are not integrated with the canonical books (cf. Table 14.4). Summary Conclusions Although full Bible and complete Old Testament manuscripts provide a rather full range of religious literature, they probably do not reveal the whole picture. More work still needs to be done on the specific contents of the partial manuscripts that contain the various blocks or sections of OT books to see whether text stabilization occurred in the sections of the Greek OT. There will be potential pitfalls since these manuscripts are intentionally incomplete. The complete manuscripts reveal a great amount of flux in the order of the books and they continue to reveal uncertainty on the books of Maccabees, whether 3 and 4 Maccabees should be included in the so-called Septuagint Canon or not. Furthermore, they do not match Rahlfs’s Septuaginta (1935). They do not include the Psalms of Solomon and not all of them include the Odes with the Psalms. These manuscripts usually contain too many or too few books to match modern conceptions of the “Septuagint Canon.”

THE SEPTUAGINT AND THE BIBLICAL CANON 223

Early Christians cited a wide range of religious literature, frequently not differentiating the way in which they cited from canonical or deuterocanonical works. However, the trifold schema of religious literature—canonical-useful-apocryphal—shows that their conception of scripture does not equal our own. Indeed, these early Christians would probably have differentiated between “canonical” scripture as their authority and “useful” scripture as books having diminished authority in cases of ecclesiastical doctrine (cf.  Athanasius, Ep. fest. 39; Amphilochius, Iambi ad Seleucum 251–60; Epiphanius, De  Mens. et Ponds. 4; Jerome, Praef. in Sol.; and Rufinus, Symb. 36). Thus, they might cite these books as scripture but intend fine distinctions among them. One of the more interesting discoveries in terms of usage was the lack of a Greek commentary on a deuterocanonical book until the fourteenth century, and slightly less interesting the uneven commentary on the additions to Jeremiah and Daniel. In sum, the Greek canon lists limit the canonical books to the twenty-two books of the Hebrew canon with one or two exceptions. They do not support the modern notion of the “Septuagint Canon.”

Printed Editions of the Greek Old Testament How was the Greek Old Testament envisioned once it began to be printed? The printed editions of the Greek Old Testament surveyed here are as follows: the Complutensian Polyglot (completed in 1517; published in 1520), the Aldine (1518/19), the Sixtine (1587), Holmes-Parsons (1798–1827), and Rahlfs (1935). Besides the books of the Jewish Bible, they all contain the additions to Esther, Judith, Tobit, 1-3 Maccabees, Wisdom, Sirach, Baruch and Epistle of Jeremiah, and the additions to Daniel. Except the Complutensian, these editions contain 1 Esdras and Psalm 151. The Sixtine entitles Psalms as “the Book of 150 Psalms” and does not number the final Psalm as 151. The Odes appear only in the Aldine and Rahlfs. Furthermore, Rahlfs adds 4 Maccabees and Psalms of Solomon. Thus, Rahlfs presents the fullest version of the so-called Septuagint canon. In 1517 the Complutensian Polyglot, sponsored by Cardinal Ximénes, was completed. He and the editors of this work used rubrics to identify “the LXX.” For the books of the traditional Jewish canon, the Complutensian presents three columns: a Hebrew text (outside), Jerome’s Latin translation (middle), and the Greek translation of the LXX with an interlinear Latin translation (inside), probably produced by the editors. For books not part of the traditional Jewish canon, there is no Hebrew text but only columns representing a Greek edition with an interlinear Latin translation and either a copy of the Vetus Latina (OL) version or sometimes Jerome’s Latin translation of the book for the middle column. For 3 Maccabees only the Greek version with an interlinear Latin translation is represented (cf. Bogaert 2014: 77–80). The significance of the Complutensian for this subject is in its Prologue and rubrics above each column of the Greek translation. The Prologue describes the books of the Old Testament under the three sections of the Hebrew canon as conveyed by Jerome (his order was Pentateuch Prophets Agiographos): Pentateuch, Agiographos, and Prophets. For the books in each of these sections, Cardinal Ximénes says they supplied the Greek translation of the edition of the Seventy. However, when he comments on the books “outside of the canon” (extra canone) a little later in his preface, he says the editors only provided these books in “Greek writing” (Graecam scripturam) and nowhere mentioned the Seventy in relation to them. The Prologue, therefore, associates the LXX with the books of the Hebrew canon and the books “outside the canon” only with Greek writing.

224

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

The rubrics over the Greek translations throughout the Polyglot demonstrate the point. Normally, the rubric for each Greek book that has a Hebrew text and is part of the traditional Jewish canon says: “Greek Translation of the LXX. with Latin Interpretation.” Normally, the Greek text of the books without a Hebrew text and that are not part of the traditional Jewish canon has a different rubric: “Greek Translation of [book title] with Latin.” That is, the book is not said to come from the Seventy. Unexpectedly, Susanna and Bel and the Dragon do not have a Hebrew column but their Greek translation is attributed to the LXX. This rubric is either an editorial mistake or meant to reflect the enduring debate over these two pericopes. In Table 14.4, the book titles without brackets have a Hebrew text and “LXX” in the rubric over the Greek text, while books in brackets Table 14.4 Select Greek Manuscripts and Printed Editions 4th–16th Vaticanus

Sinaiticus

Alexandrinus

Venetus

Ra 130

Ra 106

Octateuch

Octateuch (Defective)

Octateuch

Octateuch (> Genesis + Exodus)

Octateuch

Octateuch

1–4 Kingdoms

1–4 Kingdoms 1–4 Kingdoms

1–4 Kingdoms 1–4 Kingdoms

1–2 Chronicles

1–2 Chronicles (Defective)

1–2 Chronicles

1–2 Chronicles

1–2 Chronicles

1–2 Chronicles

1–2 Esdras

2 Esdras (Defective)

The Twelve

Esther

1–2 Esdras

1–2 Esdras

Psalms

Isaiah

1–2 Esdras

Esther

Tobit

Proverbs

Esther

Jeremiah + Baruch + Lamentations + Epistle of Jeremiah

Job

Judith

Judith

Ecclesiastes

Tobit

Ezekiel

Proverbs

1–3 Maccabees

Esther

Song

Judith

Daniel

Ecclesiastes

Psalms + Odes

Job

Job

1 Maccabees

Esther

Song

Job

Proverbs

Wisdom

4 Maccabees

Tobit

Wisdom

Proverbs

Ecclesiastes

Sirach

Isaiah

Judith

Sirach

Ecclesiastes

Song

Esther

Jeremiah (Defective)

1–2 Esdras

The Twelve

Song

Wisdom

Judith

The Twelve (Defective)

1–4 Maccabees

Isaiah

Wisdom

Sirach

Ep. Mar.

Jeremiah + Baruch + Lamentations + Epistle of Jeremiah

Sirach

Isaiah

THE SEPTUAGINT AND THE BIBLICAL CANON 225

Hypoth. Psalms

Ezekiel

Isaiah

Jeremiah + Baruch + Lamentations + Epistle of Jeremiah

Tobit

Psalms

Psalms + Odes Susana + Daniel + Bel

Jeremiah + Baruch + Lamentations + Epistle of Jeremiah

Ezekiel

The Twelve

Proverbs

Job

Tobit

Ezekiel

Susanna + Daniel + Bel

Isaiah

Ecclesiastes

Proverbs

Judith

Susanna + Daniel + Bel and the Dragon

The Twelve

Jeremiah + Baruch + Lamentations + Epistle to Jeremiah

Song

Ecclesiastes

1–4 Maccabees

The Twelve

1–2 Maccabees

Ezekiel

Wisdom

Song

Chronographion from Adam to Justinian

Susanna + Daniel + Bel

Sirach

Wisdom

Psalms + Odes

Sirach Ra 68 (= Ra 122)

Ra 46

Complutensian Polyglot

Octateuch

Octateuch

Octateuch

1–4 Kingdoms

1–4 Kingdoms

1–4 Kingdoms

1–2 Chronicles

1–2 Chronicles

1–2 Chronicles

Isaiah

1–2 Esdras

Esdras-Nehemiah

Jeremiah + Baruch + Lamentations + Epistle of Jeremiah

Esther

[Tobit]

Ezekiel

Judith

[Judith]

Susanna + Daniel + Bel

1–4 Maccabees

Esther [Add. Est.]

The Twelve

Tobit

Job

Job

Isaiah

Psalter 150

Psalms

Jeremiah + Baruch + Lamentations + Epistle of Jeremiah

Proverbs

Proverbs

Ezekiel

Ecclesiastes

226

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Ecclesiastes

Susanna + Daniel + Bel

Song

1–2 Esdras

The Twelve

[Wisdom]

Esther

Proverbs

[Sirach]

Wisdom

Ecclesiastes

Isaiah

Sirach

Song

Jeremiah

Judith

Job

Lamentations

Tobit

Wisdom

[Baruch]

1–3 Maccabees

Sirach

Ezekiel

(1 Maccabees frag. in 122)

Daniel [Dan.+] The Twelve [1–3 Maccabees]

do not have a Hebrew text or “LXX” in the rubric over the Greek text. That is: the books, which are not part of the traditional Jewish canon and do not have a Hebrew text, are not considered part of the original scope of the “Septuagint canon” according to the Complutensian rubrics with the exception of Susanna and Bel and the Dragon. Having focused mainly on evidence that defines the Old Testament canon according to the Hebrew canon, we still should show how this evidence intersects with the perceived translation work of the Seventy(-Two).

VIEWS OF THE SCOPE OF THE OEUVRE OF THE “SEVENTY(-TWO)” The scope of the work encompassed by the term “Septuagint” has been notoriously difficult to identify, mostly due to modern usage of the phrase “Rahlfs’s Septuaginta” in particular. Probably, all the books of the Jewish canon were translated into Greek by the first century CE with most of them translated by 130 BCE as noted in the Prologue to Sirach. Regarding the work of the Seventy(-Two) translators, Jewish sources limit the scope to the books of Moses (cf. The Letter of Aristeas; Aristobulus, frg. 3; Philo, Vit. Mos. 2.25-44; Josephus, Ant. 12.48ff; and the Rabbis [e.g., b. Meg. 9a–b]). Early Christians, on the other hand, beginning with Justin Martyr (d. c. 165; Dial. 68 et al.), extend the work of the Seventy to encompass the whole Old Testament (see Gallagher 2012: 94). Evidently, when Christians read the term νόμος (“law”) in Aristeas, they understood it to refer to the entire Old Testament (Gallagher 2012: 93 n. 92). Cyril of Jerusalem (Cat. 4.33) and Epiphanius of Salamis (De Mens. et Pontds. 4-5) state that the SeventyTwo translated all twenty-two books of the Old Testament. Cyril and Epiphanius are important since they list the books of the Hebrew canon and attribute their translation into Greek to the Seventy-Two. Hilary attributed the translation of the Psalter to the original translation of the Seventy under King Ptolemy (which one is not specified), and Theodoret referenced the Seventy as translating the Psalm superscriptions as they did the rest of the Scriptures (Gallagher 2012: 95). In the sixth century, the Dialogue of Timothy and Aquila (Conybeare 1898: 66) lists the twenty-two books of the Old

THE SEPTUAGINT AND THE BIBLICAL CANON 227

Testament canon and then adds, “For the Seventy-Two translators transmitted for us Tobit, the Wisdom of Solomon, and the Wisdom of Jesus, son of Sirach as the apocrypha.” Apparently, the author thought that the Seventy-Two translated these books, including Wisdom of Solomon, from a Hebrew source but also separated them from the twenty-two books (cf. also Epiphanius, De Mens. et Ponds. 5 for the view that the Seventy-Two also translated seventy-two [anonymous] books of the apocrypha in addition to the twenty-two canonical books). At the end of the fourth century, Jerome and Augustine would have a debate over this matter (Gallagher 2012: 98–103; Law 2013: 151–66). Jerome would continue to define the scope of the translation work of the Seventy(-Two) along the lines already presented, though he would employ the Hebrew criterion more strictly, while Augustine would affirm the Septuagint as the Bible of the Church (Doctr. chr. 2.22; Civ. 18.43), including the six deuterocanonical books as though part of the translation. Augustine seems to know that the Maccabean literature comes after the time of Ptolemy (Civ. 18.36), even though he includes these books in the wider canon and attributes it to the Septuagint (on the complexity of Augustine’s view, see Gallagher 2012: 101–3). Moving to 1517, the scope of the “Septuagint canon” according to the Complutensian Polyglot consists only of books, and even sections within Esther and Daniel, that have a Hebrew source; that is, the editors of the Complutensian thought that the Septuagint Canon was equivalent to the Hebrew Canon. All other books or sections of books in Greek without a Hebrew text were not considered to be part of the work of the Septuagint. Not only does this approach echo Jerome, it also echoes most ancient Greek patristic biblical theory as well, which held that the Seventy(-Two) translated the books of the Hebrew canon. Only Epiphanius in the late fourth century and the Dialogue of Timothy and Aquila in the sixth century attributed the translation of several of the apocryphal books to the Seventy(-Two). In the West, Augustine appears to attribute his wider canon to the activity of the Seventy(-Two) with some chronological difficulty over the books of Maccabees. Up to the Complutensian, the scope of the Septuagint appears to be defined rather closely according to the Hebrew text. After 1517, the rest of the printed editions up to Rahlfs’s (1935) edition include significantly more books under the title “Septuaginta.”

ONGOING RESEARCH QUESTIONS Scholars continue to ask how our sources relate the Septuagint to the matter of canon. If one begins with Rahlfs’s Septuaginta (1935), the scope of the Septuagint canon will be wider than in any manuscript, testimony, and patristic canon list. On the other hand, if one begins with the collective of the three magisterial codices from the fourth and fifth centuries with their four extra books of Judith, Tobit, Sirach, and Wisdom, then that canon will not agree with Greek canon lists due to too many books, or some Latin canon lists due to too few. Therefore, if “Septuagint Canon” stands in opposition to the Hebrew Canon as an alternative canon, one will be unable to interpret the sources consistently. However, the ancient description of the Seventy as translators of the Hebrew text presents a simple solution that explains most of these data. The canon lists and clear statements of biblical  patristic theory affirm that there is only one canon, the twenty-two (or twentyfour) books of the Jews that were translated into Greek. Some patristic testimony says the Seventy(-Two) also translated some apocryphal books. Most of these Greek fathers would have been surprised to learn that their Greek texts did not equal the Hebrew.

228

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Those fathers that did learn this truth employed theories to explain the discrepancies, usually citing translation method or scribal corruption that would preserve the equality of the two texts (Gallagher 2013b: 676). Cardinal Ximénes and his editorial team applied this same solution by identifying the work of the Seventy by recourse to the presence or absence of the Hebrew text. The outliers of this theory are of course Augustine and much of the Western church up  to the Council of Trent (1546). It is not clear whether they would have cited the Seventy(-Two) as the source for their canon or not. What is clear is that many churches accepted the so-called deuterocanonical books early in the West, and therefore Augustine included them in the canon (Doctr. chr. 2.8.12.14-13.29). In the seventeenth century, the Eastern Orthodox continued to debate the borders of its canon, moving between the Protestant canon and the Roman Catholic canon (Pentiuc 2014: 126–31).

RECOMMENDED RESOURCES 1. Gallagher and Meade (2017). This book presents the Greek, Latin, and Syriac canon lists up to about 400 CE in original languages and English translation with notes and commentary. 2. Sundberg (1964). This work offers the view that Christians after their split from Judaism formed their own canon before the Jewish canon was closed and therefore included more books than the Jewish canon.

PART FOUR

Reception

230

CHAPTER 15

The Septuagint and Second Temple Judaism BENJAMIN G. WRIGHT III

INTRODUCTION AND OVERVIEW The translations that comprise the Septuagint/Old Greek (LXX/OG) were products of Second Temple or Hellenistic Judaism and, as might be expected, they exerted a significant influence on Greek-speaking Jews in antiquity.1 The origins of most of the works that make up the LXX/OG corpus remain obscure, and the Jewish literary works that transmit different versions of a story of origins concern the translation of the Pentateuch, the Septuagint proper, despite the fact that later Jewish and Christian authors, as well as many modern scholars, use the epithet “Septuagint” in a more general way to refer to the larger corpus of translated and compositional books (such as the Wisdom of Solomon and 2–4 Maccabees), often designated LXX/OG. We encounter the basic story of the Septuagint translation in several ancient Jewish texts: The Letter of Aristeas (the fullest account) and the Jewish writers Aristobulus, Philo, and Josephus (who paraphrases Aristeas).2 While the historical veracity of the basic legend of the translation transmitted in these sources is doubtful, each of these sources constructs a version of the story that reflects what the Septuagint had become in its time rather than how it actually originated.3 In that sense, then, to think in this way about the Septuagint and Second Temple Judaism is to be concerned with its reception history rather than its point of production.4 In this chapter I use the term Septuagint and its abbreviation LXX to refer to the Greek translation of the Pentateuch, the narrow sense of the term. If I intend the larger corpus of collected Jewish-Greek translations (which includes some Greek compositions, such as 2 Maccabees or Wisdom of Solomon), I will use the abbreviation LXX/OG. For arguments defending this use, see Wright (2008). 2 Some rabbinic sources after the Second Temple period transmit an abbreviated version of the story developed in the direction of the miraculous. The earliest account comes from a Baraitha in the Babylonian Talmud, bar. b. Meg. 9 a-b, which tells of seventy-two translators whom King Ptolemy brought together. He did not tell them the reason for their presence, and he put them in separate locations, telling them to translate the Law of Moses. Miraculously their translations all agreed. This retelling takes a neutral stance toward translation from Hebrew to Greek. Other rabbinic sources are decidedly negative (see, e.g., Massekhet Sefer Torah 1:8-9). Christian Church Fathers also transmit more miraculous versions of the story (see the chapter by Gallagher in this volume). For a detailed analysis of the development of the translation legend from antiquity to modernity, which includes the rabbinic material and the church fathers, see Wasserstein and Wasserstein (2006). For the legend in antiquity, especially in rabbinic literature, see also Veltri (1994b). 3 For a fuller assessment of what we do know about the historical origins of the Septuagint, see the chapter by Aitken in this volume. 4 See the chapters by Porter and Hiebert in this volume. 1

232

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

In addition to relating a story of the origins of the Septuagint, all of the Jewish authors who transmit the legend and wrote in Greek also employ the text of the LXX in their work.5 Other Jews writing in Greek who do not report the origin story also depended on and employed the Septuagint text. The parade example of those who do both is Philo of Alexandria, who in his Vit. Mos. 2 offers a version of the origination story and whose detailed allegorical interpretations of the text assume the independent and equal status of the LXX to the Hebrew text. Others, such as Demetrius the Chronographer and Ezekiel the Tragedian, worked long before Philo knew and used the LXX, even if they do not report its origin story.

The Origin Story of the LXX The Letter of Aristeas The fullest surviving version of the Septuagint’s story of origins comes in the pseudepigraphical Letter of Aristeas.6 The narrator, called Aristeas, purports to be a gentile courtier of the Ptolemaic ruler Ptolemy II Philadelphus, whom Ptolemy has tasked with procuring seventy-two Judean scholars from Jerusalem. These men would translate the Hebrew Pentateuch into Greek in order to fulfill Ptolemy’s and Demetrius of Phalerum’s mission of including all the books of the world in the great Alexandrian library (§§9–11). This text, which presents itself as a letter from Aristeas to his brother Philocrates, was clearly composed by a Jew taking the guise of a gentile. The narrative’s 322 paragraphs proceed through the following scenes: Preface (§§1–8) Demetrius of Phalerum and the library project (§§9–11) The liberation of the Jewish slaves (§§12–27) Demetrius’s report to Ptolemy (§§28–34a) Ptolemy’s letter to the Jewish high priest Eleazar (§§34b–40) Eleazar’s reply (§§41-51a) The construction of gifts for the Jerusalem temple (§§51b–82) A description of Jerusalem and its environs (§§83–120) Eleazar’s farewell to the translators (§§121–7) Eleazar’s apologia for the Jewish law (§§128–71) The reception of the translators in Alexandria (§§172–86) Seven symposia held in honor of the translators (§§187–300)T The execution of the translation (§§301–7) The proclamation of the translation (§§§308–16) The translators’ departure (§§317–21) Epilogue (§322) The author of Aristeas, an unknown Jew whom we might call Pseudo-Aristeas, makes the translation of the Jewish law frame the work. He adapts the story of the Exodus as a way of reinforcing his (and his audience’s) identity as Jews in Alexandria who did not Except for the rabbis, who, since they wrote in Aramaic and Hebrew, tended not to cite the text, although there is a rabbinic tradition that the Greek translators introduced certain changes into the text (Wasserstein and Wasserstein 2006: 51–83). 6 For recent book-length treatments of Aristeas, see Wright (2015), Honigman (2003), and Carbonaro (2012). See also Niehoff (2011: 19–37) and Rajak (2009: 24–63). 5

THE SEPTUAGINT AND SECOND TEMPLE JUDAISM 233

need to escape from an oppressive pharaoh. Indeed, Ptolemy II becomes a benevolent pharaoh who facilitates the giving of the law a second time, in Greek, and thus, the Jews have no reason to flee Egypt. The purpose of the work is essentially twofold: “(1) to construct or reinforce a Jewish identity that would provide a solid justification for elite, educated Jews to participate in the larger Hellenistic world of Alexandria as Jews and (2) to offer a myth of origins for the primary basis on which a Jewish identity should be built, the Greek version of the Pentateuch, the Septuagint” (Wright 2015: 66).7 Thus, the LXX—the laws of the Jews given by their lawgiver Moses, according to Aristeas—serves a critical function in Jewish ethnic boundary marking, distinguishing them from Greeks and Egyptians (Moore 2015). At the same time, however, the fictional gentile narrator assures a Jewish audience that gentiles recognize their shared reverence for the deity and high moral standards, even an implicit monotheism. He reinforces this assurance first with the famous statement in §16 that “These people [i.e., the Jews] revere God, the overseer and creator of all things, whom all, even we, also worship, O King, using different names, Zeus and Dis”; then Eleazar’s allegorizing apologia for the law; and finally the translators’ responses to the king’s questions in the symposia (B. G. Wright 2013, 2015: 69–74). In the formulation of Pseudo-Aristeas, the LXX stands independent of and equal to the Hebrew Torah as containing everything that Moses intended in the law. In fact, Eleazar equates his allegorizing interpretations of Jewish laws in Greek with what Moses meant when he gave them (e.g., §§132, 139, 142, 144, 148, 153). In this way, Eleazar’s speech and thus the Letter of Aristeas participate in what Najman (2003) has called “Mosaic discourse.” Aristobulus The several testimonia about and the five fragments attributed to an Egyptian Jewish writer called Aristobulus have long been the subject of scholarly debate. On the one side are those who think that Aristobulus is an invention and the fragments attributed to him are inauthentic, a position most recently taken by David Wasserstein and Abraham Wasserstein (2006: 27–35). On the other are those scholars who have argued for their  authenticity, the position taken here (see the detailed discussion in Holladay 1995: 45–74). The earliest reference to Aristobulus is likely 2 Macc. 1:10 in the address of the letter: “To Aristobulus, who is of the family of anointed priests, teacher of King Ptolemy, and to the Jews in Egypt.” His writing has not survived other than in fragments attributed to him and transmitted in the Christian Church Fathers Clement of Alexandria (two fragments), Anatolius (one fragment), and Eusebius of Caesarea (four fragments). Of interest in this context is Fragment 3, which both Eusebius and Clement attributed to a work addressed to Ptolemy VI, in which Aristobulus makes three important claims about the LXX/OG: (1) There were translations of excerpts from Exodus to Joshua before Ptolemy II authorized the translation of the Pentateuch; (2) Demetrius of Phalerum is connected with the complete translation of “everything in the law” (the LXX proper); (3) Demetrius “attended to matters relating to these things” during the reign of Ptolemy II Philadelphus (Eusebius, Praep. ev. 13.11b–12.2; Clement of Alexandria, Strom. 1.22.150.1-3; Holladay 1995: 150–61). Honigman (2003) employs the concept of a “charter myth.”

7

234

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

He makes these claims as part of a larger argument that Plato knew and followed the traditions of the Jewish law as had Pythagoras, “who borrowed many things in our traditions.” Two issues stand out in Aristobulus’s report. First, there had been earlier translations than those made under Ptolemy II, and second, Demetrius of Phalerum receives credit for being responsible for the LXX. It is not likely that there were earlier translations as Aristobulus claims. The ultimate goal of his argument about the Greek scriptures, that Plato and Pythagoras relied upon Jewish tradition in their philosophies, necessitated that Jewish “traditions” in the law (τῇ καθ᾽ ἡμᾶς νομοθεσίᾳ, literally “the legislation concerning us”) would have been available long before he knew that the law had been translated under Ptolemy II, an event that he uses as something of a touchstone. His brief notice of the LXX being translated under Demetrius raises the question of whether there is any literary relationship between Aristobulus and the Letter of Aristeas. In short, an argument for a literary relationship between the two needs to resolve too many problems to be persuasive. It is more likely that Aristobulus and Aristeas used a common tradition or knew two versions of a story that made Demetrius responsible for the translation (Holladay 1995: 64–5; Wright 2015: 28–30). Philo of Alexandria Philo locates his version of the story of the translation of the Pentateuch (Vit. Mos. 2.2544) as part of his biography of Moses, with whom he intimately connects the production of the LXX. Much of the basic narrative he shares with Aristeas, and it seems likely that he knew the earlier work. In his adaptation, however, he essentially produces a new story, one that places the Greek translation side-by-side with the Hebrew text. It is not derivative of it but equal to it, the texts being “sisters,” as he says in Vit. Mos. 2.40, since they both derive from the same source, the unwritten law of nature. For Philo, the Greek text has the same authority as the Hebrew (or “Chaldean,” as Philo calls it) because the two texts are identical, the LXX presents anew the revelation at Sinai, and it is ascribed to Moses. He justifies this construction of the LXX by making several remarkable claims about the production of the text: 1. The translators did their work in an environment that replicated the primordial conditions of creation (Vit. Mos. 2.35-6). 2. The translators did what Moses had done, producing a copy of the law of nature that the Creator had prescribed (Vit. Mos. 2.34-40). 3. The translators “were possessed by an external, divine spirit, which produced utterances through them,” and thus, they were like hierophants, in essence scriptural prophets like Moses, and were inspired interpreters (Vit. Mos. 2.40). 4. The Hebrew and Greek versions of the law are both perfect copies of the law of nature, and “thus they share the same parent and enable whoever accesses them to be instructed in the life of virtue and reason” (throughout Vit. Mos 2.25-40; see also Abr. 3 where Philo treats similar issues). 5. The promulgation of the law in Greek enabled its universalization (Vit. Mos 2.15-24). These basic categories are explored in depth in Najman and Wright (2017: esp. 898–9) in their detailed examination of Philo’s construction of the LXX. With respect to the origination story, Philo dispenses with Ptolemy, Demetrius, and the Library of Alexandria for a loftier justification. Since the Mosaic law had been

THE SEPTUAGINT AND SECOND TEMPLE JUDAISM 235

written in Hebrew, it was unavailable to most of humankind, although the practices of the Jews brought them to the attention of other peoples (Vit. Mos. 2.26-8). “Then it was,” Philo says, “that some people, thinking it a shame that laws should be found in only one half of the human race, the barbarians, and denied altogether to the Greeks, took steps to have them translated” (Vit. Mos. 2.27). Ptolemy II, whom Philo praises as “highly distinguished” (Vit. Mos. 2.30), developed “an ardent affection” for the laws of the Jews, and he brought translators to Alexandria to do the work. Whereas in the Letter of Aristeas the translation work was done cooperatively among the group in the manner of good textual scholars, Philo treats them as in effect little Moseses whom the deity inspired with the spirit of Moses, “the purest of spirits” (Vit. Mos. 2.40) to produce this perfect copy of the law of nature. Philo’s version is not the miraculous product of separate  translators as we see later in some Christian Church Fathers. He instead constructs the translators as having prophetic inspiration, like Moses, rather than producing a scholarly translation as in the Letter of Aristeas. This theoretical grounding works for Philo in that it provides an authority for a Greek scriptural corpus, since for him the life of virtue and reason depends on following the heretofore unwritten law of nature, which now takes written shape in the LXX. It also provides a basis for Philo’s entire hermeneutical project, which relies completely on the text of the LXX, since he claims that the text has a deeper—and universal—meaning that can only be known by allegorical interpretation, which transcends any set of particular laws. For the Letter of Aristeas, the LXX also was independent of the Hebrew and of equal status, but it was a Jewish law meant for Jews, even if its deeper meaning revealed shared values with Greeks. For Philo, the entire justification for translating the law was to bring to all of humankind the universal value of the Mosaic law for living a life of reason and virtue in accordance with the law of nature. Josephus Josephus opens his Antiquities 12 with the beginnings of the Ptolemaic dynasty. Alexander had died, and Ptolemy I had taken Egypt. According to Josephus, the first of the Lagids conquered Jerusalem by entering on the Sabbath pretending to sacrifice and by taking advantage of the Jews due to the nature of the day. Ptolemy I administered a harsh rule in which he took war captives to Egypt (cf. Ep. Arist. §§12–13). Ptolemy II succeeded him, “and he had the Law translated and released from slavery some hundred and twenty thousand natives of Jerusalem, who were slaves in Egypt” (Ant. 12.11). From this point, Josephus paraphrases a good portion of the Letter of Aristeas, giving versions of §§9– 46, §§51–81, §§172–187, §§292–305, and §§308–321. He omits the preface (§§1–8), the translators’ names (§§47–50), the travelogue (§§82–120), Eleazar’s farewell and his allegorical interpretations of the law (§§121–171), the symposia (§§188–291), the mention of ritual washing (§306–7), and the epilogue (§322).8 We know that he used Aristeas not only because his narrative often reproduces the text of Aristeas verbatim but also because he tells his reader in Ant. 12.99-100 that he would not relate the twelve symposia in which all the translators answered the king’s questions. “Anyone who wishes to find out the details of the questions discussed at the banquet,” he writes, “can learn them by reading the book which Aristaios composed on this account.” He immediately turns in Ant. 12.119 to ways that Seleucid kings and Roman rulers honored the Jews. On the basis of these lacunae, some have argued that Josephus’s omissions from Aristeas demonstrate On the way that Josephus treats Aristeas, see Pelletier (1962).

8

236

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

that they were later additions to that work, but these arguments have not been persuasive. Josephus had his own reasons for leaving out those sections (see Honigman 2003: 17–25; Wright 2015: 35–43). Throughout the Antiquities, Josephus’s text reveals that he knew the broader corpus of LXX/OG translations, which comprised some of his source material for his narration of early Israel, although as in the case of Aristeas, he paraphrases and adapts generously. Whereas Aristeas is primarily concerned with the LXX proper and Philo also relegates his exegesis primarily to the Pentateuch, Josephus ranges throughout the corpus.

The Use of the Septuagint in Greek-Jewish Texts Second Temple Jewish authors who wrote in Greek knew and employed the LXX to varying degrees starting from the third century BCE. Outside of Philo and Josephus, however, very few Jewish writers cite the text explicitly. Even the Letter of Aristeas, which given its subject matter might be expected to depend explicitly on the LXX, cites it only once, even though the Greek Pentateuch certainly served as one of the sources for the story, as repeated references to the law and especially to kosher law in Eleazar’s apologia demonstrate. In §155, for instance, Pseudo-Aristeas has Eleazar cite Deut. 7:18, calling it “scripture” (ἡ γραφή), and follows the quote with an explicit allusion to Deut. 10:21. Aristobulus, more than Aristeas, cites directly from the text of the LXX. For example, in his famous explanation that the anthropomorphisms in scripture that attribute limbs to God are to be understood metaphorically, he quotes from Exod. 13:9, 3:20, and 9:3 in succession (Fragment 2; Eusebius, Praep. ev. 8.10.8). Later Aristobulus alludes clearly to Gen. 2:2: “Now, as for what is shown plainly in our code of laws, namely, that God ‘ceased’ working on the Sabbath, this does not, as some suppose, substantiate the view that God no longer does anything” (Fragment 5; Eusebius, Praep. ev. 13.12.11; Holladay 1995: 183). These examples from texts that I discussed above are emblematic of how other Jewish texts in Greek employ the Septuagint. Within the space permitted here it would be impossible to survey every text that utilizes the LXX/OG. In what follows, I will discuss representative examples of how Jewish authors writing in Greek engage the LXX/OG translations. Demetrius the Chronographer In the latter part of the third century, probably in Alexandria, a Jewish historian named Demetrius composed a work meant to resolve difficulties—mostly dealing with chronology and genealogies—in the Jewish scripture.9 We know nothing of his life, and his work does not survive except in six fragments quoted by Eusebius and Clement of Alexandria. Five of the fragments concern the narratives in Genesis and Exodus (Eusebius, Praep. ev. 9), while the sixth concerns the number of years from the destruction of Samaria until Ptolemy IV (Clement, Strom. 1.21.141.1-2). Demetrius is the first Jewish author known to have written in Greek. His sole source is the LXX, which served as the basis for his exegetical solutions, and thus, he is the earliest witness to the translation.

For introduction, Greek text, and English translation, see Holladay (1983a: 51–91).

9

THE SEPTUAGINT AND SECOND TEMPLE JUDAISM 237

The rather brief Fragment 5 illustrates Demetrius’s close reading of the LXX: “But someone asked how the Israelites obtained weapons, since they left (Egypt) unarmed; for they said that after they had gone three days’ journey and had made sacrifices they would return again. It seems, then, that those who did not drown used the weapons of those who did.” In the LXX of Exod. 13:18, the translator read the Hebrew ‫חמׁשים‬, usually translated “equipped for battle” (cf. NRSV), as “fifth generation.” Thus, in the LXX as opposed to the MT, when the Israelites left Egypt they were weaponless, but when they later went to battle against the Amalekites (Exodus 17), they must have had weapons. Where did they acquire them? Demetrius reasons that the Israelites (“those who did not drown”) returned to the sea and took the weapons of the Egyptians who had drowned, and thus, he resolves the textual difficulty.10 Ezekiel the Tragedian Somewhere in the second century BCE, a Jew named Ezekiel wrote a Greek tragedy based on the narrative of the Exodus. As with Demetrius and so many other Jewish writers of the period, Ezekiel’s work does not survive, except in fragments (269 lines) preserved in Eusebius (Praep. ev. 9.28-9), Clement of Alexandria (Strom. 1.23.155.1), and PseudoEustathius (Hexameron, unattributed).11 Ezekiel’s play is the only example from the ancient world of Jewish use of tragedy. Unlike Demetrius, who relied solely on the LXX, Ezekiel knew and was influenced by classical Greek tragedy, especially Euripides. He introduces characters and events that do not appear in the biblical text. Like other Jewish writers in Greek, such as Pseudo-Aristeas, he adapts the LXX text to a Hellenistic genre, in Ezekiel’s case the tragedy, but the LXX is present nonetheless. Holladay (1983b) cites many examples. From the very beginning of the extant lines one can see that the LXX serves as the underlying basis for Ezekiel’s work, both in Ezekiel’s general adherence to the Exodus narrative and in its specific details. As one example, in lines 14–21, Ezekiel writes, Then the mother who bore me hid me three months [cf. Exod. 2:2], as she told me. Unable to escape notice, wrapping me in a covering, she set me out at the edge of the river in a thickly grown marsh [cf. Exod. 2:3]. Miriam my sister kept watch close by [cf. Exod. 2:4]. Then the king’s daughter, together with her maids, came down to clean her youthful skin by bathing. Immediately she saw me, she took me, lifted me from the water [Exod. 2:5].12 The language of this passage, although not a citation of the LXX, follows the narrative closely and uses several terms clearly derived from the Exodus narrative. Artapanus One of the stranger Greek-Jewish texts that survive in fragments is that of Artapanus. Of his life we know nothing, and he is often dated between the mid-third to mid-second century BCE. Three fragments of his work survive, preserved in Eusebius (Praep. ev.  9): a short notice about Abraham, who taught the Egyptian pharaoh astrology, a longer fragment about Joseph and his time in Egypt, and a long fragment treating Moses.

Another argument usually put forward for Demetrius’s use of the LXX is his spelling of names, which agrees with the spellings in the LXX. On the methodological difficulty inherent in this argument, see section III below. 11 For introduction, Greek text, and translation see Holladay (1983a) and Jacobson (1983). 12 Translation is from Holladay (1983a: 349). 10

238

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Clement of Alexandria reports one short episode that overlaps with part of fragment three from Eusebius, attributing it to Artapanus (Strom. 1.23.154.2-3).13 Artapanus produced what John Collins (1985: 891) calls “competitive historiography.” In each case, the Jewish hero, Abraham, Joseph, or Moses, acts as a founder and hero of Egyptian culture. Although the LXX serves as a source for Artapanus, he freely adapts it and introduces numerous characters and events not found there. So, for example, his recounting of Moses bringing the plagues on the Egyptians presupposes the account in LXX Exodus, although he dispenses with the plague of turning the Nile to blood, instead claiming that Moses made the Nile flood and recede, the first time this had happened in Egypt. Among the remarkable achievements of Moses, Artapanus says that he divided Egypt into districts called nomes to which he assigned each its god in the form of cats, dogs, and ibises. Other Writers These three examples do not exhaust the number of Second Temple texts that use the LXX, nor do they fully represent all the ways that the LXX is employed. Other writers, such as Eupolemus, Aristeas the Exegete (not the author of the Letter of Aristeas), and Philo the Epic Poet rely on the LXX as a source. Most of these authors likely come from Alexandria (or at least Egypt), where the LXX was translated. The evidence of all these writers suggests that by the end of the third into the second century, the LXX had indeed become what the author of the Letter of Aristeas later represents it as being, namely an independent replacement for the Hebrew text that held an authoritative status for Greekspeaking Jews in Egypt and likely in other locales as well. Yet, despite its high status, in the third to second centuries BCE we rarely find in this period specific quotations of the text with introductory formulae, unlike in later periods where the LXX/OG will be quoted frequently. Because of the fragmentary nature of many of the Greek-Jewish writers whose extant literary remains do not transmit a story of LXX origins, exactly how they thought of the LXX remains something of a mystery. If we take Aristeas as any sort of general indication, the LXX served these writers as the foundational literature of a people and constituted their ethnic laws and customs. For later writers, such as Philo and Josephus, the LXX had become sacred scripture, so much so that for Jews like Philo each word had potential significance for living a virtuous life in accordance with the law of nature.

DEBATES, THEORIES, AND VIEWS Perhaps the thorniest question facing someone writing about the LXX in Second Temple Judaism is the extent to which the texts that discuss the origins of the translation reflect historical events, since no contemporary testimony exits for the origins of the LXX.14 Most of the attention is given to the Letter of Aristeas, because it narrates a much fuller account than Aristobulus and the accounts in Philo and Josephus largely rely on Aristeas. The historical value of Aristeas has been debated since as long ago as 1684, when the Oxford don Humphrey Hody (1705) disparaged Aristeas as “a suppositious foundling”

For introduction, text, and translation, see Holladay (1983a); introduction and translation in Collins (1985). See also James K. Aitken’s chapter in this volume.

13 14

THE SEPTUAGINT AND SECOND TEMPLE JUDAISM 239

in an attempt to argue that the LXX/OG could not have been inspired scripture.15 While almost all scholars recognize that Aristeas was written in the second century BCE and have noted its improbable elements, those who look to Aristeas for historical clues to the origins of the LXX have taken one of two paths. They claim either that features identified as historically problematic are not really so or that, despite its historical difficulties, Aristeas presents an accurate, or at least plausible, picture that substantiates the origins of the LXX. Perhaps the best example of the first tack is Nina Collins (2000), who tries to show that Demetrius of Phalerum was indeed the librarian under Ptolemy II and thus Aristeas does witness to the LXX actually being a project inaugurated by Ptolemy II. Her argument hinges on establishing that various sources about Demetrius are independent of one another and that they do not rely on the Letter of Aristeas, an argument that cannot be sustained. Thus, they do not provide the historically valuable evidence that Collins finds in them (Wright 2015: 112–16). The scholars who have taken the second tack often focus on slightly different data to try to maintain the historicity of the Letter of Aristeas. They accept as consistent with the Ptolemaic period and ancient practice the claim of the work that from the beginning the LXX had royal involvement and was meant to reside in the Alexandrian Library and/or to serve as an independent scriptural corpus for Greek-speaking Jews. So, for example, Elias Bickerman (1959: 9) simply noted that “the traditional account [i.e., that of Aristeas, Aristobulus, and secondarily Philo] is confirmed by the intrinsic probabilities of the case.” Arie van der Kooij (1998: 220) suggests that what we see in Aristeas fits ancient cultural norms that the men who were appointed to execute the project would be “respected persons, both in the sense of being of noble birth and of being learned.” Moreover, van der Kooij (2007: 299) understands Aristeas’s claim that the translation was destined for the Alexandrian Library to be “natural.” Similarly to van der Kooij, Natalio Fernández Marcos (2009b) considers Aristeas’s portrait of the translation being made for the library as an historical likelihood. In his view only within such a scholarly context as the library could such a project be carried out. Tessa Rajak (2009) reconstructs the impetus for the translation of the LXX as coming from the Jewish community of Alexandria, but she also thinks that royal involvement and placement in the library are likely, the motivation being the prestige of having their law included in this important institution. She concludes, “All in all, the substantial assertions in the Letter of Aristeas are quite consonant with what we know of the early Ptolemaic environment there—the cultural activities of its kings, the breadth of interest of its Greek thinkers, and the position of the Jews on their mental map” (2009: 66). Other scholars have adopted a more dubious view with respect to Aristeas’s claim that the LXX was meant to be sacred scripture from its origins and that it had royal support. Honigman (2003) doubts that the LXX would have been considered scripture, and she calls Aristeas a “charter myth” that offers a justification for regarding it as scripture in a later period. Yet, Honigman also seems to accept Aristeas’s claim about the LXX as a replacement for the Hebrew text. She wants to take seriously the tradition about the library, theorizing that an exemplar of the LXX was placed there in a subsequent period and that the narrative in Aristeas supports seeing that copy as authoritative (2003: 36).

See Wright (2015: 6–7).

15

240

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

In my own work on Aristeas, I have concluded that Aristeas’s narrative is a fiction created to justify the changed status of the LXX from its origins as a translation dependent on its Hebrew parent to one of independence and scriptural status by the latter part of the second century BCE (Wright 2015: 6–15). If we compare the way that Aristeas constructs the LXX with the linguistic character of the LXX itself, we find Aristeas’s narrative contradicted in numerous places. In short, at its point of production, the LXX was meant to serve as a gateway to the Hebrew that remained the authoritative text for Jews. Aristeas witnesses to the reception history of the LXX and constructs a myth of origins for its scriptural status to which the author’s claims of independence, royal involvement, and high literary quality all contribute (Wright 2011, 2015). These conclusions build on the “interlinear paradigm,” the theoretical model that served as the basis for the NETS (New English Translation of the Septuagint) project (Pietersma and Wright 2007: xiii–xx; Hiebert 2010; Ross in this volume). Scholarly debate about these issues surely will continue, however, since they bear on more issues in ancient Judaism than the historical origins of the LXX. Answers to such questions also can do much to help us understand the nature of Greek-speaking Judaism in the Second Temple period—how Jews engaged broader cultural trends in their world, how they understood themselves vis-à-vis that world, and how they then constructed Jewish identities within their Hellenistic context.

METHODOLOGY FOR RESEARCH When thinking about the use of the LXX in Second Temple Judaism, one key methodological question is how to recognize when an author is using the LXX specifically, since, as I noted above, these writers rarely quote it directly. What criteria can determine such dependence on the LXX? This issue pertains especially to the fragmentary Greek-Jewish writers, but texts like Aristeas are not necessarily excepted, since both the fragmentary writers and Aristeas were transmitted in later Christian sources. For Philo and Josephus, who refer to the LXX and OG texts frequently, this problem does not present as difficult an obstacle. One criterion that sometimes gets applied is the orthography of proper names in both the LXX/OG and the Greek-Jewish texts. Holladay (1983a: 56), for instance, uses orthography of names as one piece of evidence to show that Demetrius the Chronographer had the LXX as a source. Yet, we only know Demetrius’s work thirdhand. First a Gentile named Alexander Polyhistor, who flourished in the mid-first century BCE, compiled the texts of numerous Jewish writers like Demetrius in a work entitled “On the Jews.” Alexander’s work itself unfortunately did not survive, but Eusebius of Caesarea knew it and used it in his Preparatio Evangelica in which he explicitly acknowledges that he took citations from Alexander Polyhistor’s work. Within this lengthy process of textual transmission, it is entirely probable (likely?) that later scribes who held the LXX/ OG translations to be sacred scripture would recognize some elements, especially items such as names, from the LXX/OG and harmonize them toward the LXX/OG forms, if they did not already reflect that spelling in the scribes’ exemplars. From a methodological point of view, other factors would need to be taken into account when looking for the use of the LXX in these texts in order to build a compelling cumulative argument. So, in Ezekiel’s tragedy, he follows the narrative of LXX Exodus closely, as we saw above, and uses specific phrases found in the LXX within larger units that might not adhere so closely overall to the language of the LXX. Only when these factors are taken together do they

THE SEPTUAGINT AND SECOND TEMPLE JUDAISM 241

point most convincingly to the LXX as one source that Ezekiel employed. Of course, this caveat applies whenever we study ancient Jewish texts whose primary transmission into modernity have been through later Christian sources that themselves have undergone long histories of transmission. Since these texts cite the LXX/OG so infrequently, they have little text-critical value for the history of the text of the LXX.

ONGOING RESEARCH QUESTIONS Most of the questions that remain open with respect to the Septuagint in Second Temple Judaism have to do with the origins of the Septuagint, especially the social contexts within which the LXX proper originated. But these concerns are the focus of other articles in this volume. Distinguishing between the LXX at its point of production—as well as the translations of non-Pentateuchal books at their points of production, as much as they can be known—from the reception history of these texts will help disentangle issues that have often been mixed together. All of the texts that I have discussed here belong to the reception history of the LXX. As suggested here, they have little bearing on the actual origins of the translations. Focusing on reception history and the place of the LXX/ OG translations in Second Temple Jewish works will shed light on a range of questions that scholars have recently begun to ask about how Jews, especially Greek-speaking Jews in diaspora, positioned themselves and constructed their identities in these contexts. Traditionally, scholars treated the LXX primarily as a witness to the development of the biblical text. But, more and more, as the LXX/OG texts have been recognized as literary products of Second Temple Judaism, as attention has been paid both to how the LXX/OG translations interpreted the Hebrew text and how the LXX/OG translations themselves were translated, scholars are better positioned to incorporate the study of the Septuagint more fully into the study of Second Temple Judaism, although that incorporation is still in its early stages.

RECOMMENDED RESOURCES 1. Rajak (2009). A monograph treating a broad range of issues with respect to the LXX in ancient Judaism, including Rajak’s assessment of LXX origins, the LXX and culture in Alexandria, the use of scripture in Hellenistic Judaism, and the LXX and Jewish identity. 2. Wright (2015). The most recent full-length commentary on Aristeas and the first in English, containing an extended introduction, bibliography, English translation, textual notes, general commentary, and comments on specific words and phrases.

242

CHAPTER 16

The Septuagint in the New Testament STEVE MOYISE

INTRODUCTION It is not an exaggeration to say that without the Septuagint, there would be no New Testament. Its influence is seen in its three hundred or more explicit quotations, along with over fifteen hundred allusions and verbal parallels. It can also be seen in specific concepts, such as “glory” and “covenant,” as well as the high frequency of words like κύριος (“Lord”) and Χριστός (“anointed one”). Less appreciated, perhaps, is the value of the New Testament for textual criticism of the Septuagint. It is true that the New Testament authors sometimes modified the text of their sources to facilitate application and that the scribes who later transmitted the texts (NT and LXX) sometimes sought to harmonize them. Nevertheless, there are still plenty of occasions where the New Testament is a valuable witness to the Septuagint text as it existed in the first century, as well as a witness to some of its various revisions and recensions.

Quotations The fifth edition of The Greek New Testament (United Bible Societies) lists 316 quotations of the Old Testament in the New Testament, taken from 266 different verses (Aland et al., 2014: 857–63). The most popular books cited are Psalms (58×), Isaiah (55×), Deuteronomy (36×), and Genesis (26×). Although over half of these quotations are cases where the OG is in substantial agreement with the MT (making it difficult to tell which source is being used), most scholars are convinced that the OG was the predominant source of Scripture for the New Testament authors.1 This is supported by three lines of evidence. First, there are examples where the OG has a substantially different meaning to the MT. Now we know from the Dead Sea Scrolls that the Hebrew text existed in a number of different forms and so it is possible that the NT author is quoting (and translating) from one of these. But unless there is specific evidence for this, it is more probable that the NT author is quoting from the OG, especially if the agreement is exact. For example, in Amos 9:11-12, God promises to “raise up the booth of David that is fallen” so that Israel will “possess the remnant of Edom and all the nations who are called

Along with many scholars, I use the term Old Greek (OG) in preference to LXX, which properly refers only to the initial translation of the Pentateuch.

1

244

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

by my name.”2 The OG does not refer to Edom and has a rather more positive role for non-Jews, saying instead that God would act “in order that those remaining of humans and all the nations upon whom my name has been called might seek out me” (NETS, emphasis added). This translation is either based on a different Hebrew text, or ‫ירׁש‬ (“possess”) has been read as ‫“( דרׁש‬seek out”) and ‫“( אדום‬Edom”) as ‫“( אדם‬humans”), with the latter being made the subject of the former. This reading is the form quoted by James (in Jerusalem!) in Acts 15:16-18 to support the inclusion of the Gentiles into the people of God (McLay 2003: 17–23). As a second example, Isa. 11:10 says: “On that day the root of Jesse shall stand as a signal to the peoples; the nations shall inquire of him, and his dwelling shall be glorious.” The OG differs in three ways, all of which are followed by Paul verbatim in Rom. 15:12. First, the signal (‫—)נס‬elsewhere rendered by σημεῖον (“sign”) in Isa. 11:12, σημαία (“military standard”) in Isa. 30:17 and χάραξ (“pointed stake”) in Isa. 31:9—is here rendered by the verb ἄρχω (“to rule”). Ross Wagner (2002: 322) suggests this reading might have arisen because of the similarity between ‫ נס‬and ‫“( נסיך‬leader, prince,” rendered by ἄρχων in Josh. 13:21). Second, instead of the nations “inquiring” (‫ )דרׁש‬of this figure, the OG has them “hoping” (ἐλπίζω) in him. Third, instead of a distinction between “peoples” and “nations,” the OG has ἔθνος (“nation”) for both. The result of this translation is followed word for word by Paul in Rom. 15:12: “The root of Jesse shall come, the one who rises to rule the Gentiles; in him the Gentiles shall hope.” Not quite as clear-cut are examples where the OG offers a plausible but idiosyncratic rendering of the Hebrew. For example, the Hebrew of Isa. 40:13 reads, “Who has directed the spirit of the LORD, or as his counsellor has instructed him?” but the OG has “mind” (νοῦς) instead of “spirit” (‫ )רוח‬and “know” (γινώσκω) instead of “directed” (‫)תכן‬. There is not a huge difference in meaning but in Rom. 11:34, Paul also uses νοῦς and γινώσκω.3 Similarly, Prov. 3:34 uses a cognate noun and verb to say that God is “scornful to the scorners” (‫ )ללצים הוא־יליץ‬but the OG uses the noun ὑπερήφανος (“proud”) and the verb ἀντιτάσσω (“oppose, resist”). The resulting text, which reads, “The Lord resists the arrogant, but he gives grace to the humble” (NETS) is followed in 1 Pet. 5:5, apart from θεός substituted for OG κύριος. A second line of evidence is when the OG has additional words, known as a “plus,” compared with the MT. Thus in the dialogue following Moses’ killing of an Egyptian, Exod. 2:14 has the man ask: “Do you mean to kill me as you killed the Egyptian?” But the OG has an additional word, the adverb ἐχθές (“yesterday”), which also occurs in Stephen’s speech in Acts 7:28. Ps. 118:6 proclaims: “With the LORD on my side I do not fear.” The OG has the additional word βοηθός (“helper”), which is also present in Heb. 13:6. Perhaps the most interesting example is Paul’s use of Isa. 52:5 in Rom. 2:24. Paul wants a text that shows that those who are “physically uncircumcised but keep the law” will condemn Jews who “have the written code and circumcision but break the law” (Rom. 2:27). He finds it in the OG of Isa. 52:5, which states, “Because of you, my name is continually blasphemed among the nations” (NETS). But the MT does not have the incriminating “because of you” (δι᾽ ὑμᾶς) or indeed the context of “among the nations” (ἐν τοῖς ἔθνεσιν).4

Biblical quotations are taken from the New Revised Standard Version unless otherwise stated. The OG has also introduced the idea of “becoming” (γίνομαι) God’s counsellor, which is also followed by Paul but absent from the MT. 4 One might think there would be a corresponding category of “minuses,” where the OG omits words from the Hebrew but there are only a few examples among the quotations (e.g., Isa. 61:1 in Lk. 4:18) and these are easily explainable.

2 3

THE SEPTUAGINT IN THE NEW TESTAMENT 245

A third line of evidence is when the NT quotation is long. Given the variety of ways that a Hebrew text might be rendered into Greek (contrast the differences between the versions known as Aquila, Symmachus, and Theodotion when available), the presence of long quotations in the NT that agree exactly or nearly exactly with the OG is strong evidence for the use of the OG by the NT author. Pride of place for this category goes to the quotation of Jer. 31 (OG 38):31–34 in Heb. 8:8-12. The NT quotation runs to 131 words long with only a few minor differences, which include λέγει for φησὶν, ἐπιγράψω for γράψω, συντελέσω and ἐποίησα for the first two occurrences of διαθήσομαι, and the omission of δώσω and αὐτῶν. There are also two agreements in the NT quotation with the OG against the MT. First, ἠμέλησα αὐτῶν (“I was unconcerned for them”) rather than a translation of ‫“( בעלתי בם‬though I was their husband”). In this case the MT is an odd reading and thus the OG may be evidence of a different Hebrew Vorlage. Second, both the OG and Hebrews omit the phrase ‫“( נאם יהוה‬says the Lord”) in Jer. 31 (OG 38):34. Some other long quotations in the New Testament are Ps. 95 (OG 94):7-11 in Heb. 3:7-11, which stretches over sixty-seven words with five differences: (1) ἐδοκίμασαν/ἐν δοκιμασίᾳ;5 (2) εἴδοσαν/εἶδον; (3) ἐκείνῃ/ταύτῃ; (4) εἶπα/εἶπον; (5) an additional διό. Perhaps the most impressive long quotation is that of Isa. 6:9-10 in Acts 28:26-27, which extends fifty-four words, with just a single omission of the pronoun αὐτός.

Allusions and Verbal Parallels In addition to the 316 quotations listed in The Greek New Testament, there is also a list of more than 1,500 “allusions and verbal parallels” (2014: 864–83). There is much debate as to what exactly constitutes an “allusion,” but here it is combined with the more general category of “verbal parallel.” The distinction is, however, significant. For example, Paul says in Phil. 1:19 that “through your prayers and the help of the Spirit of Jesus Christ this will turn out for my deliverance.” The text is perfectly understandable as it stands, but the last phrase (τοῦτό μοι ἀποβήσεται εἰς σωτηρίαν) has an exact parallel in the OG of Job 13:16. Is this an intended allusion or an unconscious borrowing of biblical language? Richard Hays thinks it is intended, noting that this is Paul’s only use of ἀποβαίνω (used literally of disembarking a boat in Lk. 5:2; Jn 21:9). From this, Hays (1989: 22) draws a wide-ranging conclusion, that “[b]y echoing Job’s words, Paul the prisoner tacitly assumes the role of righteous sufferer, as paradigmatically figured by Job.” Hays (1989: 29–32) offers seven tests for determining whether a particular set of words is in fact an intended allusion: (1) Availability; (2) Volume; (3) Recurrence;

Susan Docherty (2009: 203) notes that our earliest witness to OG Psalms (P.Bod. 24) was not available to Rahlfs but agrees with Hebrews, so that there may only be four differences.

5

246

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

(4) Thematic coherence; (5) Historical plausibility; (6) History of interpretation; and (7) Satisfaction. Others have sought to refine these tests (Litwak 2005; Carey 2009), but the idea is clear—the more tests that are passed, the more likely is the suggested allusion. In the case of Job 13:16, Hays thinks that criterion (1) is met by the fact that Paul quotes Job elsewhere, and that criterion (4) is met by the common theme of suffering exacerbated by so-called friends.6 However, this example does not fare well in the other tests. Volume (criterion 2) is only moderate, as we are dealing with just five words (though ἀποβαίνω is distinctive). Furthermore, there is no recurrence (criterion 3) of the allusion in Paul’s writings and it has not figured in the history of interpretation (criterion 6). Hays acknowledges that Satisfaction (criterion 7) is a rather subjective category, but the conclusion of Hays’s book is that this is just the sort of thing that the apostle Paul would do. More persuasive is the claim that Phil. 2:10-11 (“at the name of Jesus every knee should bend … and every tongue should confess that Jesus Christ is Lord”) is a deliberate allusion to Isa. 45:23 (“By myself I have sworn … ‘To me every knee shall bow, every tongue shall swear.’”). Not only is the conceptual parallel strong—both texts involving knees bending and tongues confessing—but Paul also follows the OG in using ἐξομολογέω (“confess”), whereas the MT has ‫“( ׁשבע‬swear”). The allusion is undoubtedly significant, for words spoken by God are being applied to Jesus, even though the previous verse had God declaring: “For I am God, and there is no other” (Isa. 45:22). N. T. Wright (2013: 684) concludes from this that, “Jesus is to be seen as part of the identity of Israel’s God, and vice versa.” The book of Revelation is well known for having more allusions to the OT than any other NT book, but without ever formally quoting it (Moyise 1995: 15–17). John appears to be familiar with a Hebrew text (especially in Ezekiel), but parallels with the OG can also be found. For example, the statement that “fire came down from heaven and consumed them” (καὶ κατέβη πῦρ ἐκ τοῦ οὐρανοῦ καὶ κατέφαγεν αὐτούς) in Rev. 20:9 agrees with 4 Kgdms (2 Kgs) 1:10, apart from the final pronoun. The song sung in Rev. 15:3-5 is an amalgam of texts, but a number of the phrases are drawn from Ps. 86 (OG 85):8-10. Interestingly, the opening of Rev. 4:1, ἃ δεῖ γενέσθαι μετὰ ταῦτα (“what must take place after this”) agrees exactly with Theodotion’s version of Dan. 2:45, against the OG rendering, τὰ ἐσόμενα ἐπ᾽ ἐσχάτων τῶν ἡμερῶν (“what will be at the end of the days”; NETS). Now it is possible that this example is John’s own translation of the Aramaic (‫מה‬ ‫)די להוא אחרי דנה‬, but most commentators recognize it as first-century evidence of such a Greek text form. Another example is Rev. 20:11, καὶ τόπος οὐχ εὑρέθη αὐτοῖς (“and no

Phil. 1:16–17: “These proclaim Christ out of love, knowing that I have been put here for the defense of the gospel; the others proclaim Christ out of selfish ambition, not sincerely but intending to increase my suffering in my imprisonment.”

6

THE SEPTUAGINT IN THE NEW TESTAMENT 247

place was found for them”), which agrees exactly with Theodotion Dan. 2:35, but the OG reads, ὥστε μηδὲν καταλειφθῆναι ἐξ αὐτῶν (“so that nothing remained of them”). This might suggest that the process that led to Theodotion supplanting the OG in the manuscript tradition was well advanced by the first century, though Gregory Beale (1999: 881) also notes that Rev. 17:14, κύριος κυρίων ἐστὶν καὶ βασιλεὺς βασιλέων (“he is Lord of lords and King of kings”) draws on OG Dan. 4:37, θεὸς τῶν θεῶν καὶ κύριος τῶν κυρίων καὶ βασιλεὺς τῶν βασιλέων (“God of gods and Lord of lords and King of kings”), which is missing from Theodotion and MT.

Other Modes of Influence Vis-à-vis Stereotyped Equivalents Karen Jobes and Moisés Silva (2015: 220–3) offer examples of particular translation choices in the OG that influenced the NT authors. This source of influence is often about frequency of usage rather than mistranslation. For example, although καρδία (“heart”) can be used for the seat of the mental faculties in ordinary Greek, it is quite rare. However, it is very frequent in the OG as a translation of ‫“( לב‬heart”), and this sense is carried over into the NT use of καρδία. Similarly, the use of ἄγγελος in the NT for a supernatural being (“angel”) is partly due to its use in the OG to render ‫מלאך‬ (“messenger”). This might have been significant when the NT authors read OG Deut. 33:2, which suggests that “angels” were present at the giving of the law (Gal. 3:19; Acts 7:53; Heb. 2:2), a detail absent from the MT. A third example is the use of διαθήκη (“will, testimony”) to render ‫“( ברית‬covenant”), a dual meaning that Paul consciously exploits in Gal. 3:15-16: “I give an example from daily life: once a person’s will (διαθήκη) has been ratified, no one adds to it or annuls it. Now the promises were made to Abraham and to his offspring.” The combined evidence of quotations, allusions and stereotyped equivalents has convinced most scholars that the influence of the OG is everywhere in the NT. Even in examples where the OG is a rather “wooden” rendering of the Hebrew and hence the quotation could theoretically come from either source, the evidence above makes it likely that it comes from the Greek. Timothy Law expresses it in the title of his book, When God Spoke Greek. The Septuagint and the Making of the Christian Bible (2013). However, it has taken some time before this conclusion has resulted in significant study of the OG. Even in 1996, Mogens Müller entitled his monograph The First Bible of the Church: A Plea for the Septuagint.

The New Testament and Textual Criticism of the Septuagint Given the paucity of OG manuscripts that can reliably be dated before the first century CE, the NT is potentially an important collection of witnesses to the OG text. However, before this evidence can be used, several questions need to be answered. For example, exegesis in the NT has been variously characterized as “eschatological,” “messianic,” “christological,” or even “ecclesiological” (Hays 1989: 154–92). Did this “core” belief influence the NT authors to change the quoted text, so as to make the parallels more exact? Second, given that a number of important manuscripts contained a text of both the OG and the NT (e.g., Vaticanus, Sinaiticus, Alexandrinus), were the scribes who transmitted these texts tempted to conform one to the other, whether by modifying the OG or modifying the NT quotation?

248

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Examples of Deliberate Changes to the Quoted Text in the New Testament Perhaps the most striking example is when Matthew inserts the word οὐδαμῶς (“by no means”) into his quotation of Mic. 5:2, effectively reversing the meaning: Bethlehem is “by no means least among the rulers of Judah” (Mt. 2:6), whereas Micah was asserting the opposite. Mk 1:3 quotes Isa. 40:3 according to the OG, allowing him to locate the voice of John the Baptist in the wilderness but not necessarily the preparatory action. In order to facilitate the application to Jesus, Mark uses a pronoun (“make his paths straight”) rather than the OG’s τοῦ θεοῦ ἡμῶν (“of our God”). In John’s version of the last supper, Jesus quotes Ps. 41:9 in the form, “The one who ate my bread has lifted his heel against me” (Jn 13:18). The OG (40:10) has the word πτερνισμός (“trickery”) instead of πτέρνα (“heel”) and Maarten Menken (1996: 133–4) argues that John modified the text by drawing on 2 Sam. 18:28 to avoid any suggestion that Jesus was betrayed by Judas’s trickery. At the climax of Paul’s argument in Rom. 11:26, he quotes Isa. 59:20 verbatim except for one word: the redeemer will now come “from (ἐκ) Zion” not “for (ἕνεκεν) Zion” (MT has ‫“ לציון‬to Zion”). As Ross Wagner (2002: 284) notes, the change reflects a fundamental shift in perspective, namely, “the Lord’s coming in person from a restored Zion to bring deliverance to his people who are scattered among the nations.” At the end of 1 Corinthians 15, Paul wants a proof-text that God gives “victory (νῖκος) through our Lord Jesus Christ” and finds it in Hos. 13:14, which he quotes in the form: “Where, O death, is your victory (νῖκος)? Where, O death, is your sting?” (1 Cor. 15:55). However, the key word νῖκος is absent from the OG, which has δίκη (“penalty”; MT ‫דבר‬ “pestilence”). It is to be noted that Jewish exegesis at the time did not necessarily regard such changes as underhand or deceptive. Whereas today we would prefer a clear demarcation between quoted text and subsequent interpretation, Jewish exegesis was happy to combine these into a single modified quotation (Brooke 1985). Matthew could assume that his hearers/ readers would know that Mic. 5:2 spoke about the insignificance of Bethlehem and would therefore attend to his claim that this was no longer the case. Nevertheless, it is clear that caution must be exercised before Mt. 2:6 or 1 Cor. 15:55 can be offered as witnesses to the OG of Mic. 5:2 and Hos. 13:14, respectively.

Cross-contamination in the Transmission of the Old Greek and the New Testament Rahlfs’s edition of the OG was heavily dependent on the three great codices, Vaticanus (B), Sinaiticus (S), and Alexandrinus (A), which also contained a text of the NT. This led to the suspicion that agreements between the OG and the NT quotation, especially if the witnesses were divided, were caused by cross-contamination. In other words, the scribe assumed that the original texts must have been in agreement and that any differences needed to be corrected. Thus when a NT text is cited in the apparatus of Rahlfs, generally it is not there to offer first-century support for a particular reading but rather to explain the origin of a variant reading in OG. For example, against the ὠτία (“ears”) of Ps. 40:6 (OG 39:7), in the apparatus Rahlfs prints: “σωμα BSA: cf. Hebr. 105.” The implication is that, despite the strong attestation of σῶμα in Vaticanus, Sinaiticus, and Alexandrinus, it has not been chosen for the text because it evidently entered the manuscript tradition through Heb. 10:5. The Göttingen edition (also edited by Rahlfs) offers more information, noting that ὠτία is supported by the later recensions and versions (αˈ σˈ θˈ εˈ), as well as

THE SEPTUAGINT IN THE NEW TESTAMENT 249

the Psalterium Gallicanum and manuscript G of the Vetus Latina (OL). But despite the fact that all known Greek manuscripts read σῶμα, the agreement with Heb. 10:5 proved decisive for rejecting it. How much cross-contamination between OG and NT manuscripts was there? Although it is clear that the NT authors did sometimes alter the source text in order to make their point more persuasive, there is considerable debate as to whether this should be regarded as characteristic of their approach. Alan Cadwallader (1992: 286) found 142 examples of contamination among the manuscripts of Hebrews (forty-six within the quotations) that “reflect the effort to sanitize Hbs of its offences” and that “no manuscript escaped its influence.” However, the higher figures were generally found among later manuscripts (e.g., K, L, and Ψ each score above 20 percent), but Sinaiticus is praised for its “remarkable purity” (7.6 percent) and Vaticanus and Alexandrinus for their “overall reliability” (11 percent). Similarly, of the sixty-seven verses of Isaiah quoted in the NT, eighteen of them exhibit variant readings in agreement with the NT.7 However, there are no cases where this has affected Vaticanus, only one case where it has affected Sinaiticus (omission of αὐτῶν in Isa. 6:9 in agreement with Mt. 13:14), and three cases where it has affected Alexandrinus (Isa. 9:2: καθήμενος; cf. Mt. 4:16; Isa. 40:13: συμβιβάσει; cf. 1 Cor. 2:16; Isa. 59:8: ἔγνωσαν; cf. Rom. 3:17). Correspondingly, there are about a dozen changes in the NT manuscripts (e.g., Codex Bezae reads καλεσεις in Mt. 1:23 in agreement with OG Isa. 7:14 and τοῦ θεοῦ ἡμῶν in Mk 1:3 in agreement with OG Isa. 40:3). But only three have affected the first hand of Sinaiticus and only two the first hand of Vaticanus. Indeed, in both examples of accommodation in Vaticanus, NA28/UBS5 prints the words in parenthesis, indicating considerable doubt concerning the original reading.8 From a research project at Wuppertal, Johannes de Vries and Martin Karrer (2013: 14) conclude: “Rahlfs[’s] methodological preference of selecting LXX readings based on their difference from New Testament parallels is now obsolete. This overestimation of the influence of the New Testament quotations over the transmission of the LXX means that every one of his reconstructions of the quoted texts must be rechecked.” Some of this rechecking has already been done by Joseph Ziegler for the Göttingen editions. For example, in Isa. 10:22, Rahlfs followed Sinaiticus in printing τὸ κατάλειμμα αὐτῶν σωθήσεται (“the remnant will be saved”), noting in the apparatus that the omission of αὐτῶν in Aquila was due to cross-contamination from the citation in Rom. 9:27. Ziegler, however, does not find this persuasive and prints the OG text without the αὐτῶν, for which the NT is instead construed as an early witness to it. In support of this perspective, one could also point out that there must have been a great temptation to change κατάλειμμα to Paul’s ὑπόλειμμα but all Greek manuscripts resisted it. Similarly, for the last clause of Isa. 40:4, Rahlfs prints εἰς πεδία (“for plains”), rejecting the reading of αˈScC (εἰς ὁδοὺς λείας; “for a smooth path”), because it agrees with Luke 3:5. Ziegler is again unconvinced and prints εἰς ὁδοὺς λείας, noting that the single word πεδία (“plain”) is closer in meaning to the MT (‫ )בקעה‬and thus more likely to be a revision

E.g., Isa. 7:14: καλέσουσι is read by MSS 26-106 90 130; cf. Mt. 1:23; Isa. 29:14: ἀθετήσω is read by MSS 564 301; cf. 1 Cor. 1:19; Isa. 53:4: ἀσθενείας is read by MSS 22c -93; cf. Mt. 8:17. 8 Rom. 10:15 (omission of τά; cf. Isa. 52:7) and Acts 8:33 (inclusion of αὐτοῦ; cf. Isa. 53:8). 7

250

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

toward a Hebrew exemplar. One could add that λεῖος is a hapax in the NT and so Luke would be unlikely to change a perfectly acceptable πεδία to ὁδοὺς λείας. However, it should be noted that despite praising the “overall reliability” of the three great uncials (i.e., BSA) in respect to conformity to the OG, Cadwallader did find that P46 (c. 200 CE) scored high (19.2 percent), whereas manuscript 33 (c. 800 CE) scored low (10.6 percent). Consequently, it is always possible (even if less likely) that the original reading is preserved in a later manuscript.9 Likewise, it is not always true that Rahlfs chose readings that disagree with the NT text. Thus in Isa. 42:4, he follows the majority of Greek manuscripts in declaring that the nations will hope in the name (ὀνόματι) of God, which agrees with Mt. 12:21. But in his edition Ziegler (1983: 277) prints νόμῳ (“law”) in agreement with the MT (‫ )ולתורתו‬and the later Greek versions (“το εβρ΄ και οι λ΄ π΄”). It is not clear why he does so. Maarten Menken (2004: 80) thinks the OG translator may have been influenced by Isa. 26:8 (ἠλπίσαμεν ἐπὶ τῷ ὀνόματί σου) and rejects the idea that Matthew made the change, which then influenced the whole manuscript tradition.

The New Testament and the Other Greek Versions/Recensions As we have seen, the NT authors sometimes made changes to the received text in order to make a particular theological or rhetorical point (Mt. 2:6; Rom. 11:26). However, sometimes the best explanation appears to be that they were drawing on a different version or recension of the text. Thus, the OG of Isa. 25:8 has managed to reverse the meaning of the Hebrew, so that instead of death being swallowed up forever (‫)בלע המות לנצח‬, we have κατέπιεν ὁ θάνατος ἰσχύσας (“death, having prevailed, swallowed them up”; NETS). It is not surprising that the later versions wished to correct this, with θˊ reading κατεπόθη ὁ θάνατος εἰς νῖκος (“death was swallowed up in victory”). But Paul already knows this reading, which he quotes exactly in 1 Cor. 15:54. This then explains the presence of νῖκος in his quotation of Hos. 13:14 in the following verse (1 Cor. 15:55). It is not an arbitrary change but part of his fusion of the two texts in the combined quotation. Another example is found in Rom. 10:15, where Paul quotes Isa. 52:7 in the form, “How beautiful are the feet of those who bring good news.” This form is closer to the MT than the OG but, more significantly, the adjective “beautiful” (ὡραῖος) is found in a group of later manuscripts that Ziegler calls Lucianic (22c 62 90-130-311 93 456), whereas the OG reads ὡς ὥρα (“like a season”). As Ross Wagner (2002: 172) notes, this Lucianic reading is unlikely to have originated with Paul, as it does not follow any of his other changes, including his unique plural εὐαγγελιζομένων (“heralds”). Thus the NT suggests that the process of revising the OG was already well under way by the first century and even later recensions, such as that by Lucian, appear to have made use of earlier materials.

ONGOING RESEARCH QUESTIONS New Testament Interpretation or Old Greek Interpretation? In cases where the NT differs from the OG, there is considerable debate as to whether the NT authors modified the text to suit their purposes (as with the insertion of οὐδαμῶς in Mt. 2:6) or followed a revised OG text that is now lost to us (Wilk 2006: 253–72). The 9 According to Cadwallader (1992: 287), the most reliable manuscript for Hebrews is P13 (third/fourth century), which scored only 6 percent but the worst offender was the fifth-century codex D at 31.9 percent.

THE SEPTUAGINT IN THE NEW TESTAMENT 251

difficulty of adjudicating between these options was compounded by the discovery of the Dead Sea Scrolls, not so much by the OG texts that were found there but by the variety of Hebrew texts, which could potentially offer a variety of Vorlagen for Greek texts. Thus, there continues to be debate as to whether the author of Hebrews was attracted to Ps. 40:6 (OG 39:7) because the text with σῶμα could be used as a witness to the incarnation (Karrer 2010: 126–46; Steyn 2011: 283–97) or is a rather obvious theological change made by the author of Hebrews (Jobes and Silva 2015: 216–18). One scholar who has argued firmly in favor of revised texts is Maarten Menken. In Matthew’s Bible: The Old Testament Text of the Evangelist (2004), he argues that in the majority of cases, the differences between Matthew and the OG are not explicitly Christian but could equally have arisen prior to Matthew writing. For example, in the first explicit quotation (Mt. 1:23 quoting Isa. 7:14), Matthew follows the OG’s rendering of ‫“( עלמה‬young woman”) with παρθένος (“virgin”) but instead of the second person singular καλέσεις (“you will call”), he has the third person plural καλέσουσιν (“they will call”). A number of scholars believe the explanation for this lies in the fact that the angel tells Joseph that Mary will call the child “Jesus” (1:21) but the proof-text from Isa. 7:14 says that the child will be called “Emmanuel.” The dissonance is solved by changing to a third person plural so that “they will call” refers to a different subject, namely, the later Christian community, who will come to believe that “God was in Christ” (2 Cor. 5:19) and hence Emmanuel (Gundry 1994: 25). Menken (2004: 126–31) disputes this argument. He first notes that the impulse to change the second person singular (‫ )וקראת‬into a third person singular (‫ )וקרה‬is already present in 1QIsa.a, perhaps to dissociate the naming of the child from the evil Ahaz (2 Kgs 16; 2 Chron. 28). The third person singular is probably intended in an impersonal sense (“one will call”), much as Matthew’s third person plural καλέσουσιν (“they will call”). Second, Menken argues that there is no evidence that Matthew wished to relegate the name/title “Emmanuel” (or “God is with us”) to the later Christian community, while preserving “Jesus” for his earthly ministry. Indeed, most scholars recognize that Mt. 28:20 (“I am with you [μεθ’ ὑμῶν] always, to the end of the age”) forms an inclusio with Mt. 1:23. Jesus’s abiding presence is also affirmed in Mt. 18:20, where he makes the promise: “where two or three are gathered in my name, I am there among them,” though the actual phrase μεθ’ ἡμῶν is not used. Menken concludes that Matthew wished to endorse both names and saw no conflict between them. Thus he thinks it is more likely that Matthew is following a revised OG manuscript that read καλέσουσιν rather than deliberately altering one that read καλέσεις.

Collections of Texts If we were to posit a revised OG text for every difference in NT quotations, we would end up with an absurdly large number of revisions. Martin Albl (1999) has therefore revived Rendel Harris’s testimony hypothesis (1920) to account for the form of many of the quotations in the NT. Harris argued that the early Christians most likely produced small collections of texts (“testimonia”) that could be used in debate with outsiders. Such collections would have focused around a theme, such as Messianic proofs for addressing non-believing Jews. The theory fell into neglect for lack of specific examples (Harris relied heavily on Justin Martyr), but the discovery of such “testimonia” or “florilegia” at Qumran has given it fresh impetus. Albl offers three reasons for why this is often a better explanation than a revised OG text (1999: 160):

252

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

(1) It is unlikely that Paul interrupted his dictation to unroll a large scroll every time he wanted a quotation; (2) The textual variety found in Paul’s quotations is explained by suggesting that he copied extracts from scrolls during his travels; and (3) It explains why some texts are used in a sense that appears foreign to their original contexts. The classic example of the testimony hypothesis is the combination of Isa. 8:14 and 28:16 in Rom. 9:33 and 1 Pet. 2:6. As well as combining these two texts, for which the OG addition of καὶ ἐὰν ἐπ᾽ αὐτῷ πεποιθώς (“and if you trust in him”) at the beginning of Isa. 8:14 may have been the catalyst, 1 Peter and Romans both use σκάνδαλον instead of πτῶμα in Isa. 8:14 and τίθημι instead of ἐμβάλλω in Isa. 28:16. Since scholars think it very unlikely that Paul made use of 1 Peter or vice versa, it seems probable that both are drawing on a collection of “stone” quotations, especially as 1 Peter also cites Ps. 118:22 (“The stone that the builders rejected”) in the very next verse. A further example is the quotation of Deut. 32:35a (“Vengeance is mine, and recompense”) in Rom. 12:19 and Heb. 10:30. The OG differs from the MT by beginning with ἐν ἡμέρᾳ ἐκδικήσεως (“ in the day of vengeance”) and using a verb ἀνταποδώσω (“I will repay”) instead of a noun (“and recompense”). Paul and Hebrews both have ἐμοὶ ἐκδίκησις ἐγὼ ἀνταποδώσω (“vengeance is mine, I will repay”), agreeing with the MT for the first two words (‫ )לי נקם‬and the OG for the verb, resulting in a text quite close to the Targums. This mixed text is not easily explained as part of a revision of OG Deuteronomy toward a Hebrew text; it looks more like an ad hoc summary that both Paul and the author of Hebrews have made use of.10 A weakness of the testimony hypothesis is that some textual variations only seem to carry significance within their specific NT contexts. For example, in Paul’s discussion of speaking in tongues, he cites Isa. 28:11-12 in 1 Cor. 14:21, a promise that God will speak to “this people” with ἑτερογλώσσοις (“other tongues”) and χείλεσιν ἑτέρων (“other lips”). The OG has φαυλισμὸν χειλέων (“contempt of lips”) and γλώσσης ἑτέρας (“other tongues”), so that either Paul or his source removed the idea of contempt and conformed the “lips” clause to the “tongues” clause. However, it is difficult to see why an earlier excerpter would produce a text about “other lips and tongues” outside of this context, but easy to see why Paul might have done. For the testimony hypothesis to be convincing, the modified text needs to be useful to Christians in general, not just a single author speaking to a particular context (see Law 2013: 91–4 for other examples).

Composite Citations Sometimes what looks like an arbitrary change to a text is actually a conflation of one text with another, also known as a composite citation. For example, the quotation of Hab. 2:3-4 in Heb. 10:37-38 begins with the words ἔτι γὰρ μικρὸν ὅσον ὅσον (“for yet a little while”), which is secure in the manuscript tradition.11 Most commentators think

On the ancient view that Paul wrote Hebrews, no such explanation would be required, but few scholars hold to such a view today. 11 Cadwallader (1992: 271) notes that D reads μικρὸν ὅθεν ἐρχόμενος, though this is not mentioned in the apparatus of NA28. 10

THE SEPTUAGINT IN THE NEW TESTAMENT 253

this quotation is simultaneously drawing on Isa. 26:20, where Israel is told to hide “for a little while” (μικρὸν ὅσον ὅσον) until the wrath is past. Whether the first hearers would have recognized this or simply assumed that it was part of the “righteous living by faith” quotation is a matter of debate. The phenomenon of composite citations has been studied in two volumes edited by Sean Adams and Seth Ehorn. The first volume looked at a variety of Jewish, Greek, and Roman authors (2016), while the second focused on the NT (Adams and Ehorn 2018). In the latter, Catrin Williams has written the chapter on John’s Gospel and finds eight examples of this phenomenon (6:31; 7:38; 12:13; 12:15; 12:40; 13:18; 19:36; 19:37). Thus in Jn 19:36, a citation is offered for why the soldiers did not break Jesus’s legs as he hung on the cross: ὀστοῦν οὐ συντριβήσεται αὐτοῦ (“None of his bones shall be broken”). Considering the prominence of Passover motifs in John’s Gospel, Williams suggests the source is either Exod. 12:10, 46 (καὶ ὀστοῦν οὐ συντρίψετε ἀπ᾽ αὐτοῦ) or Num. 9:12 (καὶ ὀστοῦν οὐ συντρίψουσιν ἀπ᾽ αὐτοῦ). If this is the case, then it raises the question of the origin of the passive form συντριβήσεται. It could be John’s modification of the text but Williams suggests that it has been taken from Ps. 34:20 (OG 33:21). The context is God’s care of the righteous and ends with the promise, “He keeps all their bones; not one of them will be broken” (φυλάσσει πάντα τὰ ὀστᾶ αὐτῶν ἓν ἐξ αὐτῶν οὐ συντριβήσεται). If this is correct, John has not arbitrarily changed the wording of his Pentateuchal text, but has conflated it with another text that was significant to him, though how many of his readers would have perceived this is open to discussion.

Availability of Old Greek Manuscripts Faced with the difficulty that some of the quotations in the NT seem to be taken out of context, scholars have suggested a variety of sophisticated theories to explain what the NT authors were trying to do. For example, consider the use of Isa. 52:5 in Rom. 2:24. The immediate context of Isa. 52:5 is that God’s name is being despised because of the piteous state of Israel. However, this piteous state will soon be reversed, as 52:7 makes clear (“How beautiful upon the mountains are the feet of the messenger who announces peace, who brings good news”). In Rom. 2:24, Paul appears to ignore this positive outcome and uses the quotation as proof of Israel’s guilt (“The name of God is blasphemed among the Gentiles because of you”). N. T. Wright (2013: 811–15) claims that Paul has not taken Isa. 52:5 out of context but is interpreting it in the light of Ezek. 36:20-23 (which does accuse Israel). Wright thinks that Paul makes this clear when he goes on to speak about “the spirit, the law in the heart, the fresh keeping of Torah’s requirements” (2013: 814) in Rom. 2:25-29, which would immediately be recognized as a reference to Ezek. 36:26-27, the same context as Ezek. 36:20–23. If this explanation is correct, it would seem that Paul can assume that the house churches in Rome have ready access to OG manuscripts and are fairly well acquainted with them. Is this claim credible? Christopher Stanley (2004) thinks not, arguing that not only were manuscripts expensive to produce but levels of literacy were such that only the educated elite would have been able to read them (also Law 2013: 90). According to 1  Cor. 1:26, not many of the Christians came from such an elite social class and so either Paul completely misjudged the abilities of his readers (not unknown among teachers) or modern scholars have misunderstood the rhetoric of Paul’s letters. Stanley opts for the latter, suggesting that instead we should focus on what Paul makes overt in his quotations, rather than overly subtle nuances derived from supposedly parallel texts. For Stanley,

254

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

scholars like Hays, Wagner, and Wright deduce more from what Paul does not quote than from what he does actually quote. It is clear that the OG had an enormous impact on the language, theology, and rhetoric of the NT writings. In some cases, the NT author’s argument depends on an idiosyncratic rendering of the Hebrew (e.g., Amos 9:11-12 in Acts 15:16-18) but even when the OG is similar to the MT, it is often the Greek text that is the vehicle for communicating the point (e.g., the use of διαθήκη [“will, covenant”] in Gal. 3:15). The NT is also a valuable witness to the text of the OG in the first century, though its evidence has to be used with care. It is always possible that later scribes conformed the NT quotation to the OG text, whether by changing the quotation or by changing the OG. The NT also offers early evidence for the existence of various revisions of the OG and even the later recensions (such as that of Lucian) seem to have drawn on earlier materials. There is debate as to whether the NT authors were aware of this pluriformity or simply used the texts to which they had access (Norten 2011). It is not an exaggeration to say that without the OG, there would be no NT.

RECOMMENDED RESOURCES 1. Moyise and Menken (2004). The first of five volumes that deal with the use of specific books (Genesis, Deuteronomy, Psalms, Isaiah) or sections of Scripture (Minor Prophets) in the NT. 2. Hatina (2006). The first of a series of volumes that cover the use of Scripture in individual Gospels. 3. Menken (2004). A collection of studies arguing that Matthew’s quotations derive from Mark’s Gospel and a revised text of the OG. 4. Watts (1997). Argues from the initial composite quotation (Mk 1:2-3) that themes from Deutero-Isaiah and Malachi are key to interpreting Mark’s Gospel. 5. Litwak (2005). Argues for the influence of a broad range of Scriptures rather than focusing on one or two key books. 6. Hays (1989). The book that began the serious study of allusions and echoes as well as quotations in the letters of Paul. 7. Docherty (2009). Focuses on the textual background of the quotations and the Jewish exegetical procedures of the author. 8. Beale (1999). A collection of studies arguing that John was a careful exegete of Scripture, as well as a seer of visions.

CHAPTER 17

The Septuagint in Patristic Sources EDMON L. GALLAGHER

INTRODUCTION Jewish believers in Jesus had been using Greek translations of Scripture already for a century before any of them acknowledged the fact in writing. The first Christian writer to mention the Septuagint was Justin Martyr, who not only transmitted a version of the translation legend in his First Apology (ch. 31) but also—in his Dialogue with Trypho (esp. chs. 70–73)—promoted the authenticity of the original translation against competing translations (discussed below).1 Not long after Justin, Irenaeus of Lyon (Haer. 3.21.13) more explicitly attacked recent Jewish translations as inferior to the Septuagint, a position Irenaeus supported through several arguments, including his own retelling of the translation legend. These earliest Christian comments on the Septuagint attribute to the Seventy translators not merely the Greek Pentateuch, as in our preserved Jewish sources, but the entire Greek Old Testament, a transformation of the work of the Seventy that becomes crucially important for patristic reflection on Scripture. When the Fathers encountered differences between their manuscripts of the Septuagint and other Greek translations, they explained these differences in various ways, but they rarely questioned the authority of the Seventy translators, whom they thought God had inspired to make the Hebrew Scriptures available to the world. Jerome’s promotion of the alternative view that the Seventy were mere translators, not prophets, and that their translation sometimes contained mistakes, caused controversy throughout the West, particularly with his erstwhile friend Rufinus of Aquileia and his correspondent Augustine of Hippo. Eventually Jerome’s own translation, later christened the Vulgate, replaced the Septuagint as the Old Testament in the Latin-speaking West. But Jerome’s work had no similar effect in the East, and the Greek Orthodox Church continues to use primarily the Septuagint as its Old Testament (Pentiuc 2014: 62–100).2

For discussion of the major revisions and recensions of the Septuagint, see Mäkipelto in the present volume. On the legendary account of the origin of the Septuagint, see the chapter by Wright. 2 For a discussion of the Septuagint in the Orthodox tradition, see Seleznev in the present volume. 1

256

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

THE RECEPTION OF THE SEPTUAGINT IN PATRISTIC LITERATURE The Septuagint Translation Legend The earliest Christian account of the production of the Septuagint assumes that the translation encompasses the prophetic books of the Old Testament (Justin, 1 Apol. 31). (The ensuing argument shows that Justin regards the entire Old Testament as prophetic.) Similarly, it is in a discussion of the Immanuel prophecy of Isa. 7:14 that Irenaeus brings up the authority of the Septuagint. Only Jerome, two centuries later, will point out that the Jewish sources such as the Letter of Aristeas and the account in Josephus limit the work of the Seventy to the Mosaic Law. Jerome’s comments in this regard have little effect on subsequent discussion. Major features of the translation legend derive from Jewish sources (Gallagher 2012: 147–52). The Letter of Aristeas narrated the full story in which seventy-two Jewish sages travel to Alexandria from Jerusalem, work together on the island of Pharos to produce an accurate translation, completing their work in seventy-two days, “as if according to some design” (οἱονεὶ κατὰ πρόθεσιν τινα; §307). The same features appear in Josephus’s epitome of the “Book of Aristeas,” as he calls it (Ant. 12.11-118), and Aristobulus in the second century BCE shares the basic contours of the account (frg. 3 in Holladay 1995: 156). We encounter new elements in Philo, most influentially the assertion that God inspired the translators to produce a perfect translation. Philo recognizes, as he says, that translation is an inexact science because no two languages correspond perfectly, but in the case of the Septuagint God overcame this difficulty by endowing the translators with his own Spirit, thus enabling them to find Greek words that exactly matched the Hebrew words both in sense and etymology (Vit. Mos. 2.37-40; Kamesar 2009: 65–72). Philo does not so much argue for this position as assert it. He imagines that, should someone learn both languages, Greek and Hebrew (or, to use Philo’s term, “Chaldean”; see Gallagher 2012: 123–4), he would regard the original text and the translation as “sisters, or rather … one and the same both in sense and in words” (Vit. Mos. 2.40). While Aristeas hinted that divine Providence aided the translation, Philo declares that the Greek Pentateuch is just as inspired as the Hebrew Torah and that the two say precisely the same thing, because God willed it so. The Jewish versions of the story include several elements that would become influential in Christian tradition. The translation was accomplished during the early Ptolemaic era. Seventy-two Jewish sages worked on the translation. The translation corresponded well—or perfectly, as Philo would say—with the Hebrew text. Aristeas hints at divine participation, Philo insists upon it. What Philo does not fully explain is whether he imagines that the translators cooperated in their work or labored separately. Aristeas says expressly that the translators compared their results before submitting a final draft (§302), but Philo attributes to God’s invisible prompting the fact that all the translators used the same words (Vit. Mos. 2.37). He seems to imagine them working independently and miraculously arriving at the same result (pace Wasserstein and Wasserstein 2006: 45), in a way similar to a version of the story found in the Talmud (Meg. 9a–b). At any rate, an account along these lines will become popular among Christians. But, as mentioned earlier, Christians largely ignore or reinterpret the Jewish statements that the Seventy translators worked on the Jewish law, apparently understanding the term “law” as encompassing not just the Pentateuch but the entire Old Testament (in line with some New Testament passages; cf. Jn 10:34; 1 Cor. 14:21; etc.).

THE SEPTUAGINT IN PATRISTIC SOURCES 257

Most patristic accounts of the translation echo Philo’s belief in the divine inspiration of the Septuagint, and they regard the Septuagint as a supremely accurate rendering of the Hebrew Bible, though they usually refrain from following Philo completely in his assertion that the Hebrew and Greek represent the exact same text; Christians recognize ambiguities and difficulties in the Septuagint that, they surmise, did not exist in the Hebrew text (Pentiuc 2014: 97–9; and see below). Justin actually does not mention the inspiration of the translation, but his entire discussion, both in the First Apology (ch. 31) and in the Dialogue with Trypho (ch. 71), presupposes that the translation of the Seventy accurately represented the original Hebrew prophecies. In Justin’s mind, the Jewish leaders are at fault for refusing to acknowledge the accuracy of the ancient translation and proposing a revised translation/interpretation, particularly with regard to Isa. 7:14, a verse that should read “virgin” (παρθένος) rather than “young woman” (νεᾶνις). It was essential to Justin’s argument that the ancient Hebrew prophecies agreed completely with his Greek Bible in their predictions of the Christ. Irenaeus argues extensively for the inspiration of the Septuagint. Whether or not Philo imagined that the translators worked independently, Irenaeus certainly does, asserting that King Ptolemy desired to test the translators by separating them and demanding each to produce a translation. Comparison of the seventy-two translations showed that they all contained the same words, proving not only that the translators produced an accurate translation but much more that “the Scriptures had been translated by the inspiration of God” (κατ᾽ ἐπίπνοιαν τοῦ θεοῦ εἰσιν ἑρμηνευμέναι αἱ γραφαί; Haer. 3.21.2; apud Eusebius, Hist. eccl. 5.8.14). This story of separated translators producing identical results was repeated many times—with some variations—by patristic authors (see Wasserstein and Wasserstein 2006: 95–131), such as Clement of Alexandria (Strom. 1.22), Ps.-Justin (Cohort. Graec. 13), Epiphanius (De Mens. et Pond. 17), John Chrysostom (Hom. Matt. 5.2), and Theodoret of Cyrus (Comm. Isa. 7:14). It was Jerome who pointed out that the earliest Jewish sources, Aristeas and Josephus, in fact say nothing about the separation of the translators (Praef. in Pent.). Jerome’s caution seems to have made little impact on his contemporaries. Even Augustine, who early seemed not completely convinced of the miracle story (Doctr. chr. 2.15.22), later affirmed it more confidently (Civ. 18.43).

The Apostles and the Septuagint The textual form of Old Testament quotations in the New Testament is a complex topic defying simple solutions.3 Nevertheless, patristic authors often considered the matter rather straightforward: the apostles quoted the Septuagint (Kamesar 1993: 29–34). This interpretation so dominated patristic interpretation that hardly anyone felt the need to justify it through detailed comparison of texts. At most, an author might compare Mt. 1:23 with Isa. 7:14 in its available versions (Justin, Dial. 67, 71; Irenaeus, Haer. 3.21.1). Jerome, again, sought to complicate the picture by pointing to several New Testament quotations that did not match the Septuagint. He claimed to have found the Vorlage for these quotations in the Hebrew text (Kato 2013). Not everyone found Jerome’s argument convincing (e.g., Rufinus, Apol. Hier. 2.37-38; Gallagher 2012: 203–5), but Augustine, for one, did come to acknowledge that the apostles quoted both the Septuagint and the Hebrew text (Civ. 18.44; Gallagher 2016). On this topic see the chapter by Moyise in the present volume.

3

258

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Exegesis The Septuagint (as traditionally understood) served as the textual basis for patristic interpretation, usually without recourse to any other textual witness. Early Christian exegetes did not, as a rule, bother with Hebrew at all, even to the point of treating the Greek text as the original text. To cite an oft-discussed example, some interpreters saw in the 318 men taken by Abram in search of Lot (Gen. 14:14) a reference to the name of Jesus, because in Greek the number eighteen is written with the first two letters of Ἰησοῦς, and the number 300 is written with the Greek letter tau, which resembles a cross (Barn. 9.7-8; Clement of Alexandria, Strom. 6.85). This sort of interpretation could not have occurred to a reader of Hebrew. Whereas patristic interpretation of many biblical passages might have arrived at the same basic results regardless of the textual form used because “the Septuagint and the Hebrew are either the same or similar” (as Jerome asserts near the end of the preface to the first book of his Comm. Isa.; see also Dorival 2016b: 22–3), there are many passages for which a Greek text as opposed to the Hebrew text—or, alternatively, the Septuagint as opposed to other Greek translations (such as “the Three”)—could suggest distinctive avenues for interpretation. For example, Origen begins his Exhortation to Martyrdom (c. 235 CE) with an exposition of the Septuagint text of Isa. 28:9-11, where the Hebrew and Greek texts diverge significantly: the MT has “precept upon precept” (‫)צו לצו‬ where the Septuagint reads, “receive affliction upon affliction” (θλῖψιν ἐπὶ θλῖψιν προσδέχου), a reading that makes this verse relevant to the theme of martyrdom. Readers of the Septuagint learn in Gen. 14:13 that Abraham is an emigrant (περάτης), whereas the Hebrew text calls him a Hebrew (cf. Philo, Migr. 20; further references at Johnson 2006: 115; Harl 2010: 159). The Septuagint of Gen. 5:24 presents a fairly straightforward translation, but the use of μετατίθημι gave rise to the tradition of making Enoch a model of repentance (cf. Sir. 44:16), an idea hardly derivable from the Hebrew. (These examples are taken from Harl 1988: 292–3; see also 304–11.) One can also gain a sense of what early Christians made of the distinctive interpretive possibilities of the Greek and Hebrew by perusing Jerome’s commentaries on the Hebrew prophets, which often present alternative interpretations for the Hebrew text and for the Septuagint. After presenting five pages of excerpts of patristic interpretations based specially on the Septuagint, Swete emphasizes two points (1900: 464–71): first, the peculiar wording of the Septuagint shaped the Christian imagination in liturgy and Bible study; and, second, the Greek version contributed its language to theological formulation in these crucial early centuries. He points out how critical was the wording of Prov. 8:22, κύριος ἔκτισέν με ἀρχὴν ὁδῶν αὐτοῦ εἰς ἔργα αὐτοῦ (“the Lord created me as a beginning of his ways for his works”). Since the speaker of this line, ostensibly divine Wisdom, was universally acknowledged among Christian exegetes to be the pre-incarnate Christ (cf. 1 Cor. 1:24), this verse could be used to demonstrate that the Son is a created being (Simonetti 1994: 127–8). Those inimical to such an idea devised various explanations, the most popular being Athanasius’s proposal that the apparent reference to the creation of Christ pointed to the birth of Jesus from Mary (Or. 2.44-56). Other interpreters compared different textual traditions. Eusebius of Caesarea points out, in part, that Aquila, Symmachus, and Theodotion each used in Prov. 8:22 not κτίζω but κτάομαι, a verb also appearing in Gen. 4:1 in the context of giving birth (Eccl. theol. 3.2).4 According to Eusebius, these Eusebius sometimes viewed the recentiores, particularly Symmachus, as clearer than the Septuagint (Barthélemy 1978: 179–93).

4

THE SEPTUAGINT IN PATRISTIC SOURCES 259

alternative translations clarify what is apparently a special use of κτίζω in Prov. 8:22, which does not mean “create” but—in light of the translations of the Three—something more like “acquire” (DelCogliano 2008). The wording of the Septuagint could be reinterpreted but not abandoned, and the versions of “the Three” could help to explain its more peculiar elements or otherwise be enlisted for Christian exegesis (Kamesar 1993: 35–8; Salvesen 2015). There were also other times that patristic exegetes acknowledged the Hebrew origins of their Scriptures. The etymology of Hebrew proper names was an enduring part of Greco-Latin biblical exegesis (Graves 2014: 65–70), going back to Philo and before (Grabbe 1988). Origen (Harl and de Lange 1983: 447–57) and Jerome (Kamesar 1993: 103–26) used Hebrew etymologies extensively in their exegesis, a practice that continued throughout the Middle Ages owing to the wide influence of Jerome’s work On Hebrew Names (Poleg 2013). As Augustine said, “many Hebrew names that are not interpreted by the authors of these books undoubtedly provide no meager power and assistance in solving the enigmas of the Scriptures, if someone can interpret them” (Doctr. chr. 2.16.23). The Antiochene school of biblical exegesis tended to take a critical view of the Septuagint as a translation (Gallagher 2012: 195–6). Particularly the early Antiochene exegete Eusebius of Emesa, who knew no Hebrew but whose native language was Syriac, did not hesitate to criticize the Septuagint as a poor translation, as he did in his comment on Gen. 1:1, where he asserted that the Seventy should have used the word ἔκτισε rather than ἐποίησεν to render the Hebrew verb for “create” (Haar Romeny 1997: 169–74). While the Septuagint continued to be the “point of departure” for Eusebius (ibid.: 112), “recognition of the priority of the Hebrew text” was “his starting-point” (ibid.: 139).

Textual Theory among Greek Patristic Authors The translation legend and the traditional Christian use of the Septuagint extending back to the apostles undergirded the authority of this translation against potential rival translations. Even in the second century, Christians took notice of later Greek translations of the Jewish Scriptures. Whereas Justin merely referenced the translation νεᾶνις in Isa. 7:14 as promoted by Jewish teachers (Dial. 43; 68), Irenaeus named Theodotion and Aquila as producing translations with this new reading (Haer. 3.21.1). Origen also knew the translation attributed to Symmachus, and his grouping these three translations— Aquila, Symmachus, and Theodotion—together with the Septuagint and the Hebrew text in his Hexapla ensured that these three came to be considered the most important versions after the Septuagint.5 Patristic authors frequently cite “the Three” later Jewish translators, both positively—to clarify the Septuagint or to gain a better understanding of the Hebrew text—and negatively, as examples of faulty translation or even malicious Jewish corruption of the biblical text. Early Christians regarded the differences between the Septuagint and the Three as a problem demanding attention because the Jewish advocates of the later versions promoted these texts as superior, more closely approximating the Hebrew Bible. Christians developed several strategies for countering this proposition. As in many areas, Origen’s work proved to be seminal to the subsequent discussion.

On the Hexapla of Origen, see the chapter by Gentry in the present volume.

5

260

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

The second-century authors provide little indication that they investigated the new versions to any great extent. Justin and Irenaeus—the only pre-Origen Christians whose comments on these textual differences are preserved for us (but cf. Tertullian, Marc. 3.13; Adv. Jud. 9.8)—regard the new translations, or at least the proposed translation of Isa. 7:14 with the wording νεᾶνις in place of παρθένος, as at best inaccurate and at worst the willful perversion of Scripture designed merely to score points against Christians. Neither Justin nor Irenaeus seems to consider seriously the possibility that the wording of the Septuagint might diverge from the Hebrew. Irenaeus knows that the Septuagint must be the correct translation because the Jews themselves translated it (Haer. 3.21.1), the miracle story demonstrates its authority (3.21.2), the translation long preceded the coming of Christ so that there was no possibility of Christian bias, and the apostles used the Septuagint (3.21.3). Justin does not mention the miracle story or the apostolic use of the Septuagint, but he stresses that the Jews themselves produced the Septuagint in ages long past, and the newer translations or interpretations result from anti-Christian prejudice (cf. Dial. 68-73). Justin and Irenaeus insist that the Septuagint alone accurately represents the Hebrew text (Gallagher 2012: 174–8). Origen knew better (Gallagher 2012: 178–89). His enormous research project known as the Hexapla demonstrated numerous differences between the Hebrew text as Origen knew it and the Septuagint as it existed in the early third century. A detailed inspection of the columns of the Hexapla would, of course, reveal many differences of translation such as Justin and Irenaeus observed in Isa. 7:14, though the Hexapla itself could not clarify for the reader without Hebrew which of the Greek translations most accurately reflected the first column. But a mere glance at this parallel Bible sufficed to expose the quantitative differences among the editions (cf. Origen, Ep. Afr. 4-5)—that is, where the Septuagint had more or less material than the Hebrew text—in which cases Aquila, Symmachus, and Theodotion almost always matched the Hebrew text. One who had seen the Hexapla could not claim that the Septuagint and not the Three best conformed to the Hebrew Bible. Origen’s views on this matter are complex and subject to varying interpretations. At times he seems to advocate a Septuagint revised according to the Hebrew text, as when he speaks about “healing” the text of the Septuagint (Comm. Matt. 15.14), or in his adapting the Greek word order toward the Hebrew word order, or correcting the spelling of proper names. These aspects of Origen’s textual theory suggest that he believed that the Septuagint textual tradition had suffered corruption and he could restore it to its original form by conforming it to the Hebrew text, a theory that he worked out in practice with his Hexapla and the Septuagint recension resulting from it. But elsewhere Origen argues stridently against the suggestion by Julius Africanus that questions surrounding quantitative differences between the Septuagint and the Hebrew Bible should be decided in favor of the latter. Though Origen frequently refers to the translators as “the Seventy,” he never recites the miraculous legend of the origins of the Septuagint. Instead, he argues for the authority of the translation from his confidence that divine Providence has ensured that the church has the correct Bible, regardless of these textual issues (Ep. Afr. 8). These are cases, apparently, in which Origen did not think the Septuagint had suffered corruption. Origen offers a detailed explanation for why the Septuagint sometimes contains more material than the Hebrew Bible: the Jewish leaders have excised narratives unflattering to themselves (Ep. Afr. 13-15). His argument implies that the Hebrew Bible itself may have suffered in transmission, in which case its differences from the Septuagint could hardly encourage revision toward the corrupt Bible of the Jews.

THE SEPTUAGINT IN PATRISTIC SOURCES 261

Whereas Justin and Irenaeus had assumed that the Septuagint corresponded more closely with the Hebrew text than did the later Jewish translations, Origen’s work demonstrated that the Septuagint diverged from the Hebrew Bible often. As just discussed, he entertained multiple explanations for this potentially troubling situation. Sometimes he regarded the Septuagint as textually suspect. At other times, he regarded the Hebrew Bible as textually suspect. In both cases, Origen’s underlying assumption was apparently that the original Septuagint constituted a faithful translation of its Vorlage. There were also occasions when Origen claimed that what might at first blush appear to be a difference did not, upon critical examination, turn out to be so. Such is the case for παρθένος and νεᾶνις at Isa. 7:14, the difference that had so troubled earlier Christians. While Justin and Irenaeus had rejected the reading νεᾶνις as patently inaccurate, Origen interprets these two translation options as synonyms in the biblical idiom, both meaning (or, at least, allowing for the meaning) “virgin” (Cels. 1.34-35), an argument that became standard in later patristic writings (Kamesar 1990). But Origen also proposed a further explanation for observed textual differences: he sometimes attributed to the Seventy translators a desire to highlight certain aspects of biblical religion by putting forward a turn of phrase not precisely equivalent to what they found in their Hebrew Vorlage. For example, some prophecies of Christ, which appear with the future tense in the Hebrew text and in the Three, are rendered by the Seventy in the past tense. Origen suggests that the Seventy wanted to represent God as omniscient (Sel. Ps. 2:1, PG 12.1104c), whereas the Three translate “more clearly” (σαφέστερον; Sel. Ps. 42:3, PG 12.1420d). Sometimes the Seventy offer spiritual rather than literalizing translations, such as might be the case at Ps. 3:8 when they (perhaps) substitute the term “without cause” for the more literal “cheek” (Sel. Ps. 3:8, PG 12.1129b–c). Nevertheless, Origen usually considered the Septuagint to be a very faithful translation, which contributed to its high status in his estimation (Kamesar 1993: 15–16). Patristic textual theory after Origen stood on the foundation laid by him (Gallagher 2013b). These writers usually still regarded the Septuagint as an inspired text, provided to the church by God, and often they based this belief on the traditional reasons, namely the miraculous origin story and the apostolic use of the translation (Wasserstein and Wasserstein 2006: 109–24; Pentiuc 2014: 90–3). (We have already noted that not everyone regarded the Septuagint as a perfect translation; see the earlier comments on Eusebius of Emesa.) The work of Origen demonstrated that the Septuagint diverged frequently from the Hebrew Bible, while Aquila, Symmachus, and Theodotion had faithfully represented the Hebrew text. For the most part, patristic authors of the fourth and fifth centuries adopted Origen’s principles for explaining such differences between the authoritative Septuagint and the Hebrew: divergences resulted from corruption in the manuscript tradition of one of the two texts—a supposition in some ways confirmed in the twentieth century through manuscript discoveries in the Judean Desert. Just as Origen rarely suggested that the Seventy translators had strayed from their Vorlage, so also this explanation did not feature prominently in the work of his followers. These writers seem to have assumed that originally the Septuagint and the Hebrew Bible stood united, a unity sundered only by the intervention—willful or not—of copyists. For example, Gregory of Nyssa accuses Jewish scribes of eliminating from the Hebrew Psalms certain messianic superscriptions present in his copy of the Septuagint (Inscr. Psa. 2.8-9). On the other hand, Theodoret of Cyrus recognizes the same textual difficulties in the Psalter superscriptions, but he is willing to concede that the problem resides in the Septuagint manuscript tradition (Comm. Psa. 93 LXX), though the original translators, he believes,

262

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

“slavishly followed the Hebrew text” (τῇ Ἑβραίων γλώττῃ δουλεύσαντες; Comm. Cant. 3:6, PG 81.120a).

The Septuagint in Latin Christianity The origins of the Latin Bible are obscure, but the earliest Latin Old Testament translations were based on the traditional Greek versions and probably originated in second-century North Africa (Bogaert 2013; Houghton 2016: 3–18). The Old Testament seems to have been translated by Christians (not Jews) from the Greek versions of the books (Kraus 2003). The evidence suggests a single translation for each book (Houghton 2016: 12). These Latin translations were also called “Septuagint” (i.e., the translation of the Seventy, in Latin Septuaginta) by Latin writers (Linde 2011: 9–13, 126–7). By the late fourth century, the textual tradition of these Latin translations had suffered corruption to the extent that revision was necessary, and Jerome stepped forward to perform the task. His various efforts to improve the Latin text of the Bible eventually resulted in the Vulgate (though it received this name only much later), a new biblical text that proved controversial at its creation. In a radical move, Jerome ultimately based his translations of the Old Testament on the Hebrew text available in his day, rather than on the Septuagint. Perhaps uniquely among Latin-speaking fourth-century Christians, Jerome learned Hebrew well and put this knowledge to use in his biblical scholarship (Graves 2007; Newman 2009). He advertised himself as a Vir trilinguis (Ruf. 2.22; 3.6), a three-language man (Latin, Greek, Hebrew), garnering him authority as a premier biblical exegete. He promoted knowledge of the Hebrew Bible, the Hebraica veritas, as important and beneficial to Christians, and others seem to have thought similarly as they wrote him questions requesting his insight into biblical passages. (On Jerome’s correspondence, see Cain 2009.) His commission from Pope Damasus in the early 380s to revise the Latin text of the Gospels (mentioned only in Jerome’s Praefatio in Evangelio; Cain 2009: 48–52) initiated his career of revising or translating the text of Scripture, which took up much of his time over the next two decades. Aside from the Gospels, Jerome worked on the biblical text of the Old Testament exclusively. (He did write commentaries on some other New Testament books.) At first, he revised the Latin translations of the Septuagint, basing his work on Origen’s hexaplaric Greek recension complete with asterisks and obeli, a project that elicited the commendation of Augustine (Epist. 28.2). However, Jerome perhaps did not finish this project (though see Kamesar 1993: 53–4); at any rate, the only surviving portions of this effort include the Psalter (the so-called Gallican Psalter), the book of Job, and the Song of Songs, along with prefaces to Chronicles and the Solomonic books (Kamesar 2013: 661). From around 390, he devoted himself to an all-new project: translating the entire Hebrew Bible directly into Latin, which he completed around 405. He also produced translations of Tobit and Judith, from a “Chaldean” (Aramaic) Vorlage, according to his prefaces to these translations (Gallagher 2015). Jerome’s version became well known during his own lifetime, and it would eventually dominate the Latin biblical tradition, especially from the ninth century (Linde 2011: 34–8). Multiple factors led to Jerome’s magnification of the Hebraica veritas as opposed to the Septuagint. We have seen that Greek patristic authors often explained the differences between the Septuagint and the Hebrew as resulting from textual corruption in one of the two textual traditions, but Jerome regarded both to have been preserved in his day essentially intact (Gallagher 2012: 197–203). The differences between the texts are

THE SEPTUAGINT IN PATRISTIC SOURCES 263

owing to the translators, who could err, since translators are not prophets (Praef. Pent. 29-34). While Christians generally trumpeted the Septuagint as the version quoted by the apostles, Jerome was able to identify in the Hebrew text the source of several New Testament quotations of Scripture that had baffled other exegetes (Kato 2013). Probably the best example, Mt. 2:15 quotes a prophet as saying, “out of Egypt I have called my son,” wording which corresponds closely to the MT of Hos. 11:1, while the Septuagint of that verse reads, “out of Egypt I recalled his children” (NETS). Such quotations form the basis for Jerome’s (dubious) claim that the apostles routinely cite the Hebrew text, relying on the Septuagint only when it agrees with the Hebrew (except for Luke, who uses the Septuagint because he knows no Hebrew; cf. Comm. Isa. 28:9-13). Nevertheless, the Septuagint retained an esteemed position in Jerome’s exegetical practice, and even his translation “according to the Hebrew” (iuxta Hebraeos) makes use of the Septuagint quite extensively (Kraus 2017: ch. 4). His translation from the Hebrew so aggravated many in the Christian world that he routinely defended his work in the prefaces to his translations. In his Adversus Rufinum, he reveals that he preached from the Septuagint (Ruf. 2.24). His commentaries constantly refer to the Septuagint, sometimes negatively, often positively, occasionally without judgment. He usually gives a historical interpretation of the Hebrew text and a spiritual interpretation of the Septuagint (Jay 1985: 276–9). Even where Jerome believes the Septuagint text results from mistranslation, Jerome often offers an interpretation for the benefit of its readers (e.g., Comm. Isa. 9:6; Comm. Jer. 2:23-24). Augustine harbored mixed feelings about Jerome’s work (Gallagher 2016). The bishop of Hippo was a lifelong proponent of the inspiration and authority of the Septuagint, based on the traditional arguments developed in the Greek church, particularly the  translation legend (Doctr. chr. 2.15.22) and the notion that the apostles quoted the Septuagint (Epist. 71.6; 82.34-35). He initiated a correspondence with Jerome on the proper Old Testament text for the church, encouraging the translator to produce no translation without the critical signs (obelus and asterisk) signaling differences from the traditional Hebrew—as Jerome had done in his translations of Origen’s hexaplaric Septuagint. (Augustine mentions having seen the translation of hexaplaric Job by Jerome; Epist. 28.2, written c. 394). Continued study and reflection compelled Augustine in later decades to acknowledge the value of Jerome’s translations from the Hebrew text, though he always bestowed the preeminent position upon the work of the Seventy translators. Jerome had claimed that the Seventy translators did not always translate precisely the text in front of them but introduced several changes (Kamesar 1993: 64–7; Gallagher 2012: 202–3). Augustine agreed, but since he—unlike Jerome—considered these translators to be true prophets filled with God’s Spirit, he evaluated these changes positively as imparting new revelation from God. But neither did he want to deny that God had also inspired the Hebrew Bible, and so Augustine developed the idea that God had delivered two inspired texts of Scripture, one in Greek and one in Hebrew, just as he had inspired both Isaiah and Jeremiah to say different things (Civ. 18.43). Augustine uses the example of Jon. 3:4, where the prophet declares the number of days until the destruction of Nineveh (Civ. 18.44). The Hebrew text says forty days, while the Greek text says three. According to Augustine, the Hebrew text reports accurately the words of the historical prophet Jonah, while the Seventy translators altered the number to three in order to prompt the reader to contemplate the three days between the death and resurrection of Jesus. A similar spiritual meaning can also be found in the Hebrew text, due to the forty days between the resurrection

264

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

and ascension of Jesus (Acts 1:3). Thus, while the Septuagint maintains its position in Augustine’s thought and practice as the Christian Old Testament, he recognizes that the Hebrew and the Greek text both contain edifying spiritual content, and both were used by the apostles (cf. Civ. 15.14). The subsequent history of the Latin Bible attests the growing dominance of Jerome’s translations iuxta Hebraeos, not necessarily because the majority of readers found Jerome’s textual arguments convincing but more because of Jerome’s elegant Latin. In the early seventh century Isidore of Seville praised Jerome’s translation as preferable to others due to its accuracy and clarity (Etymologies 6.4.5). In the early ninth century, two editors of the Latin Bible, Theodulf of Orléans and Alcuin of York, both chose to incorporate Jerome’s translations into their editions, reflecting and effecting its increasing prominence (Linde 2011: 34–8).

VIEWS AND DEBATES The enormity and profundity of patristic literature ensures an unceasingly fertile field of research. In many cases, the fundamental task of literary scholarship—editing texts—remains to be done, and this lack of critical editions can have a direct bearing on interpretations of patristic engagement with the Septuagint, and even on the textual history of the Septuagint itself. Patristic quotations of the Septuagint have served as one of the linchpins in unraveling the complicated textual history of the Greek Old Testament, but this work is hampered by unreliable editions of patristic writings. As one example among many, Ceulemans (2016) has recently called upon scholars to give attention to the manuscript tradition of the Psalms commentary by Theodoret of Cyrus, since this commentary is an essential witness to the Antiochian text of the Psalter and thereby holds immense importance for determining the text of the Old Greek Psalter. While a century of textual scholarship on both the Septuagint and patristic literature has finally rendered inaccurate Hatch’s lament from 1889 that “The quotations from the LXX. in the Greek Fathers are an almost unworked field” (133; later quoted approvingly by Swete 1900: 406), this task always requires more researchers than those who are willing to devote their energy to it. Swete pointed to a lack of critical editions of patristic texts as partly to blame for the situation in his day; he would rejoice to see the progress of the major series of patristic texts today, especially Die griechischen christlichen Schriftsteller der ersten Jahrhunderte (GCS; now published by de Gruyter), the Greek and Latin series of Corpus Christianorum (Brepols), and Sources chrétiennes (Cerf ). Still, too often Migne’s Patrologia Graeca (PG) and Patrologia Latina (PL) from the mid-nineteenth century provide the most recent edition of a patristic text. The secondary task of using early Christian literature to establish the text of the Septuagint always benefits from continuing analysis and reflection. An example concerns the text of Ps. 39:7, ὠτία δὲ κατηρτίσω μοι (“ears you prepared for me”) as it reads in Rahlfs’s edition (1931: 143), more-or-less corresponding to the MT. But all Greek manuscripts of the Psalter attest σῶμα instead of ὠτία, thus agreeing with the quotation in Heb. 10:5, where the psalm serves as a prophecy of Christ. Dorival (2016b: 30–2) has recently demonstrated that patristic literature attests knowledge of three readings for the psalm verse—ὦτα, ὠτία, and σῶμα—but, against Rahlfs, Dorival argues that ὦτα enjoys the earliest attestation whereas the diminutive ὠτία is an innovation of “the Three.” As a further area of ongoing research, we can highlight the recent surge of interest in patristic biblical interpretation that has clarified early Christian exegetical methodology

THE SEPTUAGINT IN PATRISTIC SOURCES 265

and hermeneutics. A list of works that in the last two decades have explored how various early Christians interacted with the Bible would include Martens (2012) and Heine (2010) on Origen, Ernest (2004) on Athanasius, Graves (2007) on Jerome, but there are many other such studies on various patristic authors going back decades. Most of these works have to some extent examined the way Scripture in Greek influenced the theology and exegesis of the author under discussion, and they provide models for similar treatments of other patristic figures. While we have seen that patristic authors often considered the Septuagint a specially revealed edition of the Bible for the church, modern scholars have understandably been much less ready to describe the Septuagint as an inspired translation. Nevertheless, with the increased attention on the Septuagint in recent decades has come an increased appreciation for the patristic view of this text. This appreciation has only been aided by the discovery of the Dead Sea Scrolls, which have demonstrated both that the protoMasoretic Text was not as dominant in Antiquity as scholars might have thought and that the Septuagint differs from the Masoretic Text (at times, at least) because it was based on a different Hebrew Vorlage. This recognition has increased in the minds of some Christian theologians the plausibility of the patristic argument that use of the Septuagint within the New Testament really amounted to apostolic approval of the translation. If there were multiple available text forms in Antiquity, then perhaps Paul’s—or Luke’s—use of a particular text form represents a divine sanctioning of that form. If the New Testament substantiates its theological claims by citing the text of the Septuagint, even where it diverges from other textual forms, then can modern Christians continue to adhere to those theological claims without also adhering to the text that undergirds them? Looking more to the patristic contribution itself, we might ask the same sort of question: if the formative theological debates of the second through fifth centuries were conducted in constant dialogue with the Septuagint, so much so that the triumphant orthodoxy formulated its doctrines in reliance on the language of this particular version, to what extent does this fact obligate Christians who affirm these formulations to accept also the Septuagint as their Scripture? These questions have been asked for decades (Benoit 1963/1964; Barthélemy 1978: 111–26; Müller 1996) but have recently become more prominent within biblical scholarship (Johnson 2002: 11–23; Hays 2014: 107–8; cf. Provan 2017: chs. 10–11). My own contribution to this discussion has been to emphasize the novelty of Augustine’s approach to the matter (Gallagher 2016). Augustine developed the idea that the Septuagint differed from the Hebrew text in some places because God willed it so; in these passages, the Seventy translators received divine guidance to issue new revelation, whereas in other passages they were inspired to produce an accurate translation (Civ. 18.42-44). We saw earlier that Origen had hinted at this sort of idea, but only Augustine fully expressed it, and he developed this view in conversation/dispute with Jerome, who emphasized the authority of the Hebraica veritas far more forcefully than any previous Christian. Some scholars have read earlier Christian literature, I would argue, through the lens of Augustine, as if patristic authors routinely assumed that the Septuagint differed from the Hebrew Bible because God inspired them for this purpose. On the contrary, it seems to me, Christians before Augustine argued that God inspired the Seventy translators for the purpose of producing an accurate version, but they did not have a Jerome pushing them to rethink this assumption (Gallagher 2013b; 2020). This discussion can move forward only through detailed examination of individual authors with these questions in mind.

266

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

RESEARCH METHODOLOGIES The Bible is so pervasive in every form of patristic literature that the main method for studying the Septuagint in this literature is simply to read the literature. A study examining the use of the Septuagint in a particular early Christian writer or work will not lack material. Sometimes, as we have seen, these patristic authors discuss the version of the Bible that they are using, but often they do not. In either case, the researcher will want to investigate the textual form of the quotations, making use of the various tools available, including especially the Göttingen editions of the Septuagint. The situation becomes a little more complicated for Latin authors, particularly in regard to those biblical books for which the Vetus Latina Institute (Beuron) has not yet produced an edition. Moreover, in the centuries after Jerome, his translation from the Hebrew gradually replaced the Septuagint as the dominant text of the Old Testament, but that development for the most part occurred only after the patristic era. Beyond the textual analysis of quotations, approaches to the patristic use of the Septuagint include investigation of exegesis and hermeneutics, which can be interesting particularly in those cases in which the textual form quoted diverges from other attested forms. Online tools have aided the process of tracking down quotations of scripture in patristic literature. The BIBLIndex website (https://www.biblindex.info/) is replacing and moving beyond the older print collection Biblia Patristica, which in seven volumes (with a supplement on Philo) provided an index to biblical quotations in a number of patristic authors. To see the actual patristic texts referenced at BIBLIndex, one could turn to the Thesaurus Lingae Graecae and the Library of Latin Texts (Brepols), both subscriptionbased services that provide online access to an enormous amount of literature in their respective language traditions. The TLG and LLT databases also allow for searching for specific words or phrases throughout the corpus covered or in specific authors and works. Of course, these tools are based (usually) on the best available edition of the patristic work, and we have already observed that the textual base of such editions can sometimes be problematic.

ONGOING RESEARCH QUESTIONS Though the issue of the textual form employed by early Christians has been a research question for centuries, there is still the need for continued work in this area, due in part to the vastness of the corpus and in part to the problem of proper editions of patristic works. There remains the need to explore further the biblical texts quoted in the literature and how the specific text affected the theological appropriation of Scripture. While there are a number of recent studies on the biblical interpretation of individual authors (as we have seen), it would be useful to have more focused treatment on the set of issues outlined in this chapter—which text form did the author quote, and how did this form affect his appropriation of the passage?—and treatments covering a larger swath of patristic literature. And to the extent that the patristic works allow for it, we need studies of the theoretical reception of the Septuagint in its relation to other texts—not just whether early Christians thought it was the best translation, but why they thought this, and how they sought to establish the point. Such questions should be pursued for individual patristic authors and—in some cases—for individual patristic works. (For a brief survey of patristic authors with these questions in mind, see Gallagher 2012: 173–209.)

THE SEPTUAGINT IN PATRISTIC SOURCES 267

Only detailed and systematic study over a wide selection of authors can resolve certain questions that thus far have mostly received summary treatment based on general impressions. How much patristic exegesis and theology is based on a biblical text specific to the Greek tradition, or to a textual base diverging from the Masoretic Text? To what extent were the individual authors aware of textual divergences with regard to the quoted passages? In such cases, what effect did this awareness have on their methodology and conclusions? It is not uncommon for an ancient Christian writer to acknowledge the presence of variant readings in the manuscript tradition of the biblical text (cf. Kamesar 1993: 4–40), but there has been no systematic study of patristic discussions of variant readings in the Septuagint tradition. An Old Testament study along the lines of what Amy M. Donaldson (2009) has done for the New Testament would be extraordinarily useful, though such a project would be extraordinarily daunting. In the same vein, early Christians often noticed that certain apparent quotations of Scripture within the New Testament (e.g., Mt. 2:23; 1 Cor. 2:9) proved difficult or impossible to locate within their copies of the Old Testament (see Gallagher 2014). The patristic handling of such problematic quotations and how such quotations affected views of the correct form of the biblical text would also provide a fruitful avenue of research.

RECOMMENDED RESOURCES 1. Gallagher (2012). Attempts to untangle some of the ways that patristic authors conceived of the relationship of the Septuagint to the Hebrew Bible. 2. Harl (1988). A helpful overview from one of the giants in the field. 3. Kamesar (1993). According to Alison Salvesen (2005: 131), “Kamesar’s study is certainly the one work that I would insist that any graduate student working on Origen’s Hexapla, the Christian exegetical tradition, or Jerome’s Iuxta Hebraeos version should read from cover to cover, and more than once”. 4. La Bible d’Alexandrie, Paris: Cerf (1986–). This essential series of translations of the Septuagint into French includes extensive introductions and notes, routinely incorporating patristic interpretation and thereby providing convenient access to many scattered comments. 5. Pentiuc (2014). Pentiuc is a professor within the Greek Orthodox communion, and here he explains the position that the Septuagint has enjoyed, and continues to enjoy, in that religious tradition. 6. Wasserstein and Wasserstein (2006). As its title indicates, this monograph, with extensive quotations from ancient and medieval sources, traces the translation legend of the Septuagint from its origins to modern times.

268

CHAPTER 18

The Septuagint in Byzantine Judaism CAMERON BOYD-TAYLOR

Between the transfer of the Roman capital to the city of Byzantium in 330 CE and the fall of Constantinople in 1453, there was a continuous Jewish presence in many of the principal administrative centers of the Byzantine empire.1 Greek was the spoken language of the vast majority of Byzantine Jews throughout this period (de Lange 2010: 40).2 At the outset it also served as their primary written language, and the scriptures were read and interpreted in Greek. By the sixth century, however, a shift toward Hebrew book learning was under way, and by 1000 CE Hebrew had essentially replaced Greek as the language of Jewish worship and written culture (Jacoby 2009: 162). Nevertheless, Byzantine Jews continued to read the scriptures in Greek. Although no Jewish literary source refers unambiguously to the practice (de Lange 2010: 45), the ongoing use of Greek biblical texts by Jews is attested by a wide range of documents, and represents a significant, though largely neglected, area of research within biblical studies.3 The present chapter provides a brief historical review of the subject, outlines the principal scholarly debates and methodological issues, and identifies a number of avenues for future research.

INTRODUCTION That Greek scriptural texts were employed by Jews in Caesarea on the threshold of the Byzantine era is certain, and the inference to other Greco-Jewish centers is secure.4 Although the prestige of the Hebrew language was growing, use of it as a vernacular had virtually disappeared, and translation was thus an essential aspect of Jewish education (Alexander 1999).5 Rabbinic sources indicate that monolingual reading of the scriptures prevailed in those areas where Greek speakers were predominant (Smelik 2007: 141). In For a concise historical overview see Jacoby (2009), and Bowman (2014). More extensive treatment is provided by Starr (1939), Sharf (1971) and Bowman (1985). Panayotov (2014) offers a useful demographic survey. 2 The evidence for Jewish knowledge of Greek comes primarily from the frontiers of the empire, that is, the Near East (Syria, Palestine, and Egypt), and southern Italy (Bowman 2014: 38). 3 In 2006, Nicholas de Lange secured funding for a research project based in Cambridge, the Greek Bible in Byzantine Judaism (GBBJ), which assembled a corpus of texts in the form of a digital edition. Documents in the corpus are numbered and identified as either glosses (e.g., G1) or continuous translations (e.g., T1), and will be referred to accordingly in this chapter. The corpus is available online at https://www.gbbj.org/. 4 A Talmudic story (y. Sot. 7.1, 21b) referring to the recitation of the Shema in Greek bears witness to its use in Caesarean synagogues in the latter part of the third century CE. 5 The process is described by Theodoret, Questions on Genesis, 61. 1

270

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

deference to this state of affairs, a number of prominent sages explicitly sanctioned the use of Greek versions, including R. Simon ben Gamaliel in the second century, as well as Bar Qappara and R. Abbahu of Caesarea, both active in the third century.6 Christian apologists could still trade on the assumption that the Greek scriptures were being read in the synagogue (Boyd-Taylor 2010a: 275).7 It is reasonable to posit a situation of textual pluriformity at this time. That much is evident from the Hexapla, a polyglot bible produced by the Christian scholar, Origen of Alexandria (185–254 CE), and an unparalleled witness to the state of the Greek Bible in the late Tannaitic period. The so-called younger versions incorporated within the Hexapla—including Aquila, Symmachus, and Theodotion (often known as “the Three”)— are known principally through lexical variants recorded in Christian manuscripts of the Septuagint, but each had Jewish origins, and each presumably enjoyed some degree of use by Greek-speaking Jews.8 According to Origen, Aquila was the preferred version among Jews without Hebrew in his day, an observation confirmed by both Jerome and Augustine.9 Aquila’s translation was also consulted in the rabbinic academies, and is explicitly quoted by Talmudic authorities (Tov 2005: 394).10 Yet to conclude that Aquila had completely displaced the Septuagint by this time is unwarranted. A third-century papyrus codex containing the Septuagint version of Genesis 2-3 is in all likelihood Jewish (Kraft 2003: 61).11 Moreover, Origen’s text of the Septuagint, which reflects partial revision toward the Hebrew, may itself have been based on current Jewish exemplars (Boyd-Taylor 2018: 372). The period from the fourth century CE until the Arab conquests of the early seventh century was characterized by a significant degree of social and political continuity in Byzantium. The widespread use of Greek biblical versions by Jews seems altogether likely. Yet it is weakly documented. Quite simply, the reading practices of Greek-speaking synagogues at this time are unknown (de Lange 2015: 59). There are at present no securely identified Jewish manuscripts of the Septuagint beyond the third century CE,12 and while two fragments of Aquila’s version dated to the sixth century were likely copied by Jewish scribes (T1. Fragments of 1-2 Kings and T4. Fragments of Psalms),13 this is by The Mishnah (m. Meg. 1.8) attributes to R. Simon ben Gamaliel the opinion that Greek was the one language, beside Hebrew, in which the scriptures were permitted to be written. Bar Qappara is said to have interpreted Gen. 9:27 as meaning: “May the words of the Torah be recited in the language of Japheth [i.e., Greek] in the tents of Shem [i.e., Israel] (Gen. Rab. 36.8).” R. Abbahu (b. Meg. 9b), on the authority of R. Yoḥanan, confirmed the opinion of R. Simon ben Gamaliel that it was permissible to write the scriptures in Greek. He also permitted the teaching of Greek to one’s daughter (y. Šabb. 6.1). 7 Tertullian (c. 160–230 CE) implies that the Jews continued to read the Septuagint publicly (Apol. 18.8). The Cohortatio ad Graecos 13, written sometime after 260 CE, likewise implies that the Septuagint could be found in any synagogue. 8 On the Hexapla and the major recensions, see Gentry and Mäkipelto in the present volume, respectively. 9 Origen, Ep. Afr. 2; Jerome, Comm. Ezech. 3.5; and Augustine, Civ. 15.23. 10 The quotations occur in the Palestinian Talmud, Genesis Raba, Leviticus Rabba, Shir Hashirim Rabba, Echa Rabba, Esther Rabba and Qohelet Rabba, but not in the Babylonian Talmud. For a philological analysis, see Veltri (2002: 83–92). Typically, the Greek rendering occurs in Hebrew transliteration followed by its Hebrew translation. Alexander (2014: 243) points out that these lists do not necessarily attest to the circulation of complete versions of Aquila in rabbinic circles. 11 P.Oxy. 1007 (= P.Lit.Lond. 199. British Library, London.) 12 For a survey of the Jewish manuscript evidence see Kraft (2003). 13 T1. Cambridge UL, T-S 12.184; T-S 20.50. Fragments of a palimpsest. Top writing: a Hebrew liturgical text. Bottom writing: A Greek manuscript of Kingdoms (3 Kgdms 21:7-17; 4 Kgdms 23:11-27). Edited by Burkitt (1897). T4. Cambridge, UL, T-S 12.186 + Cambridge, UL, T-S 12.187 + Cambridge, UL, T-S 12.188. Fragments of a palimpsest. Top writing: Yerushalmi, Ta’aniyot and Mashqin. Bottom writing: A Greek manuscript of Psalms, including Pss. 90(89):17-92(91):10; 96(95):7-12; 98(97):3; 102(101):16-103(102):13. Edited by Taylor (1900). 6

THE SEPTUAGINT IN BYZANTINE JUDAISM 271

no means certain. What little epigraphic evidence there remains is, however, consistent with a situation in which both Aquila and the Septuagint were in use (Cappelletti 2009: 138). External support for this hypothesis comes from an unlikely witness. On February 8, 553, Justinian I, Byzantine emperor (527–565 CE), issued an edict (Novella 146 Περὶ Ἑβραίων) authorizing the use of both versions by his Jewish subjects. The edict ostensibly addresses a Jewish petition (de Lange 2015: 64). Taken at face value, this would suggest that the reading of Greek scriptural texts in the synagogue had been contested by certain Hebrew-literate Jews. This era saw the consolidation of rabbinic Judaism, and with it a growing resistance to the scriptural authority of the Septuagint. The trend is attested in various rabbinic sources by the appearance of lists of purported textual alterations by the “Seventy.”14 It is telling that these translations are said to have been written for King Ptolemy—that is, either for his sake or on his behalf—which distances them from authentic Jewish tradition. In this regard, a post-Talmudic tradition compares the translation of the Septuagint to making the golden calf.15 Apparently there was an attempt to mark the event liturgically by a day of fasting (Alexander 2014: 242). Whether the object of this polemic was the status of the Septuagint as a written text (ketav) or its didactic use in oral form (targum) is uncertain (Veltri 2009: 148). Despite such a negative attitude there is no unequivocal evidence that the Septuagint was ever formally proscribed by rabbinic authorities.16 With the Arab conquests came the political separation of Egypt, Palestine, and Syria from the empire.17 The subsequent two centuries are often referred to as a dark age in Byzantine history due to the almost complete lack of sources (de Lange 2015: 26). The continuing significance of the Septuagint for Greek-speaking Jews can, however, be glimpsed in the excavation of the one known synagogue building in Caesarea. The inscriptions, which include scriptural quotations, are almost exclusively Greek. Of particular interest is the dedication of the synagogue floor, which opens with a quotation from the Old Greek version of Isaiah. The archaeological context and orthographic features point toward the seventh century CE (Roth-Gerson 1987: 111–24). Thus, even in Caesarea, where rabbinic influence was considerable, the Septuagint still played a role in the life of the worshipping community. The importance of the Greek language for biblical lexicography during this period is witnessed by fragments of a Hebrew–Greek glossary dated to the ninth century CE (G4. Exodus and Jeremiah).18 Written in Greek script, the glosses were evidently drawn from some form of learned tradition, as significant affinities with Aquila have been noted. From this one may infer the ongoing transmission of Greekspeaking Jewish scholarship (de Lange 2007: 37). Byzantine Judaism of the late tenth century to the beginning of the thirteen century is much better documented than the preceding period. The territorial reassertion of

The principal sources are: b. Meg. 9a; y. Meg. 1, 1, 4; Mek. Exod. 12:40; Midr. Hagadol Exod. 4:20; ʾAbot R. Nat. version B, 37; Soferim 1.7; Yal. Shimoni Genesis 3; Midr. Tan. Exodus §22 (see Tov 1984: 65). Alexander (2014: 240) suggests that the scholarship underlying the lists is more likely scribal than rabbinic. 15 Soferim 1.7 which is probably not earlier than the eighth century (de Lange 2015: 3). See also Sefer Torah 1.6. Veltri (2009: 149) concludes that both witnesses are post-Talmudic. 16 Veltri (2006: 104–46) has made a plausible case for rabbinic opposition to the Septuagint in the post-Talmudic period, but to infer outright rejection is risky. Tov (2005: 396) rightly points to the complexity of the evidence. A crux is b. Meg. 9a–b, which juxtaposes two contradictory evaluations of the Septuagint. 17 Despite this rupture the Jewish communities of this region retained many of their connections with Byzantium. 18 Cambridge, UL, T-S F 17.4. A bifolio from a medium-sized informal codex. Fragment of a palimpsest in which the Talmud has been written over a Hebrew–Greek biblical glossary in parallel columns (Exodus 5-28; Isaiah 66; and Jeremiah 2-38). Edited by Tchernetska et al. (2007). 14

272

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the empire in the east had created opportunities for immigrants (including members of the Karaite movement), and Greek-speaking Jewish communities, some newly recolonized, flourished (Bowman 2014: 43).19 A number of cultural shifts are apparent. Hebrew was now firmly established as a liturgical language (Jacoby 2009: 162), with obvious implications for the use of Greek versions. At the same time, exclusion from Christian schools had resulted in a precipitous decline in Greek literacy among Jews, a phenomenon witnessed in changing scribal practices. From the tenth century onward Jewish scribes typically write Greek in Hebrew characters, and the text is embedded in what are essentially Hebrew books, while the morphology and syntax is adapted to the norms of the vernacular, and the spelling becomes increasingly phonetic (de Lange 2008: 114).20 Yet the use of Greek by Hebrew authors is widespread. The eleventhcentury Rabbinate commentator Tobias ben Eliezer employs Greek glosses liberally in his Lekah Tov, as do many twelfth-century Karaite commentaries (Bowman 2014: 51). Hebrew–Greek biblical translation during this period is firmly attested by a range of primary sources, the bulk of which were recovered at the end of the nineteenth century from the Ben Ezra Synagogue in Fustat, Old Cairo (the so-called Cairo Genizah), and are now in Cambridge.21 The Geniza texts bespeak a Sitz im Leben in which the Greek language played an important but ancillary role in biblical study and worship (de Lange 2008: 113). They fall primarily into one or the other of two classes: (1) Greek glosses in Hebrew biblical manuscripts;22 and (2) Greek notes embedded within scholia or commentaries written mostly in Hebrew.23 The exception is a fragment from a Greek version of Ecclesiastes written in Hebrew characters.24 Certain paratextual features of this text are consistent with liturgical use, and it may derive from a prayer book for the Feast of Tabernacles or Sukkot (for which it is a prescribed reading). Each verse is preceded by its opening word in Hebrew, suggesting the practice of verse-by-verse translation. Although the Greek text exhibits some degree of contact with the ancient versions, its lexical stock is primarily vernacular, a pattern exhibited by other contemporary sources.25 Karaite migrants settled alongside existing Jewish communities. The degree of their acculturation to the Byzantine milieu is witnessed by the appearance of Greek vocabulary in their scholarship (Bowman 2014: 51). 20 When Greek characters do appear, they are crudely formed majuscules. De Lange (2008: 112) notes that the progression from Greek to Hebrew script was not straightforward. 21 Approval to bring the manuscripts to Cambridge was secured by Solomon Schechter in 1897 with the financial support of Charles Taylor. The Taylor-Schechter collection represents the most important assemblage of medieval Jewish documents in the world. See Reif (2000) for an account of the discoveries. 22 G11. Proverbs. Oxford, Bodleian Library, MS Heb. e. 43, fol. 51. Hebrew manuscript with Greek interlinear glosses (Prov. 17:16–19:3). Edited by Rüger (1959). Compare G5. Fitzwillam Greek Glosses. Cambridge, Fitzwilliam Museum, MS 364*. Hebrew Manuscript of the Former Prophets (Josh. 3:10-2 Kgs 25:15) with Hebrew and Greek scholia. See Olszowy-Schlanger (2003) and de Lange (2003). See also G6. Judges. Cambridge UL, T-S NS 250.7-8. Two fragments with Greek glosses. 23 G1. Genesis and Exodus. Cambridge, UL, T-S C6.117 + Westminster College, Talmudica I.110. Edited by de Lange (1996: 85–116). Hebrew scholia with Greek glosses. G3. Genesis and Joshua. Cambridge, UL, T-S C6.133 (part). Edited by de Lange (1996: 117-26). Hebrew scholia with Greek glosses. G7. 1 Kings no. 1. Cambridge, UL, T-S C6.133 (part) + Ox. Bod, MS Heb. d. 43, fol. 25-26. Hebrew commentary with Greek glosses (1 Kings 7:25–10:21). Edited by de Lange (1996: 127–154). G8. 1 Kings no. 2. Cambridge, UL, T-S K24.14. Hebrew philological notes with Greek glosses (1 Kings 6:20–8:37). Edited by de Lange (1996: 155–63). G9. Ezekiel and the Twelve. Jerusalem, JNUL Heb. 4o 577.7/1 + Cambridge, UL, T-S C2.87 + T-S F2(1).211+ T-S 32.1 + T-S K27.46 + T-S K25.288 + T-S K27.47. Edited by de Lange (1996: 165-294). Hebrew commentary containing Greek words and phrases. G10. Malachi and Job. Cambridge UL T-S NS 309.9. Hebrew–Greek glossary covering selected words. Edited by de Lange (1996: 79–84); see also de Lange (1980). 24 T5. Fragment of Qoheleth. Cambridge, UL, T-S Misc.28.74. Edited by de Lange (1996: 71–8). 25 See Gentry (2015) for a textual critical evaluation of the fragment. 19

THE SEPTUAGINT IN BYZANTINE JUDAISM 273

The period between the Latin conquest of Constantinople in 1204 and its fall to the Ottoman Turks in 1453 was a time of unprecedented social upheaval in the Byzantine world, not least of all for its Jewish inhabitants. Due in part to the political fragmentation, there was now a broader diffusion of Jewish communities, combined with greater concentration in the centers of trade (Jacoby 2009: 166).26 A diverse array of Hebrew manuscripts has been dated to this period, including numerous prayer books testifying to the local rites that had developed over the preceding centuries (de Lange 2015: 122). In the Byzantine (or Romaniote) rite, Hebrew is the primary language, but there are Greek elements, especially in the hymnic compositions. Evidence for the use of Greek biblical texts at this time comes from a version of the book of Jonah (the prophetic reading for the afternoon service of the Day of Atonement) preserved in two prayer books from Crete (T2. and T3. A Greek Translation of Jonah).27 The text is written in Hebrew characters, and the language is medieval Greek with some archaic features.28 For further clues to Jewish reception at this time one may look to two intriguing manuscripts of uncertain provenance. Graecus Venetus VII, a fourteenth-century codex, contains a late medieval Greek version of the Pentateuch, Proverbs, and the Megilloth (except Esther and Daniel).29 Vaticanus Graecus 343, which bears the date April 22, 1450, is a medieval Greek translation of the Psalter.30 In both cases Jewish antecedents are likely. With the advent of Ottoman rule came a measure of regional stability, as the Balkans were again united under one empire. Yet the new political order brought in its wake profound demographic changes for the Jewish population. As part of the Ottoman policy to repopulate the capital, large numbers of Greek-speaking Jews were forcibly transferred from the provincial cities to Constantinople. Shortly thereafter there was a mass, eastward migration of Sephardic Jews who were expelled from Spain (Jacoby 2009: 181).31 Romaniote tradition survived primarily in the Venetian sphere,32 and in Constantinople, which remained the center of Greek-speaking Judaism (both Rabbinate and Karaite) throughout the Ottoman period (Bowman 2014: 44). A key witness to this period is Elijah Kapsali of Candia (1490–1555), who refers to the reading of a Greek translation of the Book of Jonah on Yom Kippur in the synagogues of Candia and Corfu. Around the same time an ambitious publishing venture was undertaken in the capital by the Italian Jewish printer Eliezer (Albert) Soncino.33 The so-called Constantinople Pentateuch, a polyglot edition printed in 1547, contains in addition to the Hebrew text parallel Greek and Judeo-Spanish (Ladino) translations together with Targum Onkelos and Rashi. The Greek version, which is written in Hebrew characters, almost certainly draws upon earlier Romaniote sources. As such it represents not only a monument of the

The consolidation of Venetian and Genoese interests in Constantinople and the Aegean region was a key stimulus to Jewish migration (Jacoby 2009: 177). 27 T2. A Greek Translation of Jonah no. 1. Bodleian Library, MS. Opp. Add. 8o 19 (Neubauer 1144). T3. A Greek Translation of Jonah no. 2. Florence, University Library, Ms Ebr. 3574 (Modona 12). Prayer book for Pentecost, New Year, and the Day of Atonement containing a Greek version of Jonah. Edited by Hesseling (1901). 28 In one book the Greek follows the Hebrew, while in the other only the first and last verses of each Hebrew reading are given (de Lange 2008: 113). 29 Edited by Gebhardt (1875). 30 See Devreesse (1937: 18). 31 While in the past Romaniote congregations had successfully assimilated migrant Jews from both the Near East and Latin West, the rapidity of the influx after 1492 precluded this (Jacoby 2009: 179). 32 Areas conquered by the Ottomans after 1455 retained their Romaniote communities (Bowman 2014: 44). 33 The Constantinople Pentateuch was produced during the reign of Suleiman al Qanuni (1520–66). 26

274

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Greek language, but the culmination of a tradition of Hebrew–Greek translation with ancient roots (Fernández Marcos 2009a: 50).

VIEWS AND DEBATES From its inception in the nineteenth century, critical study of the use of Greek biblical versions by Byzantine Jews has revolved around the thorny question of transmission. During the middle ages new Greek translations appear, which, despite their contact with antiquity, are for the most part vernacular. It would seem that these medieval texts superseded the Koine Greek versions of the Hellenistic and Tannaitic periods (the Septuagint and the Three). Attempts to understand this shift have occasioned four principal areas of debate: (1) The terminus ante quem for Jewish scribal transmission of the Koine versions; (2) The characteristic form of the medieval Greek translations to which they yielded; (3) The connection of these new versions with their predecessors; and (4) Their affinity with early-modern sources.

Jewish Transmission of the Ancient Versions It is relatively certain that at some point subsequent to the Tannaitic period Jewish scribes stopped making copies of the Koine Greek versions, yet when this occurred is difficult to establish. While the sixth century CE is widely cited as a terminus ante quem on the evidence of Justinian’s Novella, the interpretation of this document is fraught with difficulties. It is questionable, for instance, how well informed Justinian’s advisors were on Jewish matters. Moreover, there is the possibility that the Novella represents Christian propaganda.34 On balance, however, it seems unlikely that it is an outright fabrication. De Lange (2010: 43) proposes that it arose in response to the efforts of a Hebraist party to impose the Hebrew text on congregations that had hitherto read the Bible in Greek.35 As to which version was in use, one can only speculate. The manuscript evidence, such as it is, would favor a textual descendent of Aquila. The Aquila fragments in the Taylor-Schechter collection (T1. Fragments of 1-2 Kings, and T4. Fragments of Psalms) figure prominently in the discussion of Jewish transmission. The original manuscripts, written in fine Greek majuscules, were likely produced in the Near East sometime between the fifth and seventh centuries. Most scholars follow the first editors in assuming a Jewish provenance. Fernández Marcos (2000b: 113) infers that they were still in the possession of Jews when they were turned into a palimpsest in the eleventh century. More plausible is de Lange’s (2010: 53) suggestion that they were discarded long before that time. This would be consistent with de Lange’s overall picture of Jewish reception, according to which the ancient versions gradually gave way to vernacular translation in the period following the Arab conquests. The manuscripts are part of a larger group of fourteen palimpsest fragments with Hebrew upper texts and Greek lower texts that were recovered from the Cairo Geniza.36 It

See Veltri (1994a) and Rutgers (2003). There is an extensive literature on the Novella. See de Lange (2005). 36 Sokoloff and Yahalom (1978) have demonstrated that the Hebrew quires consisted of leaves put together at random from the most usable remnants of the Greek manuscripts. 34 35

THE SEPTUAGINT IN BYZANTINE JUDAISM 275

turns out that some of the lower texts are from the New Testament and hence undeniably Christian. According to Sokoloff and Yahalom (1978), the Aquila manuscripts must also have been copied by Christians, since reverence for the written word would have prevented Jewish scribes from writing over Jewish works. If they are correct, then there is no firm textual evidence for Jewish transmission of any of the ancient versions subsequent to the time of Origen, a consequence with far-reaching implications.37 Their argument has, however, been challenged by Tchernetska (2002: 251), who discovered that at least one of the lower texts among the Genizah palimpsests is undoubtedly Jewish. Henceforth the question of provenance must be decided on other grounds. While it remains prima facie unlikely that Jewish scribes would knowingly copy over a biblical text, it is doubtful that the Aquila manuscripts were intact when they came back into Jewish hands. The fact that the Greek translation is that of Aquila rather than the Septuagint points in the direction of Jewish production, as does the scribe’s use of paleo-Hebrew letters for the tetragrammaton.38 Admittedly neither consideration is conclusive (Gallagher 2013a). How this debate is settled will have considerable bearing on the story of Jewish reception.

The Form of Medieval Hebrew–Greek Translation The bulk of the evidence recovered to date comprises isolated Greek glosses on the Hebrew text. Fernández Marcos (2009a: 48–50) has suggested that the glosses were drawn from book-length Jewish translations that circulated in written form during this period, for which he posits a two-fold typology: (1) Texts in the tradition of Aquila, which functioned in a subsidiary role with respect to the Hebrew, perhaps in an oral milieu; and (2) Texts in the tradition of Symmachus, intended to satisfy a literate audience. As evidence for the latter, he points to Graecus Venetus VII.39 There is indeed a degree of typological resemblance to Symmachus. Aslanov ( 1999a: 173) observes that the medieval version combines fidelity to the Hebrew with a need to produce literary Greek.40 Leaving aside the Byzantine translator’s identity (which remains a vexed question),41 what is significant for the present inquiry is the likelihood of his dependence on contemporary Jewish sources.

The possibility of Christian provenance cannot be excluded. Jerome in his letter to Rufinus acknowledges that the latter acquired texts of Aquila at great cost. Whether they were copied by Jews is unknown. 38 Ceulemans (2012) finds no decisive evidence that Christians of this period had access to the Three other than through extracts from the Hexapla. On scribal use of Paleo-Hebrew letters, T1 uses these characters for the Tetragrammaton (in the form YHYH [sic]) except in one instance, at the end of a line, where it is replaced by the Greek letters ΚΥ (an abbreviation of κυρίου) due to space limitations (de Lange 2015: 161). In T2 the Tetragrammaton is consistently written in paleo-Hebrew script. 39 Another textual witness adduced by Fernández Marcos (2009a: 47) for this period is a Greek version of the Psalter (Vaticanus Graecus 343). Yet, as he concedes, there is a strong likelihood that it was directly translated from a Christian copy of the Septuagint version (it includes Odes 7, 8, and 9 which lack a Hebrew Vorlage). To what extent, if any, it draws upon a medieval tradition of Hebrew–Greek translation remains uncertain. 40 The translator of Graecus Venetus VII often makes use of classical and poetic vocabulary in place of the Hellenistic Greek of the Septuagint. 41 Whereas Gebhardt (1875: lxv–lxix), who published an edition of the manuscript, assigned it to a Karaite milieu, Delitzsch, in his preface to Gebhardt’s edition, identified the translator as Elisaeus Judaeus, a Byzantine Jew known from other sources (see Gebhardt 1875: xi); Mercati (1916) identified him as Simon Atumanus, a fourteenth-century Christian bishop uniquely credited with mastery of Greek and Hebrew. Aslanov (1999a: 156–8) rightly questions whether Christian authorship accounts for the Jewish characteristics of the translation. 37

276

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

In a detailed study of Canticles in Graecus Venetus VII, De Crom (2009: 298) concludes that the Greek text bears traces of the exegesis of both Rashi and David Kimchi. Moreover, the translation appears to be indebted to Aquila’s version not only for the translation and interpretation of individual passages, but also for its lexical stock. In De Crom’s (2009: 300) estimation, the extent and nature of these traces of Aquila confirm the Jewish provenance of the translation. It is doubtful, however, that a case can be made for the influence of Symmachus, putting a question mark beside the twofold typology of Fernández Marcos (insofar as it is historically motivated). Dependence upon a tradition linked to Aquila would be consistent with the evidence of other medieval Jewish sources. As to the existence in the middle ages of written book-length versions akin to Aquila, there can be no doubt. One such text is extant in two copies (T2. and T3. A Greek Translation of Jonah), and there is a portion from what was likely another (T5. Fragment of Qoheleth). The remainder of the textual evidence, however, comprises principally glosses, and while they testify unambiguously to the ongoing practice of Hebrew–Greek translation, the question of transmission is distinct, and cannot be determined a priori. It thus continues to be a matter of debate whether the medieval texts (taken as a whole) presuppose the circulation of complete biblical manuscripts.42 That they often employ words or forms that were no longer in the spoken language is admittedly significant, as Jews of this era had no formal Greek education. On the other hand, there are tell-tale features of orality.43 De Lange (2010: 51) conjectures that Byzantine Jews were taught to translate Hebrew readings into Greek orally, using words and phrases from the ancient version, but extemporizing to some extent.

Continuity of the Medieval Greek Translations with the Ancient Versions Another source of debate is the influence of the ancient versions on the medieval glosses. Some degree of continuity is probable, as an exact correspondence is observed in many instances. Considerations of both lexicography and translation technique suggest that certain Hebrew–Greek matches established in antiquity were carried forward in Jewish tradition. Nevertheless, the task of establishing these links and identifying their textual character has occasioned controversy. Blondheim (1924) cited a Septuagintal reading from the Arukh, a Talmudic dictionary compiled in the late eleventh century by Nathan bar Yehiel of Rome that contains Greek glosses.44 Another interesting example comes from a late medieval translation of Jonah.45 Given these suggestive correspondences, there is a tendency in the secondary literature

See Aslanov (1998: 44–5) with reference to G9. Ezekiel and the Twelve (fragments of a Hebrew commentary). First, there is the paucity of biblical manuscripts as opposed to fragmentary quotations; second, the lexical stock of the translations (including the book-length texts) tends to be medieval and colloquial. 44 Nathan notes that the Hebrew word ‫“( טנא‬basket,” HALOT, ad loc.) may be glossed by κάρταλλος (LSJ, ad loc.), which is the rendering of the Septuagint at Deut. 26:2. The word is not well attested in later Greek, nor does it stem from Aquila, who uses ἀγγεῖον (LSJ, ad loc.). 45 T2. A Greek Translation of Jonah 1 and T3. A Greek Translation of Jonah 2. While Hesseling (1901) held that the translation exhibited no trace whatsoever of the Septuagint, Blondheim (1924) identified an exception at Jon. 4:6, where both the Septuagint and the medieval text render the Hebrew word ‫( קיקיון‬which denotes some sort of plant, see HALOT, ad loc.) by κολοκύνθα, typically a gourd or pumpkin (LSJ, ad loc.). The equivalence appears to be unique to the Septuagint, since Aquila, Theodotion, and Symmachus can all be ruled out as sources in this instance. In the Middle Ages a tradition arose in which the plant was identified as a castor oil plant (ricinus) (Boyd-Taylor 2010a: 286). 42 43

THE SEPTUAGINT IN BYZANTINE JUDAISM 277

to emphasize the connection between the Byzantine sources and the Septuagint.46 Yet the evidence generally points instead to Aquila.47 Thus, where Hexaplaric readings are available for comparison, the ninth-century bilingual Hebrew–Greek glossary edited by de Lange and his colleagues (G4. Exodus and Jeremiah) consistently agrees with the Three against the Septuagint, and with Aquila against Symmachus and Theodotion. Consequently, there is a strong likelihood that the text represents Aquila in cases where Hexaplaric evidence is absent (de Lange 2008: 113; Boyd-Taylor 2010a: 284). Notwithstanding the evident legacy of Aquila’s version, there remains a significant amount of material which cannot be attributed to this source. In his textual study of a Geniza manuscript containing glosses on the special vocabulary of Solomon’s building campaign (G8. 1 Kings 2), Law (2009: 282) concludes that although some of the readings can be linked to Aquila with confidence, others occur frequently in the books of the Septuagint, and therefore might not represent any single version. Rather he concurs with de Lange (1995: 263) that such glosses are probably ad hoc translations. It may be that the tendency of modern investigators to emphasize continuity with the ancient versions has obscured the implications of the linguistic evidence. De Lange (2008: 115) stresses that the later translations are texts in their own right. They are not repositories of readings from the ancient versions. Even where there is agreement with Aquila, one cannot assume that the scribe possessed a copy of this version (de Lange 1993: 208). Much of the medieval material currently known to scholarship is decidedly late in both lexical stock and orthography, underscoring the fluidity of the tradition.

The Relationship between Early-Modern Sources and the Medieval Evidence The so-called Constantinople Pentateuch (CP) represents a crucial point of reference for the Jewish transmission of Greek biblical versions. Yet its significance in this regard has been variously interpreted. Hesseling (1897: ii) concurred with Belleli that the text was independent of both the Septuagint and the Hexaplaric versions, suggesting a break with earlier tradition. More recent scholarship asserts its continuity. Fernández Marcos (2000b: 185) views the Greco-Jewish milieu that produced the CP as nothing less than a cultural survival of Hellenistic Judaism. According to Aslanov (1999b: 387) the early modern version both imitates the translation technique of the Septuagint and reacts against it.48 Krivoruchko (2008: 276) concludes her close reading of the colophon of CP with the observation that the Ladino and Greek texts were likely produced ad hoc on the basis of existing oral practices, although she does not exclude a role for translational aids such as

See for instance Aslanov (1998: 44–5) with reference to G9. Ezekiel and the Twelve. There are, however, significant exceptions. Boyd-Taylor (2010a: 284–6) draws attention to an echo of the Septuagint in a biblical manuscript of the Former Prophets dating to the eleventh or twelfth century and containing 123 Greek glosses (G5. Fitzwillam Greek Glosses). At 1 Sam. 17:5 Goliath is described as wearing ‫שריון‬ ‫( קשקשים‬presumably a “coat of mail”). The word ‫ קשקשת‬is typically used in reference to fish scales, and denotes scaled armor only here (HALOT, ad loc.). The Fitzwilliam gloss, ἁλυσιδωτόν (“wrought in chain”; LSJ, ad loc.), is also the rendering of the Septuagint. The Greek word is not well attested in late antiquity or the medieval period, and, more importantly, the influence of Aquila can be ruled out. The Fitzwilliam Bible thus stands alone with the Septuagint. 48 See also Arar (2005) and Chaze (1980). 46 47

278

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

glossaries. While both texts may retain traces of the ancient versions (Krivoruchko 2008: 276), CP is first and foremost a vital witness to late medieval Jewish tradition, something that is underscored by the existence of significant lexical agreements between this version and the marginalia of certain medieval Christian manuscripts of the Septuagint. For Fernández Marcos (2009a: 43) the Christian marginalia represent what is in effect a second channel of transmission for Byzantine Jewish tradition. While there is by no means a consensus regarding its origin, the material is of undoubted textual significance. At the center of the discussion are readings from Codex Ambrosianus.49 The relevant material (collectively referred to as Fb) is written in a cursive script dated to about the twelfth century. There are numerous Hebrew–Greek glosses that are independent of both the Septuagint and the hexaplaric versions. On the assumption that the Hebrew learning they presuppose would not have been available to Christian scribes, it has been argued that they must have originated in a Jewish source. Blondheim (1924: 5), taking up a suggestion by Margolis, recognized the significance of the agreements with CP, and proposed that Fb represents an intermediate stage in a common underlying tradition. This link was confirmed by Wevers (1993a) who, working from a fresh collation of Ambrosianus, documented the textual relationship between Fb and CP in the book of Exodus. Of the 488 readings of Fb, one hundred are shared with CP. Moreover, the lexical correspondences are beyond mere coincidence. Boyd-Taylor (2008b: 32–8) has since documented agreements between Fb and two medieval Jewish sources.50 There can be little doubt that one or more of the Fb correctors drew upon contemporary Jewish tradition, further confirming Blondheim’s interpretation of the textual evidence. Fernández Marcos (2009a: 47) was inclined to think that Fb records the remnants of Jewish translations that were in circulation in the Byzantine period.51 Yet in her study of Fb in Exodus 1-24, Salvesen (2009: 126) raised the possibility that it is instead a “ragbag” collection of readings emanating from different versions and from different periods, a picture largely confirmed by Boyd-Taylor (2014). A comprehensive analysis by Fincati (2016: 426–30) demonstrates conclusively that the revision drew on both Christian and Jewish sources and is a testament to their close interaction.

METHODOLOGY To offer a fully satisfactory historical account of the use of Greek biblical translations by Byzantine Jews, one would want to draw upon a diverse array of witnesses, both direct and indirect. Yet the available evidence, in fact, almost exclusively comprises putative Hebrew–reek translations.52 So the task of scholarship is here principally philological.

G2. Hexateuch. Milan, Ambrosian Library, A 147 Inf. (Martini-Bassi 808). The agreements are between Fb and Greek marginal glosses in a Hebrew manuscript of the Former Prophets (G5. Fitzwillam Greek Glosses), as well as Greek glosses within a series of exegetical and philological notes in Hebrew on Genesis and Exodus (G1. Genesis and Exodus). Each presupposes direct interaction with the Hebrew text, and influence from the ancient versions can safely be ruled out. In one instance, the item in question is a rare medieval usage. 51 While most are anonymous, three are identified by the scribe as τὸ ἰουδαϊκόν. 52 The following discussion is thus limited to the methodological issues surrounding instances of straightforward Hebrew–Greek translation. A much greater challenge is posed by those cases where a Greek version has apparently served a mediating role in Hebrew exegesis. Aitken (2009: 69) distinguishes three types: interpretation from Greek wording, interpretation based on Greek semantics, and Greek loan words in Hebrew. 49 50

THE SEPTUAGINT IN BYZANTINE JUDAISM 279

Distinct methodologies come into play depending upon the textual source, of which there are three basic types: (1) Manuscripts and inscriptions of medieval Jewish provenance; (2) Manuscripts and print material of early modern Jewish provenance; and (3) Manuscripts of non-Jewish provenance that exhibit contact with medieval Jewish sources. Each category presents its own challenges, and only by working eclectically can one hope to piece together the larger tradition. Four lines of inquiry may be distinguished: (1) Determining whether, in a given instance, one is dealing with a Hebrew–Greek biblical translation; (2) Assessing its textual character (here knowledge of the textual history of the Septuagint is crucial); (3) Characterizing its linguistic makeup (in which lexicography has a central role); and (4) Conjecturing the relevant pragmatics, its Sitz im Leben. It should be noted that manuscripts of Christian provenance represent a special case. Given that Christian scribes had access to the Greek versions of the Tannaitic period through the Hexapla, the burden of the argument falls on the one arguing for a medieval Jewish source (Boyd-Taylor 2015: 42–3). Two things must be demonstrated: (1) The independence of the translation from the Hexaplaric versions; and (2) Interaction with the Hebrew source (to establish that the text in question is in fact a translation). The study of translation technique, transmission history, and lexicography all enter into the analysis.53 This interplay of methods is nicely illustrated by the following examples. One Byzantine witness of undoubted Jewish provenance is a fragment of a Hebrew manuscript (tenth to eleventh centuries CE) that lists difficult Hebrew words from the books of Malachi and Job together with Greek translations written in Hebrew letters (G10. Malachi and Job). Here the Sitz im Leben is evident. It is also possible to locate some of the glosses in the history of the text.54 At Job 28:16, for instance, the Hebrew lemma reads ‫לא תסלה בכתם אופיר‬, “It cannot be valued in the gold of Ophir” (NRSV).55 The verb ‫“( תסלה‬to be paid”; HALOT, ad loc. ‫ סלה‬II) is glossed in G10, by ‫( אפושקולופיזיטי‬de Lange 1996: 83), that is, ἀποσκολοπίζεται (“to remove stumbling blocks”; LSJ, ad loc. ἀποσκολοπίζω). Ziegler (1982: 160) identifies Aquila as the ultimate source of the gloss. For although the Greek word is quite rare (and occurs in no other known Hebrew–Greek

Obviously the whole endeavor is dependent upon the state of Hexaplaric scholarship. In this regard, the fruits of the Hexapla Institute are eagerly awaited. For now, one must make do with the indices provided by Hatch and Redpath (1897–1906), and Reider (1966), used in conjunction with the secondary apparatus of the Göttingen editions (where available), the Larger Cambridge edition of the Septuagint, and Field (1875). 54 For a textual critical assessment of the G10 glosses on Job see Ziegler (1982: 160). 55 The received version of the Septuagint (which appears under the asterisk and was evidently derived from Theodotion) reads, καὶ οὐ συμβασταχθήσεται χρσίῳ Ωφιρ, “And it cannot rank with the gold of Ophir” (Ziegler 1982: 332). 53

280

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

translation), it is widely attested for this version.56 Yet the textual picture turns out to be complex. As Ziegler notes, the Hexaplaric witnesses attribute ἀναβληθήσεται (“to be lifted up”; LSJ, ad loc. ἀναβάλλω I.7) to Aquila in the present context. Consequently, if G 10 is also derived from an Aquilanic text, it represents an example of the so-called double tradition.57 On the other hand, it could well be a post-Aquilanic rendering of the passage, but one drawing on the lexical stock of Aquila’s version. The second reading of G10 for Job 28:16 is somewhat less problematic. Here the noun ‫“( כתם‬gold”; HALOT, ad loc. ‫ )ּכֶתֶ ם‬is matched by ‫( אישפילומן‬de Lange 1996: 83), that is the rare word σπίλωμα (“defilement, stain”; LSJ, ad loc. I). Again the motivation is likely to have been etymological, as the verbal form ‫ כתם‬carries the sense “to stain” in Middle Hebrew (HALOT, ad loc.). As it happens, Aquila’s rendering of this passage is not elsewhere attested, but the word σπίλωμα is attributed to him in another context, and so appears to have been part of his lexicon.58 Ziegler (1982: 160) thus concludes that the gloss stems from Aquila’s version. If he is correct, then G10 testifies to a continuous history of reception spanning nearly a millennium (de Lange 2008: 114). Another witness of secure provenance is a Hebrew manuscript dated to the tenth to twelfth centuries (G1. Genesis and Exodus), which contains philological and exegetical notes on the first two books of the Pentateuch grouped according to Jewish liturgical readings, including thirty Greek glosses (twenty-five of which are written in Hebrew characters). A number of these readings are also recorded by Fb, which increases the likelihood that they were not merely ad hoc renderings by the scholiast, but were drawn from a common tradition (Boyd-Taylor 2008b: 34–8). Nevertheless, their textual character is often difficult to establish. At Gen. 49:11, for instance, the gloss καλλίκαρπος (“rich in fine fruit”; LSJ, ad loc.) occurs both in G1, where it is transliterated into Hebrew as ‫( קַ לִ י ַכּ ְרפֹוס‬de Lange 1996: 101), and in Fb, as a match for the Hebrew lemma ‫“( שרקה‬noble grape”; HALOT, ad loc. ‫)שׂרקָ ה‬, ֵ a hapax legomenon which is used in parallel with ‫“( גפן‬climbing plant”; HALOT, ad loc. ‫ ) ֶּג ֶפן‬in reference to the grape vine as a figure of prosperity and abundance.59 The Septuagint renders the Hebrew word by ἕλιξ a poetic form commonly denoting things with a spiral shape (“a tendril of the vine”; LSJ, ad loc. III.1). It is conceivable that καλλίκαρπος comes from one of the other ancient Greek versions, as the word is attested as early as the fourth century BCE. There are no readings extant for the Three, so it is impossible to rule them out. Nevertheless, given that the word is both late and rare, it likely represents an early medieval translation.

56 The verb ἀποσκολοπίζω is also attested for Aquila at Ps. 67:5 (MT 68:5), where it renders the Hebrew verb ‫סלל‬ (Qal “to pile up”; HALOT, ad loc.); at Ps. 118:118 (MT 119:118) for ‫( סלה‬Qal “to treat as worthless”; HALOT, ad loc.); at Isa. 11:16; 36:2; and 40:3 for the noun ‫“( מסלה‬a track firmed with stones or fill”; HALOT, ad loc.); and at Isa. 57:14 for the verb ‫( סלל‬Qal). See Reider (1966: 29); for Greek Psalter references, see Field (1875: 2.199, 276); and for Greek Isaiah, see Ziegler (1939: 168, 249, 266, 334). 57 There is reason to believe that there was more than one edition of Aquila. See Field (1875: I.xxiv–xxvii). 58 According to Jerome Aquila used the word σπίλωμα at Isa. 13:12 as a match for Hebrew ‫( כתם‬Ziegler 1939: 171). The word is not otherwise attested in published texts of the ancient Greek versions. Aquila renders ‫ כתם‬by βάμμα (“that in which a thing is dipped”; LSJ, ad loc. I.) at Ps. 44:10 (MT 45:10) (Field 1875: II.163). At Song 5:11 he translates the phrase ‫ כתם פז‬by λιθέα τοῦ χρυσίου (“fine stone or marble”; LSJ, ad loc. λιθεία I.) (see Field 1875: II.419). See Reider (1966: 40). 59 The MT of Gen. 49:11 reads, ‫“( אסרי לגפן עירה ולשרקה בני אתנו כבס ביין לבשו ובדם ענבים סותה‬Binding his foal to a vine and his donkey’s foal to the tendril, he shall wash his robe in wine and his garment in the blood of a bunch of grapes” NRSV).

THE SEPTUAGINT IN BYZANTINE JUDAISM 281

Firmer conclusions may be drawn at Exod. 18:2, where the Hebrew lemma makes reference to the ‫“( שלוחים‬parting gift, dowry”; HALOT, ad loc. 2) of Zipporah.60 The Jewish scholiast glosses this obscure word by ‫( פריקיאון‬de Lange 1996: 106), that is, προικίον, a vernacular form of προίξ (“marriage portion, dowry”; LSJ, ad loc. 2; cf. προικίδιον). Interestingly enough, the gloss also occurs in Fb. The Septuagint, on the other hand, reads ἄφεσις (“divorce”; LSJ, ad loc. 4), an interpretative rendering perhaps intended as an explanation for Zipporah’s presence in Midian. Symmachus evidently followed the Septuagint. Aquila and Theodotion share an etymologically motivated match.61 The influence of the ancient versions can thus be ruled out with some degree of confidence. Hence the G1 gloss probably originated in a medieval Jewish translation. In this respect, it is significant that it construes the biblical text in reference to the return of Zipporah’s dowry, an interpretation attested by a number of medieval Jewish commentaries, but unknown to Christian exegetes (Boyd-Taylor 2008b: 38).

ONGOING RESEARCH QUESTIONS Research in this area is still in its early stages, so there are many desiderata. Four areas of investigation stand out for comment: i) Codicologists will want to revisit the provenance of early codex fragments of the Septuagint. Kraft (2003: 65–6) was right to question whether they should automatically be classified as Christian. In this regard there is a need for further work on scribal habits, so that Jewish manuscripts can be differentiated. ii) Agreements between rabbinic interpretation of the biblical text and the Septuagint have long been noted. Aitken (2009: 77) argues that the hypothesis that some rabbinic interpretations presuppose the text of the Septuagint is at least as strong as the hypothesis that they merely share a common exegetical tradition. A comprehensive study of the mediating role of the Greek versions in rabbinic literature would be welcome. iii) The material collected by Fb points to the interaction of Jewish and Christian biblical scholarship in Byzantium, a subject that merits further investigation. In this respect, it should be noted that in her analysis of Fb Fincati (2016) did not consult midrashic commentaries and medieval Romaniote authors. Yet, as she observes, these works frequently have recourse to demotic Greek vocabulary. Of particular interest are the Karaite texts. Much of this material remains to be edited and philologically assessed. iv) Continued philological study of the Greek version of the Constantinople Pentateuch will prove vital to the story of Jewish reception. Textual assessment of this version remains in its infancy. For such purposes a critical edition is a must (Krivoruchko 2008: 259).

The MT of Exod. 18:2 reads, ‫“( ויקח יתרו חתן משה את צפרה אשת משה אחר שלוחיה‬After Moses had sent away his wife Zipporah, his father-in-law Jethro took her back” NRSV). 61 The noun ἐξαποστολή (“sending away”; LSJ, ad loc. I.) is attributed to both Aquila and Theodotion (Wevers 1991: 223). It signals a perceived relationship between ‫ שלוחים‬and the Piel form of the verb ‫( שלח‬Piel “to let go, dismiss”; HALOT, ad loc. 3.b.), which is regularly translated by ἐξαποστέλλω (“to send away, dismiss”; LSJ, ad loc. II.) in the Septuagint. The word ἐξαποστολή is also attested for Aquila at Isa. 7:25 (Ziegler 1939: 149) and Theodotion at Isa. 27:8 (Ziegler 1939: 214). See also Reider (1966: 85). 60

282

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

RECOMMENDED RESOURCES 1. Aitken and Carleton Paget (2014). The articles in this volume trace the Jewish encounter with Greek culture from its earliest stages to the end of the Byzantine Empire. An essential handbook for the study of Greek-speaking Judaism. 2. de Lange (1996). The key fragmentary texts are edited and translated by Nicholas de Lange accompanied by critical notes and plates. 3. de Lange (2015). This monograph by de Lange will serve as the reference point for further research. 4. de Lange, Krivoruchko, and Boyd-Taylor (2009). The Cambridge GBBJ project hosted a colloquium in 2007 that resulted in a volume of critical studies by specialists. It covers a wide range of topics and presents original research. 5. Fincati (2016). A comprehensive textual and philological analysis of Fb.

CHAPTER 19

The Septuagint in the Eastern Orthodox Tradition MIKHAIL G. SELEZNEV

INTRODUCTION The family of Eastern Orthodox churches is formed by those Christian churches that trace their origin to the Byzantine commonwealth.1 The oldest non-Greek Church in this family is the Georgian Church, founded in the fourth century. Since the ninth century, and especially after the fall of Byzantium in 1453, an ever-increasing role in the Orthodox world has been played by the Slavic states and churches, the Russian Church becoming the strongest and most numerous. The list of Eastern Orthodox churches that mutually recognize each other and therefore were officially invited to the Pan-Orthodox Council of 2016 included churches of Constantinople, Alexandria, Antioch, Jerusalem, Russia, Greece, Cyprus, Romania, Serbia, Bulgaria, Georgia, Poland, Albania, and Czech Republic and Slovakia. The canonical status of some other Eastern Orthodox churches is disputed. The churches that split off from the Byzantine commonwealth during the great Christological controversies of the fifth century CE (Coptic Church, Ethiopian Church, Armenian Church, Syriac Orthodox Church, Assyrian Church of the East, and their daughter churches) do not belong to this family, and are usually referred to as Oriental Orthodox churches. The emergence of Orthodox diasporas (Greek, Russian, Romanian) in Western Europe and North America brought inner-Orthodox theological developments and debates to the West.

OVERVIEW The Old Testament Canon of the Eastern Orthodox Churches The ambiguity that characterized the attitude of the Byzantine Fathers toward the problem of canon continues in the post-Byzantine Orthodox tradition.2 All Orthodox churches agree that the Old Testament of the Christian Bible contains some books that are not found in the Hebrew Bible. The standard, semi-official designation

Space considerations limit the number of translations surveyed. For a broader perspective on Bible in the Orthodox world see, e.g., Scouteris and Belezos 2015. 2 On the topic of canon in Orthodox Churches see Meurer and Ellingworth 1992; Scanlin 1996; Desnitsky 2006; Konstantinou 2012; Crisp 2016; de Regt 2016. 1

284

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

for these books in Greek theological literature is ἀναγι(γ)νωσκόμενα (“those read”). The terminology goes back to the thirty-ninth festal letter of St. Athanasius (367), where he lists first βιβλία κανονιζόμενα (“books canonized,” i.e., the books of the Hebrew canon), then “βιβλία … οὐ κανονιζόμενα μέν, τετυπωμένα δὲ παρὰ τῶν Πατέρων ἀναγινώσκεσθαι τοῖς ἄρτι προσερχομένοις καὶ βουλομένοις κατηχεῖσθαι τὸν τῆς εὐσεβείας λόγον” (“books not canonized, but assigned by the fathers to be read to those now coming in and wishing to be catechized in the word of piety”). Athanasius’s statement is ambiguous, since these books are, on the one hand, οὐ κανονιζόμενα, but on the other hand, ἀναγινωσκόμενα. The current Greek usage takes the second part of Athanasius’s definition: ἀναγι(γ)νωσκόμενα. The terminology may sound illogical since the canonical books of the Bible are also read in the Church. In contrast, the terminology adopted by the Russian Orthodox Church since the middle of the nineteenth century is based on the first part of Athanasius’s definition: неканонические книги (“non-canonical books” = οὐ κανονιζόμενα). However, some Russian theologians question the logical consistency of the notion of “non-canonical books of the Bible canon.” The term “deuterocanonical books,” coined by the Post-Tridentine Catholic theologians, is often used in Orthodox theological literature, in Greece, Russia, and elsewhere, though sometimes condemned as non-Orthodox and “borrowed from Catholics.” The Orthodox tradition does not use the term “Apocrypha” with reference to these books. In Greek and Russian scholarly literature, the expression “Old Testament Apocrypha” refers to what in the English-speaking scholarship is usually called “Old Testament Pseudepigrapha.” There is no agreement on the exact list of these books. All the Bibles published by Orthodox churches for Orthodox audiences include Judith, Tobit, Greek Esdras,3 three Maccabean books, Wisdom of Solomon, Wisdom of Sirach, Baruch, and Epistle of Jeremiah. The book of Psalms contains Psalm 151, and Daniel and Esther contain additions not found in the Hebrew Bible. In addition to these, printed Bibles used by the Greek Orthodox Church include 4 Maccabees, though usually as an appendix (παράρτημα). This book was never included in any edition of Church Slavonic or Russian Bible. On the other hand, printed editions of Church Slavonic Bible as well as the Orthodox editions of the Russian Synodal Bible include “3 Esdras” (= Apocalypse of Ezra = 4 Esdras of the Vulgate), which is absent from the Greek manuscript tradition and was translated into Church Slavonic and Russian from the Vulgate. The so-called Prayer of Manasseh is printed in Church Slavonic Bibles and in the Orthodox editions of the Russian Synodal Bible as an appendix to 2 Chronicles; in the Greek tradition it is included in liturgical books and compilations (e.g., in the Odes), but not in the printed Bibles. Usually in printed Orthodox Bibles the “deuterocanonical” books of the Old Testament are interspersed among the “canonical” (just as it used to be in the manuscript tradition). Within the history of the Greek Bible an exception is the Athens edition of 1843–50, which puts these books in a separate volume. In the 1990s, the Russian Bible Society published an edition of the Russian Synodal Bible with the “non-canonical” books printed as a separate corpus between the “canonical” books of the Old Testament and the New LXX Ἔσδρας α′ (Esdras α) is known as 1 Esdras in most English Bibles with apocrypha and in German scholarship as 3 Esra. This book is present in other traditions as well, known as 3 Esdras in the Vulgate and 2 Esdras in the Slavonic Bible.

3

THE SEPTUAGINT IN THE EASTERN ORTHODOX TRADITION 285

Testament. Despite the blessing of the Patriarch of Moscow, this edition was not popular among Orthodox and the Bible Society returned to the traditional order of Old Testament books. There is no clear-cut Church verdict on whether “deuterocanonical” or “noncanonical” books should be treated as being on the same level as “canonical” ones. In the Early Church the boundary between “The Bible” and other books was not as sharp as in the Post-Reformation Western churches. The vehement discussion of the scope of the Biblical Canon started in the West in the context of polemics between Catholics and Protestants. The Orthodox churches soon became involved in the discussion, some theologians siding with Protestants, the majority with Catholics. In 1629, Cyril Lukaris, the Patriarch of Constantinople, published a confession of faith that was strongly Calvinistic in emphasis. Cyril states that only the twenty-two books approved by the Council of Laodicea should be called divine Scripture, while the “Apocrypha” (Cyril is almost unique within the Orthodox tradition in using this terminology) do not have such authority. Soon after Cyril’s death (he was executed by the Turkish government in 1638) the Synod of Constantinople of 1638 anathematized both Cyril and his confession of faith. This condemnation was confirmed by the synods of Constantinople (1642), Jassy (1642), and Jerusalem (1672). Thus, Cyril’s Confession of Faith was declared not Orthodox but Calvinist. The Jerusalem synod of 1672 explicitly declared the Wisdom of Solomon, Judith, Tobit, Bel and the Dragon, Susanna, the Maccabean books, and the Wisdom of Sirach to be genuine parts of Scripture. However, these local Synods do not have Pan-Orthodox authority and are often regarded by Orthodox theologians as reflecting strong Catholic influence. In general, one can say that the Greek Orthodox theology is probably more inclined to neglect the difference between “canonical” books of the Old Testament and the ἀναγι(γ) νωσκόμενα, while the Russian Orthodox theology of the nineteenth and twentieth centuries has given more attention to the difference (which is reflected in the abovementioned expression “non-canonical books”).4 The Pan-Orthodox Conference, held in Rhodes in 1961 to prepare a Pan-Orthodox Synod, included the issue of canon in the agenda of the forthcoming Pan-Orthodox Synod. However, later preparatory consultations removed this issue from the Synod’s agenda.

The Septuagint Text of the Greek Orthodox Church Until the establishment of the independent Greek state in the beginning of the nineteenth century, the possibility for printing Greek books in Greece or Constantinople was almost non-existent. Greek books, including the Bible and liturgical books, were published in the West, for example in Venice and other Italian cities, and later also in Jassy (Moldova). In 1821, several months after the start of the Greek War of Independence, the Russian Bible Society with the support of the Greek diaspora published the so-called Moscow Greek Bible. The Old Testament text was based on the edition of Breitinger (Zürich, 1730–2), which in turn was based on Grabe’s edition of Codex Alexandrinus (Oxford, 1707–20). Alexandrinus was chosen over Vaticanus because it was deemed to be closer to the late Byzantine manuscripts and the Church Slavonic tradition (an additional factor may have been the negative—for the publishers—associations between See also Konstantinou (2012: 44–5).

4

286

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the name “Codex Vaticanus” and Vatican as the center of the Roman Catholic Church). The respect for this edition in Greece was so great that when, after independence, a new edition of the Septuagint was published in Athens (in four volumes, without the New Testament, 1843–50), it was based on the Moscow edition and entitled ῾Η Παλαιὰ Διαθήκη κατὰ τοὺς῾Εβδομήκοντα. Ἐκ τοῦ ἐν Μόσχα, ἀδεία τῆς ἱερᾶς διοικούσης Συνόδου πασῶν τῶν Ῥωσσιῶν, ἐκτυπωθέντος ἀρχαίου ἀλεξανδρινοῦ Κώδηκος Μετατυπωθεῖσα (“The Old Testament according to the LXX. Reprinted from the ancient Codex Alexandrinus printed in Moscow by the permission of the Holy Synod ruling all of Russia”). By the beginning of the twentieth century, copies of the Athens edition, not to mention the Moscow edition, had become extremely rare. In 1928 the Greek missionary brotherhood Zoe (Αδελφότης Θεολόγων «Η Ζωή») published a new edition of the Greek Old Testament. According to the preface, it was based on what were considered the best editions of the Old Testament according to the LXX, following in all doubtful cases the critical edition of Tischendorf as most accurate. The second edition of the Zoe Bible, published in 1939, was prepared by P. I. Bratsiotis (Παναγιώτης Μπρατσιώτης; 1889– 1982), probably the most well-known Greek Bible scholar of that time. In the preface to this edition, Bratsiotis stated that for the Old Testament texts read at Church services (including the full text of the Psalter and Jonah), he “relied on liturgical books in Athenian libraries, namely manuscripts and books printed in Venice, whose text almost exactly coincides with the manuscripts,” while the remaining texts (i.e., most of the OT) were reprinted from the 1935 edition of Rahlfs. For the book of Judges, Bratsiotis used the B text of Rahlfs, for Tobit the BA text of Rahlfs, and for Daniel he followed the “Theodotion” version of Rahlfs and the liturgical books. There are some minor orthographic changes to the Rahlfs’s text, bringing it closer to the Attic norm, for example, elimination of movable -ν before consonants. Proper names that were in the Rahlfs edition left unaccented, Bratsiotis supplied with accents taken from earlier editions. It is not clear which particular liturgical texts Bratsiotis used or to what extent. A trial look at the texts of Judges, used in Orthodox liturgy (Judg. 6:2, 6, 11-24, 36-39; 13:2-8, 13-14, 17-18, 21) shows that, despite his statement in the Preface, in these places Bratsiotis used not the liturgical text of the Greek Church (the Menaion), but rather the B text of Rahlfs. In comparison with Rahlfs’s edition, the Bratsiotis edition (like that of Athens of 1843–50) lacks Odes and Psalms of Solomon. The order of books slightly differs from that of Rahlfs, and Fourth Maccabees is printed separately at the end of the Old Testament as an Appendix (παράρτημα). The Bratsiotis edition has become the standard Bible of Modern Greek-speaking Orthodox churches. It was reproduced many times, without any changes, by the brotherhood Zoe until the 1990s. Starting from 1997, essentially the same text (but with a new typesetting) began to be published by Αποστολική Διακονία, the official publishing house of the Church of Greece. The Athens Septuagint of 1843–50 as well as the first edition of the brotherhood Zoe Septuagint had the blessing of the Holy Synod of the Church of Greece. It is commonly understood that this blessing concerns also the second edition of the Zoe Septuagint and its reprints. In the Eastern Orthodox liturgy, the Old Testament texts are read during Vespers and Hours. In the Byzantine period, the collection of these readings (the so-called Prophetologion) functioned as a separate liturgical book.5 The manuscripts of the Prophetologion are often referred to as Old Testament lectionaries. In the post-Byzantine On the Prophetologion see Miller (2010).

5

THE SEPTUAGINT IN THE EASTERN ORTHODOX TRADITION 287

period, copying of the Prophetologion ceased (only one printed edition is known) and the Old Testament readings were scattered through several liturgical books containing mostly non-Biblical texts and hymns, namely the Menaion, the Triodion, and the Pentecostarion. The relationship between the OT liturgical readings of the post-Byzantine Greek Church and the earlier Byzantine lectionaries is yet to be studied. According to Mihăilă (2018: 47–51), the liturgical readings from Judges 6:2, 6, 11-24 represent an eclectic text, closer to Codex Alexandrinus than to Vaticanus.

Translations of the Old Testament into Modern Greek Vernacular Greek paraphrases of the Bible (especially of the Psalter) started already in the sixteenth century. Christian paraphrases of the OT books were based on the LXX. But a new stage in the history of Bible translation into Modern Greek began with the initiatives of the British and Foreign Bible Society (BFBS) in the beginning of the nineteenth century.6 The BFBS established contact with Adamantios Korais, the leading figure of the Greek Enlightenment, who played a decisive role in laying the foundations of Modern Greek literature and the “purified” Greek language (Katharevousa). In his letter to the BFBS, dated 1808, Korais speaks against translation of the OT from the LXX: The actual state of knowledge (which is also beginning to penetrate into Greece) would not justify the measure of adopting the version of the Septuagint as a standard text … There is only one objection to be made to a version from the Hebrew itself, and that is, the veneration which our nation entertains for that of the Septuagint; but, besides that this prejudice is growing weaker and weaker in proportion as we are becoming enlightened, it would be easy to prevent these alarms in feeble minds, by inserting at the bottom of the page, in the form of variations, all the passages in which the version of the Septuagint departs from the original and by consequence from the new Greek version. (Clogg 1969: 252–3) Around 1820, the BFBS commissioned a new Bible translation from Hilarion, archimandrite of Sinai and later metropolitan of Tirnovo, but eventually decided to reject Hilarion’s translations because they did not comply with the BFBS guidelines, for example, Hilarion followed the LXX. This decision proved to be fatal. Contrary to Korais’s forecast, the rejection of the LXX by the BFBS made almost all Greek Orthodox clergy opponents of the Bible Society. The new BFBS project of Bible translation into Modern Greek was accomplished by Neophytos Vamvas (Νεόφυτος Βάμβας; 1770–1856), an Orthodox archimandrite, professor of philosophy (later dean) of the Athenian University, and one of the very few Greek Orthodox clerics who shared the translation principles of the BFBS, including the superiority of the Masoretic Text. Between 1831 and 1851, Vamvas translated the entire Bible, namely the books of the Hebrew canon (translated directly from Hebrew) and the NT. Vamvas’s translation was immediately condemned by Church officials. The Holy Synod of the Church of Greece disapproved of it between 1834 and 1836 in several letters and declarations. As a reaction to Vamvas’s translation, the Ecumenical Patriarchate in Constantinople issued an encyclical condemning any translation of the

On the history of the Modern Greek Bible translations see Vaporis (1994); Delicostopoulos (1998); Clogg (2004); Livanios (2014).

6

288

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Bible into vernacular Greek (1836). Constantinos Oikonomos (1780–1857), an eminent Greek Orthodox scholar and theologian, wrote a four-volume treatise in defense of the LXX (see Dafni 2010). Nevertheless, since 1851, Vamvas’s translation has been republished many times by the Bible Societies, first by BFBS, later by the Hellenic Bible Society (Ελληνική Βιβλική Εταιρία), established in 1992. It is used almost exclusively by the Greek Protestant community. In 1997 the Hellenic Bible Society published a translation of the Bible into the modern vernacular (Dimotiki), commonly called Today’s Greek Version. The textual basis for the books of the Hebrew canon was the MT, while the deuterocanonical books were translated from the Septuagint. The Hellenic Bible Society publishes Today’s Greek Version in two formats: with the deuterocanonical books (for Orthodox, with a recommendation letter from the Greek Church authorities as a preface), and without them (for Protestants). A translation of the LXX into Modern Greek was announced by the Hellenic Bible Society in the 1990s, with publication expected by 2005, but later delayed. Two other projects of translating the Septuagint into Modern Greek were planned in the 2000s, one by the publishing house Psychogios (Ἐκδόσεις Ψυχογιός), and the other by the theological association Artos Zois (Ἄρτος Ζωής), but as of 2019 no publications have yet appeared. An interesting, though almost forgotten page in the history of the translation of the Greek Bible is the publication of the Septuagint text and a Modern Greek translation of the Hebrew Bible in parallel columns, prepared by Athanasois Chastoupis (Ἀθανάσιος Χαστούπης), a specialist in classical and oriental languages (Athens, 1954–5). The juxtaposition of the two versions permitted a diachronic perspective on the Bible text. For example, in Ps. 137 (138):1 the Septuagint column read ἐναντίον ἀγγέλων ψαλῶ σοι (“before the angels I will sing psalms to you”), while Chastoupis’s translation of the Hebrew gave a much more archaic reading “παρουσίᾳ τῶν (ἀλλοτρίων) θεῶν ψάλλω εἰς σὲ ὕμνους” (“in the presence of (alien) gods I sing hymns to you” = ‫)נגד אלהים אזמרך‬. Chastoupis’s publication may be called a unique attempt at a “modern Hexapla.” However, without a serious commentary, this edition, with two adjacent columns explicitly contradicting each other, was doomed to create more confusion than understanding. The print run was limited and it was never reprinted.

The Septuagint and the Church Slavonic Bible The first Slavonic Bible translations were made from the Greek by saints Cyril and Methodius and their disciples in Moravia in the 860s–880s. In the following centuries, Slavonic Bible texts were often revised against the Greek manuscripts or translated anew from the Greek.7 It is often taken for granted that the Church Slavonic tradition, being later than the Syriac, Coptic, or Armenian, cannot provide any additional information on the early history of Bible texts. However, the situation with the apocryphal literature (e.g., 2 Enoch exists in Slavonic translation only) suggests that such a judgement may be premature.

General introductions to the Church Slavonic Bible tradition are Alekseev (1999), Thomson (1999), Bruni (2016).

7

THE SEPTUAGINT IN THE EASTERN ORTHODOX TRADITION 289

Serious interest in the Church Slavonic tradition of the Old Testament was expressed in the beginning of the twentieth century at the Septuaginta-Unternehmen. In 1910, Rahlfs asked Evseev, the most prominent Russian scholar of the Church Slavonic Bible of that time, to participate in preparation of the critical text of the LXX. As a first step, the Septuaginta-Unternehmen tasked Evseev with compiling the full catalogue of Slavonic Old Testament manuscripts, in a fashion similar to Rahlfs’s catalogue of LXX manuscripts. This work was done and even paid for by the Unternehmen, but printing of the catalogue in Berlin was prevented by technical and, later, political problems (the First World War). After the Russian Academy of Sciences agreed to publish the catalogue in St. Petersburg, the manuscript of the catalogue was returned to Evseev. But the Russian Revolution and the ensuing events put an end to the project. Today we have only the draft version of the catalogue in the private archive of Evseev, after his death in 1921 (Alekseev 1999: 130). Taking as his starting point the conception of de Lagarde, Evseev wanted to trace what he thought were the “Lucianic” and “Hesychian” text traditions in the Russian manuscripts. According to Evseev, Slavonic liturgical readings from the Old Testament as well as the Slavonic Psalter preserve texts going back to saints Cyril and Methodius. Since they were officially commissioned by the Church of Constantinople to translate the Bible into the language of Slavs, their work may be regarded as primary witnesses to the Bible text of the Church of Constantinople in the ninth century. Evseev called this text the Eastern Vulgate and, following the famous notice of Jerome, identified it with the “Lucianic” recension of the Septuagint. Slavonic Bible translations in the catenae manuscripts were, according to Evseev, made later, at the time of the Bulgarian king Simeon. Evseev (1911: 445–50) deemed them to belong mainly to the “Hesychian” tradition, and suggested that the Bulgarian kings turned away from the “Lucianic” Eastern Vulgate to the “Hesychian” text because of their desire to be free from the cultural influence of Byzantium. Looking back, we see that these reconstructions are significantly out of step with modern scholarship with regard to both general methodology of textual studies and the history of the Septuagint text. The modern approach to the problem of the “Lucianic” tradition—as concerns its history, main features, and scope—is completely different from the picture drawn by de Lagarde, which was the basis of Evseev’s hypotheses. A test comparison of Slavonic manuscripts of Kingdoms with the Greek text of Kingdoms in manuscripts boc2e2, which are main witnesses to the Antiochean redaction, has demonstrated that neither liturgical nor continuous texts of the Slavonic tradition were oriented toward the tradition represented in boc2e2 (Alexeev 1999: 119–23). Moreover, the hypothesis of political factors leading to a change from “Lucianic” to “Hesychian” text is an obvious anachronism. As concerns the “Hesychian” recension, its very existence has been put in doubt by modern studies.8 Evseev’s reconstructions being thus rejected, the problem of the Greek sources of the early Slavonic manuscripts still awaits its explorer. A peculiar feature of the medieval Slavonic Bible manuscript tradition is the ample evidence of direct contacts between the Orthodox Christians and the Jews, which left traces on Christian Bible manuscripts. Over the last few decades, this has become an important topic in the study of the Church Slavonic Bible.9 The East-Slavonic translation of Esther (earliest manuscripts date to the fourteenth century) follows the MT and does not contain the Septuagint additions. Most East-Slavonic Pentateuch manuscripts also See the short exposition of the history of research in Fernández Marcos (2000b: 241–2). See Alexeev (2014), Taube (2012), Grischenko (2018) with literature.

8 9

290

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

exhibit the influence of the MT, mainly in marginal glosses and in subdivisions of the text corresponding to the weekly Torah readings. The very notion of a “Pentateuch” is alien to the Byzantine manuscript tradition (where the first books of the Bible were usually transmitted as an “Octateuch”: Genesis, Exodus, Leviticus, Numbers, Deuteronomy, Joshua, Judges, Ruth) and is sometimes regarded as a sign of a Jewish influence. The influence of the Vulgate on the text of the Church Slavonic Bible is especially clear in the first full Church Slavonic Bible, the so-called Gennady Bible, a manuscript written in 1499 for Gennady, archbishop of Novgorod. In matters of canon, the Gennady Bible largely follows the Vulgate. For example, it includes the Apocalypse of Ezra (= Vulgate IIII Ezrae), which is absent from the Greek manuscript tradition and was translated from Latin.10 Several books that were unavailable to bishop Gennady and his scribes in Slavonic (e.g., Chronicles, MT Ezra and Nehemiah, Tobith, Judith, Wisdom, Maccabees, nonMT parts of Esther, non-LXX parts of Jeremiah) were translated for the Gennady Bible from Latin. The chapter arrangement in Jeremiah, and to a large extent, the text itself of Jeremiah, also followed the Vulgate. The Gennady Bible served as the basis for the printed editions of the Church Slavonic Bible, namely the Ostroh Bible of 1581, the Moscow Bible of 1663, and Elizavetinskaya Bible (“Queen Elizabeth’s Bible”) of 1751–6, which became the official Bible text of the Russian Orthodox Church. As concerns the Latin influence of the printed Church Slavonic Bibles, the situation is complicated. On the one hand, in preparation of the printed editions, most of the texts that had been translated in the Gennady Bible from Latin were re-translated from Greek. On the other hand, the editors of the printed Church Slavonic Bibles often used the Vulgate to correct the translations made earlier from the Greek. For example, it is well known that “Molech” is absent from the Greek Pentateuch, being replaced with ἄρχων (“ruler”). However, in the printed editions of the Church Slavonic Bible, “Molech” reappears in the Pentateuch, having been borrowed from the Vulgate, first as a marginal gloss (Ostroh Bible), then in the main text (Elizavetinskaya Bible). Similarly, in Hos. 11:1 the Gennady Bible and all the printed editions follow the MT/Vulgate reading “out of Egypt I called my son,” instead of the reading “out of Egypt I called his children,” the latter unanimously witnessed to by the Septuagint tradition. In this case the MT/Vulgate reading was preferred over the Septuagint for dogmatic reasons. As noted, the canon of printed editions of the Slavonic Bible goes back to the medieval Vulgate, through the Gennady Bible. However, it differs from the standard Catholic edition, Vulgata Clementina, since the OT books relegated by Vulgata Clementina to the separate Appendix at the end of the Bible (the non-canonical books of Ezra, Psalm 151, and Prayer of Manasseh) are kept by the Slavonic Bible in the body of the OT corpus (the Prayer of Manasseh is appended to 2 Chronicles). The canon of printed Slavonic Bibles also differs from the Vulgata Clementina by including 3 Maccabees and differs from the printed Greek Bibles by not including 4 Maccabees. Additionally, the chapter order in the book of Jeremiah in the Elizavetinskaya Bible follows the Vulgate, and verses present in the MT and Vulgate but absent from LXX Jeremiah are kept in translation from Latin. In ideological debates of nineteenth through twenty-first-century Russia, the Church

Starting with the Gennady Bible, the names of the books linked in the in the Slavonic and Russian Bibles with Ezra/Esdras are the following: “I Ezdra” (canonical Ezra, Vulgate I Ezrae), “Nehemiah” (canonical Nehemiah, Vulgate II Ezrae), “II Ezdra” (LXX Ἔσδρας α′, Vulgate III Ezrae), “III Ezdra” (Apocalypse of Ezra, Vulgate IIII Ezrae).

10

THE SEPTUAGINT IN THE EASTERN ORTHODOX TRADITION 291

Slavonic Bible has often been represented as the true daughter of the true Septuagint. But clearly this is far from reality, especially as concerns the printed editions.

Translations of the Old Testament into Modern Russian By the beginning of the nineteenth century, the Russian literary language established itself as a linguistic entity different from Old Slavonic, with a flourishing body of literature. In 1815, the Russian Bible Society (1814–26) launched a project of Bible translation into Russian. But because of opposition from the more conservative part of the Orthodox clergy, the Bible Society was closed in 1826 and the whole print run of the new Russian Pentateuch was burnt. After several decades of debates, the translation project was resumed in 1858 and finished in 1876. Published under the aegis of the Holy Synod, the Russian Bible is commonly called the Synodal Bible. It was (and still is) authorized for private reading only, not for liturgical use.11 Following the example of the British and Foreign Bible Society, the Russian Bible Society used the MT as the base text for the Synodal Old Testament, a decision that received much criticism and created conflict within the Church. Among the main champions of the MT was St. Filaret Drozdov, the metropolitan of Moscow (1782– 1867), who wrote a memorandum entitled On the Dogmatic Value and Conservative Usage of the Greek Septuagint and Slavonic Translations of the Holy Scripture (Filaret 1858). Contrary to the title, the memorandum seeks to defend both the LXX tradition (reflected in the Church Slavonic version as well) and the MT. First, St. Filaret mounts a series of arguments in favor of LXX readings. For example, he argues that the LXX is a “mirror of the Hebrew text as it was two hundred years or more before Christ,” and that “in the Orthodox teaching of Holy Scripture it is necessary to attribute a dogmatic merit to the Translation of the Seventy, in some cases placing it on an equal level with the original and even elevating it above the Hebrew text.” However, next he makes a series of arguments in favor of the MT readings (the quotations from Hos. 11:1 in Mt. 2:15 and Isa. 42:1 in Mt. 12:18, references to Church Fathers’ usage of the Hebrew text, etc.). The memorandum of St. Filaret is one of the most important statements of the Russian Orthodox Church on textual problems of the Bible, often quoted and referred to right up to the present time. The “deuterocanonical” books of the Synodal Bible were translated from Greek and included 3 Esdras translated from Latin. But the editions that cater to the Protestant audience are limited in their OT to the MT canon. The influence of the LXX/Church Slavonic tradition is often felt in the canonical books of the Synodal Bible as well. As a compromise between the proponents of the MT and the Septuagint, the words, clauses, and passages that exist in the Septuagint but are absent from the MT were translated from the Septuagint and inserted (in brackets) inside the translation otherwise made from the MT. One may say these brackets played the same role as the obelos in the Hexapla. As far as I know, the Synodal Bible is the only widespread Bible translation, after Origen, that tries to combine several base texts and, at the same time, to distinguish them with text-critical markers. Unfortunately, this practice was carried out very inconsistently. In the 1990s, the author of the present chapter was involved, as the editor-in-chief of the Russian Bible

See Batalden (2013, 2017) with earlier literature.

11

292

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Table 19.1 Septuagintal Additions to the Synodal Version Book(s)

Additions

Genesis

418

Leviticus – Deuteronomy

941

Psalter

153

Isaiah

9

Jeremiah

4

Ezekiel

4

Society, in checking the textual accuracy of the Synodal version. Altogether, according to my calculations, in the Synodal translation there are 2,405 additions from the Septuagint marked with brackets (and an unknown number of additions and changes not marked). The distribution of these additions is very uneven, as shown in Table 19.1. But these statistics do not correlate in any way with the number of actual discrepancies between the LXX and the MT. For example, they give the misleading impression that the discrepancies between the MT and the LXX in Genesis are one hundred times more numerous than in Isaiah or Jeremiah! Instead, these statistics reflect the individual preferences of editors of different books. Worse still, the same siglum (brackets) was used as a punctuation sign. Quite often it is impossible to tell the intended meaning of the brackets in a given place without consulting the Greek and Hebrew. A Protestant version of the Synodal translation that appeared at the end of the nineteenth century omitted almost all the words in brackets, treating them as the LXX additions, alien to the “Hebraica Veritas.” As a result, several “innocent” passages were removed that had actually been translated in the Synodal version from the MT and put in brackets for purely stylistic reasons. The Synodal translation was heavily criticized for its eclectic nature, both by those in favor of the Septuagint and by those in favor of the MT. Several alternative translations have been offered. Already in 1869, bishop Porfiry (Uspensky), scholar, traveler, and collector of ancient manuscripts, had published samples of Russian translation of several important OT texts from the Septuagint. A full translation of the Psalter by Porfiry (from the LXX) was published posthumously in 1893. In 1909–17 P. Yungerov, professor of Kazan Spiritual Academy, published his translations of Proverbs, Isaiah, Jeremiah, Lamentations, Ezekiel, Daniel, the Minor Prophets, the Psalter, Ecclesiastes, Song of Songs, and the beginning of Genesis. This project was conceived as a full-scale traditionalist alternative to the Synodal version. However, the Russian revolution of 1917 prevented Yungerov from completing a full translation of the Old Testament. The idea behind Yungerov’s project was to create a Russian Bible text as close as possible to the official Church Slavonic text (Elizavetinskaya Bible). The textual basis of most of his translations was Codex Alexandrinus. Where Alexandrinus differed from the Elizavetinskaya Bible, he checked the apparatuses of available Greek editions for readings that could support the Elizavetinskaya. If such readings were not attested in the Greek tradition, he sometimes translated directly from Church Slavonic (e.g., Hos. 11:1, the passage that in the Church Slavonic Bible had been translated from the Latin for dogmatic reasons). Usually he reflected in his apparatus the divergences between his main Greek

THE SEPTUAGINT IN THE EASTERN ORTHODOX TRADITION 293

editions and the Church Slavonic text (strangely, he forgot to do it in Hos. 11:1, where his translation had no support in the Greek tradition). All in all, Yungerov’s goal was not the translation of the LXX as such, but rather creation of a Russian version of the “Greek-Slavonic text” (the expression he often used in prefaces to his translations). Two translations of the LXX Psalter that appeared in the last decades to be used alongside the Church Slavonic Psalter (Birukova and Birukov; Timrot) follow Yungerov’s model, creating a Russian version of the “Greek-Slavonic text.” The years of Communism (1917–91) did not favor Bible studies in Russia. However, in the last decades of Communist power and especially in the 1990s, there appeared several translations of individual books of the Old Testament, made by philologists and specialists in ancient languages, mostly from the MT. In 1990, the famous Russian philologist S. Averintsev published six Psalms of the Orthodox Morning Service, translated from the Septuagint, and, at the same time, prepared a translation of almost all of the Psalter from Hebrew. The two translations differed not only in their base text, but also in their stylistic features and the register of the Russian language. This translational experiment may serve as precedent for the coexistence of two types of translation within the same cultural milieu. In the 1990s and 2000s, the necessity of having two different Bible translations, one from the Hebrew and one from the Greek, corresponding to two different stages in the development of the Bible tradition, was voiced by the author of the present chapter, at that time editor-inchief of the re-established Russian Bible Society (see Seleznev 2008). The translation of the Hebrew text of the Old Testament was supported by the United Bible Societies already in the 1990s (published in 2011 by the Russian Bible Society as part of the Contemporary Russian Version), and the project of translation of the Septuagint was discussed by the Russian Bible Society and UBS, but never realized.

Septuagint-Based Bible Translations in the Orthodox Diaspora Although the priority of the Septuagint over the supposedly “corrupt” MT is often asserted in preaching, in reality the Orthodox diasporas mostly use the same Bible translations as other Christian denominations: in English-speaking environments the NKJV or RSV, in French-speaking environments the TOB, etc. The liturgical books, of course, are translated into the target language from Greek or Church Slavonic; this is also mostly true with regard to the Psalter, one of the most important liturgical texts of the Orthodox tradition. There exist several English versions of the Orthodox Psalter, some in print, some on the internet.12 The most widely published Orthodox Bible edition in English is the Orthodox Study Bible (OSB). The first edition (1993) contained the NKJV text of only the Psalter and New Testament, with annotations written from the point of view of Orthodox theology. It was heavily criticized by Orthodox believers especially for using the NKJV text of the Psalter. Even the numbering of Psalms followed not the LXX, but the NKJV/KJV/MT model. The second edition (2008) contains the whole Bible with the deuterocanonical books. The NT is the NKJV, while the OT is based on the NKJV but is sometimes revised according to the LXX (Rahlfs’s edition), with some help from the English translation by Brenton (1844). The revisions concerned only the few points that were important from

For a (conservative) Orthodox evaluation of the existing Bible texts in English, see Whiteford (2015).

12

294

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the point of view of the editors, otherwise the OT text of the OSB follows the NKJV and diverges from the LXX even in some well-known passages. For example, in Gen. 4:8, according to the OSB, “Cain talked with Abel his brother” (without any further details, as in the MT), while according to the LXX, he told him “let us go to the plain” (διέλθωμεν εἰς τὸ πεδίον). Similarly, according to the OSB, the flood begins (as in the MT) “in the six-hundredth year of Noah’s life, in the second month, on the seventeenth day of the month,” while according to the LXX it was “on the twenty-seventh” day (ἑβδόμῃ καὶ εἰκάδι; Gen. 7:11). Recent decades have seen a marked growth of scholarly interest in the Septuagint, witnessed by—among other things—its translation into English (NETS), French (La Bible d’Alexandrie) and German (Septuaginta Deutsch).13 However, there was no contact between these projects and theological institutions of the Orthodox diasporas. It seems that modern Septuagint scholarship and these theological institutions belong to two different worlds, having little in common.

DEBATES While opening the international theological conference “Contemporary Biblical Studies and Church Tradition,” metropolitan Hilarion (2017: 31–3), the President of the Synodal Biblical and Theological Commission of the Russian Orthodox Church, stated that The Orthodox Church never canonized any one particular text or translation, or any one particular manuscript or any one edition of the Holy Scriptures. There is not a one and only generally accepted text of the Bible in Orthodox tradition … The Apostolic Church did not strive towards canonizing any one particular type of biblical text. The Orthodox Church does not do this either. Still, as the metropolitan stressed, the Septuagint is an important part of the Orthodox identity. In fact, many ideas one can encounter in modern Orthodox debates have their roots in the Patristic age. These include, for example, the idea that the Jews had deliberately corrupted the messianic passages of the OT, or the notion of inspiration of the Septuagint paradoxically accompanied by readiness to use the Hebrew or Hexaplaric material for homiletical and exegetical purposes.14 It is interesting that Orthodox anti-Latin polemics that followed the schism between the Greek-speaking Byzantine Church and the Latin-speaking Western Church did not include Catholic use of the MT-based Vulgate until the late middle ages. It appears that the first outright Orthodox attack on the “Jewish” textual basis of Vulgate was provoked by a Western attack on the Letter of Aristeas. In the extensive commentary on Augustine’s De civitate Dei, published in Basel in 1522, Luis Vives (Joannes Ludovicus de Vives) cast doubts on the authenticity of Aristeas. De Vives’s commentary was heavily criticized by Maksim (c. 1475–1556), a learned monk of Greek origin and unusual biography. As a youth, he studied Greek and Latin in Italy, where he was deeply impressed by the Dominican friar Girolamo Savanarola. Later he became a monk at Mt. Athos from where he was sent to Russia to translate spiritual literature. He was proclaimed a saint in 1988.

On these translations, see the chapter by Ross in the present volume. On patristic biblical theory see Gallagher (2012) and in the present volume.

13 14

THE SEPTUAGINT IN THE EASTERN ORTHODOX TRADITION 295

In a pamphlet entitled “Against Ioannes Ludovicus,” written in Church Slavonic, St. Maksim criticized Vives’s “aberrations,” among which was his negative attitude toward the Septuagint. This criticism turned into a full-scale denunciation of Jerome and Catholics, who—Maksim claimed—had neglected the Septuagint and turned to the Scriptures of the “deicidal Jews” (Maksim 1862: 3.203-26). After the appearance of the MT-based Protestant translations, the Septuagint has become a symbol of Orthodoxy in its struggle on three fronts: against Jews, against Catholics, and against Protestants. The polemics become especially bitter when the Orthodox churches are confronted with foreign missionary activity or when new Bible translations are launched. In many regards, archimandrite Vamvas’s work with the BFBS on the Modern Greek translation may be compared with the work of the Russian Bible Society, and its most outstanding member, archimandrite (later metropolitan) Filaret Drozdov (1782–1867), on the Russian Synodal Bible published in 1876. In both cases, new translations were initiated by the agents of the BFBS but carried out by local Orthodox clergymen. In both cases, the translations came under heavy attack from two angles: first, because of use of vernacular instead of traditional Church language, and second, because of the Hebrew text used as the basis for the OT translation. The different fate of the two translation projects—the ultimate success of the Russian Synodal Bible and the rejection of its Greek counterpart by the Greek Orthodox Church—is probably to be explained by the very special position the Septuagint occupies in Greek culture and in the Greek Church. As Delicostopoulos (1998: 297) has put it, “the Greek nation has the rare privilege of having as its mother tongue the language of the New Testament as well as of the Septuagint (LXX).” Within Greek culture, the Greek NT and the Greek Septuagint are not perceived as mere versions of the Bible, but rather as The Bible, connected in a unique way to the Greek language and Greek history. Any move to replace them with a modern translation risks being perceived as an attempt to deprive the national culture of this unique possession. The Synodal Bible created in Russia a completely new situation: a predominantly LXX-based text used in liturgy and a predominantly MT-based text authorized for private reading. Up to the middle of the twentieth century, all the official and semi-official documents of the Russian Church, as well as almost all theological literature quoted the Bible in Church Slavonic version only. Starting from the middle of the twentieth century all such quotations follow the Synodal Bible. As concerns the Greek Church, the Septuagint has always been treated as the authoritative text. Though the Orthodox version of Today’s Greek Version is published with the letter of approval of the Church authorities, this letter states that this translation can be used for studying the truth revealed by God, but that it cannot replace “in the liturgical and general use in our Holy Orthodox Church the translation of the Seventy.” This position is partly reminiscent of that of the Russian Orthodox Church: in both cases the liturgical text is LXX-based and archaic in language, while the modern language MTbased translation is authorized for private usage only. However, the sphere of usage assigned to the Today’s Greek Version is significantly narrower (at least for the time being). Other Orthodox Bible translations oscillate between the Russian Synodal model (MT as the main base text with some respect for the LXX) and the LXX-only model (see Mihăilă 2018).

296

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Differently from medieval Greek Fathers, modern Orthodox theologians must face the problem of numerous differences between the Septuagint and modern MT-based Bible translations. It is not possible to ascribe all of these differences to the supposed “corruption” of the MT. Sometimes conservative Orthodox theologians follow the option suggested by Augustine after he had become aware of the differences between the Septuagint and the Hebraica Veritas of the Vulgate: “Orthodox believe that the changes in the LXX were made under the inspiration of the Holy Spirit, and are to be accepted as part of God’s continuing revelation” (Ware 1963: 208). After the Qumran discoveries, Orthodox proponents of the “Septuagint-only” ideology announced that the Dead Sea Scrolls have shown the priority of the LXX text over the MT, an idea that has become quite popular in the Orthodox milieu. In Russia, sometimes one can hear appeals to reject the Synodal Bible because of its “Jewish” textual base (MT). An extreme case is represented by a project to translate the OT back into Hebrew from the Elizavetinskaya Bible, the Queen Elizabeth’s Church Slavonic Bible of 1756 (Shamir 2009). On the other hand, at the beginning of the twenty-first century, in Greek, Russian, and Romanian theology and Bible scholarship, one can see developments toward recognizing textual pluralism as an inherent feature of the Orthodox approach to the Bible. Miltiades Konstantinou (2012: 53), one of the main participants in the Hellenic Bible Society project translating the LXX into Modern Greek, states that “the Orthodox Church … must recognize as her own heritage both texts, the Hebrew and the Septuagint, encouraging their study and research” (see also Seleznev 2008). The Romanian Biblical scholar Mihăilă (2018: 33) formulates this in a rather aphoristic way: “The slogan for Orthodox biblical studies should be … ‘back to Hexapla,’ not ‘back to the Septuagint.’” The Hexapla analogy is not to be taken as an indication that the duality of the LXX and MT should be treated as a purely textual phenomenon, like the difference between two codices. There are important theological and ecclesiological aspects behind this duality. From the theological point of view, this duality is rooted in the double nature of the Old Testament of the Christian canon. On the one hand it is the text stemming from ancient Israel and Judah, from the world of the Ancient Near East; it is pre-Christian and pre-Hellenistic. On the other hand, it is a part of the Bible of the Christian Church from the beginning of the Common Era. There is a diachronic dimension within the Bible itself that comes into play. From the ecclesiological point of view, this duality is linked with the fact that, on the one hand, Orthodox churches are heirs to the Byzantine tradition, and on the other, a part of world Christianity. Rejecting the Septuagint would mean betraying the Orthodox identity, rejecting the MT and MT-based translations would mean going into a self-created “Orthodox ghetto.”

ONGOING RESEARCH QUESTIONS As we have seen, the Septuagint tradition after and outside Byzantium has drawn much less attention than the history of transmission and usage of the Septuagint texts in antiquity and the medieval Greek-speaking world. Among the areas that are definitely under-researched one may cite, for example, the study of Greek Septuagint lectionaries and their reflection in post-Byzantine liturgical texts; the problem of Greek sources of Church Slavonic Bible texts and their role for reconstructing the textual history of the

THE SEPTUAGINT IN THE EASTERN ORTHODOX TRADITION 297

Septuagint; and the history of Georgian OT translations, which may well go back to the Septuagint of the early Byzantine period. As concerns the debates about the place of the Septuagint tradition in the Orthodox world, they are not just objects of historical study, but they have direct bearing on the matter of Orthodox identity. Questions that modern Septuagint scholarship poses before Orthodox communities deal not only with textual matters, but also with theological presuppositions and with the nature of religious tradition.

RECOMMENDED RESOURCES 1. Pentiuc (2014). A monograph on the role of the Old Testament in Eastern Orthodox tradition, including issues of text, canon, interpretation, and liturgical usage. 2. Mihăilă (2018). A brief study of historical and theological aspects of the LXX–MT controversy in the Orthodox world. 3. Scouteris and Belezos (2015). A brief essay on the history of the Bible in different regions of the Orthodox world (Greek-speaking world, Slavonic countries, Romania, Georgia, Arabicspeaking world) from the seventeenth century onward. 4. Meurer, Siegfried, and Ellingworth (1992). Collection of papers devoted to the status of Apocrypha/Deterocanonical writings in the Christian tradition. 5. Magdalino and Nelson (2010). Collected essays on different aspects of the OT in Byzantium, including an important essay on the Prophetologium.

298

CHAPTER 20

The Septuagint in Early Modern Europe SCOT T MANDELBROTE

INTRODUCTION For our purposes, the early modern period might be defined as running from the revival of Greek studies in fifteenth-century Italy to the beginning of the career of Constantin Tischendorf (1815–74). Tischendorf made his name through the application of technological advances in manuscript studies and confirmed his reputation through the extraordinary acquisition of Codex Sinaiticus, the first pinnacle in a chain of discoveries in the Near East that eventually transformed both the methods and the materials of Septuagint (as well as New Testament) study (Porter 2015). Tischendorf himself was animated by research questions that mattered theologically because of the ongoing consequences of the Protestant Reformations of the sixteenth century. Confessional debate, especially during the seventeenth century, turned questions of critical judgment into ideological divisions, in particular between members of the Reformed Churches and adherents of the Catholic Church. At the heart of such debates lay the contested authority of the Bible as a rule of faith, expressed through discussion of the historicity of the text itself. Discoveries of new documents and the realization of the importance of the historical witness of texts in translation generated an outpouring of Septuagint scholarship from the 1570s to the 1720s. This took the form of editions of the text, discussions of its language and the preparation of tools for its understanding, critical inquiry into its composition and dating, arguments about its coherence with other evidence (including but not limited to that of New Testament), as well as the deployment of evidence or examples taken from the Septuagint in all forms of theological literature, whether academic or practical. Eighteenth- and early nineteenth-century scholarship did not necessarily conceive of itself as practicing different techniques or as having different research questions from its predecessors: instead, it was characterized by building on what were perceived to be the solid achievements of those who had come before. In practice, however, this had the effect of opening up a broader terrain of Semitic scholarship in particular, which displaced the Septuagint, to a certain extent, from its privileged position in certain kinds of theological inquiry. Changes in the philosophical and historical understanding of religion, moreover, altered the significance of technical, linguistic scholarship in theology. The authority and interests of those trained in classical studies and those trained in Semitic studies diverged and, in both cases, were no longer motivated primarily by the influence of theology (Kippenberg 2002; Marchand

300

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

2009). This development helped to ensure that even as Septuagint scholarship grew institutionally it remained a comparative backwater to the mainstream of theological study and consciousness after the middle of the nineteenth century.

VIEWS AND DEBATES Writing in 1968, Sidney Jellicoe disposed of early modern scholarship on the Septuagint in a paragraph: “All this, however, was but preparatory. A new era dawned with the English scholar Robert Holmes … and his continuator James Parsons, whose edition of the Septuagint, with its abundance of variant readings, remains today indispensable to serious text-critical study” (Jellicoe 1968: 1–2). More than a hundred years earlier, and less than twenty-five years after the completion of the labors of Holmes and Parsons, E. W. Grinfield (1785–1864) complained that contemporary scholarship “buried [the Hebrew of the Old Testament] under endless appeals to comparatively modern oriental dialects” and ignored the “safety and security” that early modern scholars had found in studying the Hebrew alongside the Septuagint (Grinfield 1850: 57–8). Grinfield believed that he was living through the end of an era, rather than witnessing the sun rise on a new age. Both Jellicoe and Grinfield felt that a primary aim of Septuagint scholarship was the production of editions of texts. What brought them to fundamentally different conclusions about the value of early modern scholarship was the openness with which they acknowledged that theological debate is subject to confessional bias. Grinfield was not only open about this, but also believed that the purpose of scholarship was to determine which Christian denomination was right. In this, he had much in common with early modern scholars themselves, who further held that unfavorable historical interactions between Christians and Jews had been critical for the formation of the text. Jellicoe downplayed this highly contentious aspect of the ongoing justification for Septuagint studies, which in fact persisted in modern theology and philology (e.g., in the work of Paul de Lagarde), as did confessional (and, at times, national) differences in styles of scholarship (Stern 1961: 3–94; Lougee 1962: 88–116; Sieg 2007; Schäfer 2016). This attitude to the differences between modern and early modern theology has had two consequences. First, it obscures the extent to which the production of a critical text of the Septuagint was itself an aim of early modern scholarship, whether in the work of Pierre Morin (1531–1608) or later that of Patrick Young (1584–1652). Second, it neglects the extent to which early modern critics (e.g., James Ussher [1581– 1656]) themselves understood the problems that modern editors have faced with the so-called mixed nature of surviving Septuagint texts. The principal reason for early modern scholars to express skepticism about the value of the Septuagint was that they did not share Grinfield’s conviction that it provided a route to improve on either the textual or theological value of the Masoretic Hebrew Bible. One of the main causes of doubt was detailed awareness of the historical complications of surviving recensions of the Septuagint (made explicit, for example, in the footnotes to the 1597 Frankfurt edition). Modern scholarship has misrepresented this aspect of early modern criticism, both by supposing that attachment to the Masoretic text was always a consequence of theological assumption rather than critical choice and by believing that the “primitive criticism” that constructed early modern Septuagint texts was responsible only for the introduction of “corruption” into the text, and that early modern variants from the

THE SEPTUAGINT IN EARLY MODERN EUROPE 301

surviving tradition could not represent historically validated readings (Walters 1973: 4–5). In fact, the turn from the Masoretic Hebrew text to other (in the first instance, primarily, Arabic) philological traditions in the editing of the Old Testament in the late eighteenth century, which is singled out for praise by Walters, was based on longstanding early modern prejudices about the changes introduced into the Hebrew language by rabbinic theologians and editors who were supposedly motivated by the need to obscure meanings favorable to Christianity. Modern scholarship undertaken in the wake of the post-Vatican II transformation of the Catholic Church has been more obviously sympathetic to early modern attitudes to the Septuagint (Harl 2004: 160–2). In particular, the discovery of Dominique Barthélemy that readings found in the Septuagint column of Ezekiel in the Complutensian Polyglot (1514–17) were similar to those attested to by a papyrus otherwise unknown before 1931, has given life to the reinterpretation of the activities of the editors of the first printed Septuagint edition (Barthélemy 1990). The research that has since been undertaken has demonstrated, first, that the editors who prepared the Septuagint text were by no means always motivated by a desire to make it conform to the Hebrew (or the Vulgate); instead they had clear criteria for preferring texts that included historical judgments about their relationship to the Masoretic text (Sáenz-Badillos 1990). Secondly, it has shown that in the Greek text as well as elsewhere the editors of the Complutensian Polyglot had access to philological materials that preserved evidence of a wider range of possible readings than is straightforwardly identifiable by textual comparison with the known manuscript tradition (Fernández Marcos 2005; Gil 2016; cf. Delitzsch 1886). Thirdly, it has encouraged a detailed examination of the editorial methods used in particular passages of the text that (whatever its other shortcomings) has nevertheless demonstrated the underlying integrity of the known sources (O’Connell 2006). The interest that has recently been focused on the Complutensian Polyglot to some extent replicates the careful retracing of the critical steps of the editors of the Roman Septuagint (eventually published in 1587) and the Sixto-Clementine Vulgate that Hildebrand Höpfl (1872–1934) pioneered more than a hundred years ago. Höpfl, whose work remains unsurpassed as a published study of the editorial progress of an early modern edition of the Bible, wrote at a time when concern within the Catholic Church about the errors of modernism and the dangers of higher criticism made the history of biblical scholarship particularly contentious. Sixteenth-century Catholic scholarship, which itself explored the value of the concept of the inspiration of Scripture and the extent to which human authors and the contexts in which they wrote might have inflected the form and meaning of the biblical text, came to have particular resonance in this context (Höpfl 1908, 1913; Mandelbrote 2016b). Growing ecumenical concern with the history of the canon and with the value of different versions of Scripture as they have circulated in Jewish and Christian communities over time has again raised questions that inadvertently draw attention to the discoveries and interests of early modern scholarship (Harl, Dorival, and Munnich 1988: 321–34; Fernández Marcos 2000b; Rajak 2009). At the same time, the revival of interest in early modern scholarly practices among historians has encouraged the recovery of debates, especially in the circle of Joseph Scaliger (1540–1609) at Leiden, about the nature of Hellenistic Judaism, the languages spoken at the time of Christ, and the historical relationships between Jews and Christians in the Hellenistic and Roman periods (Lebram 1975; Grafton 1983–93, 2016; Dunkelgrün 2016).

302

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

In the last forty years, historians have developed a growing respect for the achievements of early modern intellectual culture and an interest, in particular, in the origins and course of European scholarly engagement with non-Christian and non-European cultures (e.g., Toomer 1996; Grafton and Weinberg 2011). They have explored technical aspects of early modern Christian and Jewish scholarship in ways that have brought to the fore questions that overlap with the sources and problems that have long been familiar to classicists and theologians considering the problems of the Septuagint (e.g., Hamilton 1999; De’ Rossi 2001). Sub-disciplines such as the history of the book have focused renewed attention on the editorial activities of major European presses, notably that of Christopher Plantin, who published a major polyglot edition of the Bible (1568–73) and who printed the first recensions of Syro-Hexaplaric texts (François 2009; Dunkelgrün 2010–11, 2012). Interest in the editorial methods of early modern classicists and theologians and the printed resources that they used in preparing their own published work, moreover, has amplified the work of Höpfl in identifying the value of particular copies of the Septuagint and the way in which they were used by individual readers (Hardy 2015, 2018; Mandelbrote 2019). This research has again demonstrated the confessionalized nature of early modern engagement with the Septuagint and the extent to which the content, language, and meaning of the biblical text, as it emerged from early modern editing and as it was deployed by early modern readers, were shaped by intellectual divisions consequent on the Reformation (Mandelbrote 2016c; Hardy 2018; Twining 2020). The humanist rhetoric of a return “ad fontes” was not straightforwardly combined with the demands of the Protestant Reformation for a Christian doctrine focused on “sola fide, sola gratia, sola Scriptura.” Post-Tridentine Catholic scholarship was at least as concerned with questions of the textual history of the Bible as were members of the Lutheran or Reformed Churches. The content and organization of the Greek Old Testament, in particular its relationship to what Luther and his followers determined were apocryphal texts, were subjects of intervention (whether on the part of editors in print or readers in manuscript annotation) from all sides in editions of the Septuagint that appeared from the mid-1520s onward. The majority of such interventions in print were the product of Protestant rejection of the Vulgate, rather than concern for the manuscript tradition. The editors of the Roman Septuagint of 1587 complicated but did not resolve this problem in their efforts to reconstruct the text as it had stood at a historical moment on the basis of privileged manuscripts (e.g., Codex Vaticanus) and citations from the patristic tradition. Subsequent efforts to recreate or replicate the work of Origen and his school foundered for many of the same reasons that they proved so difficult for nineteenth- and twentiethcentury scholars. The evidence for such work was either lacking, or proved impossible for editors to control. As modern scholars have in any case realized, the pursuit of an Urtext constitutes an effort to pin down a moving target, since variations in what was written reflected genuine differences in traditions rather than demonstrable corruptions of a single text (Nongbri 2018: 9–20).

RESEARCH METHODOLOGIES AND THEORIES Grinfield was in some respects right to recognize the confessionalized nature of early modern scholarship, but he was wrong to infer that Protestant theologians had taken the lead in the use of the Septuagint: the seventeenth-century English scholarship to which he predominantly referred was an attempt to wrest control of a debate rather than a

THE SEPTUAGINT IN EARLY MODERN EUROPE 303

demonstration of command of the field (Mandelbrote 2006). What was determined by the scholarship of the early modern period, although it remained contested until the work of Benjamin Kennicott in the third quarter of the eighteenth century, was that the oldest biblical manuscripts of the Old Testament that survived were written in Greek. Whether or not the Greek text provided the best source of information about the ancient Bible or about the beliefs of early Christians was a subject of continual debate. Mid- to late seventeenth-century efforts to control the text of the Septuagint through external witnesses, whether they be quotations in the New Testament or extra-biblical chronologies such as the Egyptian or Chinese dynasty lists, represented an attempt to come to terms with the problem of the relative authority of the Greek and Hebrew Old Testaments. Such activity may be evidenced with reference to the dispute between the Protestant  critic, Isaac Vossius (1618–89), and his Catholic opponent, the Oratorian Richard Simon (1638–1712) at the end of the 1670s (a debate that soon attracted the interest of the Holy Office). That exchange provides an illustration of the scope of early modern interest in the Septuagint and the extent to which it embraced a broad range of sources relating to the Hellenistic world. Vossius and Simon agreed that the Masoretic text of the Hebrew Bible was established later than the time of the first Christian editors of the Bible (in particular, for Vossius, Origen, whose work in compiling the Hexapla would have been unnecessary had the Masoretic text survived from antiquity). Unlike Simon, however, who was more reluctant than many Catholic critics to ascribe variants in Hebrew copies of the Bible to the perfidiousness of the Jews and, in particular, their rabbis, Vossius attacked the Masoretes as ignorant men who had perverted the text of Scripture, not least (he argued) because, by their time, the sound and pronunciation of Hebrew had been irrecoverably lost. The eagerness that Vossius showed in distancing himself from the standpoint of Reformed Orthodoxy, which held that the vowel points introduced by the Masoretes were an integral part of the inviolable text of Scripture, brought him to similar conclusions to those  of mainstream post-Tridentine, Catholic scholarship, according to which the testimony of the Fathers demonstrated that the Jews had deliberately attempted to introduce false or misleading readings into the Bible to conceal its Christian message. Such Catholic scholarship could not, however, follow Vossius either in his endorsement of the unique value of the Septuagint version, or in the supporting claim that he made when attacking Simon, that the Sibylline Oracles had also had Jewish origin in Ptolemaic Alexandria and preserved a true witness to prophesies later corrupted by the intervention of rabbis who were intent on denying that Christ had been the Messiah (Mandelbrote 2012). Modern critics have largely rejected Paul E. Kahle’s technical representation of the Septuagint as a Greek targum (Jellicoe 1968: 59–63; Fernández Marcos 2000b: 53–7, 101–3). Nevertheless, it is instructive to consider the early modern reception of the Septuagint for a moment in this way. Early modern readers knew and used the Septuagint in part through its effects on liturgical practice and in part as a result of traditions of learned commentary on fragments of the biblical text. They did so because of their recognition of the role of the Septuagint as translated in the Old Latin version of the Psalms that persisted in the Vulgate (and indeed because of the difficulties that they faced in untangling the Latin text of the Psalter: see Dunkelgrün 2012: 296–319). More profoundly, their knowledge of the text depended in large part on its fragmentary transmission in the exegetical context of the Byzantine tradition of catenae. At the same time, early modern readers made themselves aware of the Septuagint alongside their discovery of other targumim, for example on the pages of the earliest printed polyglot

304

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Bibles (Houtman, Van Staalduine-Sulman, and Kirn 2014). The dissemination of such texts complicated the story of the transmission and readership of the Hebrew Bible. In similar fashion, access by scholars in early modern western Europe to Samaritan, Arabic, Syriac, or Ethiopic versions of all or part of the Old Testament raised complex questions relating to the identity and use of the original text of the Bible, which were solved for many in the seventeenth century by positing a relationship that gave primacy to the Septuagint over the Hebrew Bible. Moreover, as early as the mid-sixteenth century, scholars recognized that the reconstruction of such an Urtext of the Greek Old Testament might require its recovery from manuscripts preserving it in translation: in particular, the Syro-Hexaplaric version of Joshua (Masius 1574; cf. Norberg 1787; Bugatus 1788). Early modern scholarship on the Septuagint therefore made linguistic comparisons from the start, whether those are held to lie in the Complutensian Polyglot or in the activities of Aldus Manutius and his heirs in Venice (who initially planned a polyglot Bible of their own and who eventually issued an edition of the Septuagint in 1518 alongside a reprinting of the Greek New Testament first prepared by Erasmus in 1516). A translation was immediately available in Latin in the interlinear version provided by Juan de Vergara to accompany the Septuagint text of the Complutensian Polyglot, which was quickly reprinted separately by Andreas Cratander at Basel in 1526. The Roman Septuagint of 1587 was accompanied a year later by a Latin version, which later formed the basis for subsequent interlinear versions such as that which accompanied the London Polyglot Bible (1654–57). The London Polyglot, in turn, both included historical versions in Syriac and Ethiopic that derived from the Septuagint tradition and printed some evidence of the textual variation between the earliest surviving manuscripts of the Greek Old Testament (Codex Vaticanus, Codex Alexandrinus, and the Cottonian Genesis). The principal Septuagint lexicon, however, edited by Conrad Kircher and published in 1607, encouraged readers to think of Septuagint vocabulary in terms of their understanding of the Hebrew text (Ross 2018b: 16–20). The language of the Septuagint (and of other Greek versions associated with it) was understood more broadly within a long-running debate over the nature of the Greek of the New Testament, which represented the primary basis for most contemporary study of Greek. Over the course of the seventeenth and eighteenth centuries, Dutch, German, English, and Italian scholars, in particular, exchanged views on the extent to which the language of the New Testament was a dialect of ancient Greek and on whether that language was the same as or different from a “Hellenistic” dialect spoken by Jews in Alexandria. They asked whether writers like Philo or even Josephus knew either Hebrew or Aramaic and they considered how far the peculiarities of the language of the New Testament might be the product of individuals trying to translate concepts found in Hebrew or Aramaic into the different modalities of Greek. They wondered about the extent to which the theological vocabulary of the Evangelists had been shaped by the writings of Aquila, Symmachus, and Theodotion, and how far these in turn differed from the presumed original language of the Septuagint itself (Ros 1940; Lebram 1975; van Miert 2018). The effort to trace the development of the vocabulary used in Greek versions of the Old Testament was one of the motives of the revised lexicon of the Septuagint produced by Abraham Trommius (1633–1719) in 1718. It, in turn, depended on new work that Bernard de Montfaucon (1655–1741) had published in 1713 on surviving evidence for the Hexapla. Extreme views were expressed on all sides in considering the language of the Septuagint, but the debate also built on and encouraged broader consideration of the citations of the Old Testament

THE SEPTUAGINT IN EARLY MODERN EUROPE 305

by the authors of the New Testament and, indeed, of the intertextual relationships to be found in Scripture more generally. The standard for this was set by the Franeker professor, Johannes Drusius (1550– 1616), who demonstrated that the citation of the Old Testament by the authors of the New Testament had not represented a simple act of translation from Hebrew to Greek. Equally, he showed that the Greek of the New Testament was replete with phrases and idioms that indicated the influence on it of Hebrew. He was, however, generally confident that much of this influence could be explained by the mediation of the Septuagint. He argued that Erasmus had been wrong to suggest that the translators of the Septuagint of Isaiah had departed from the true prophetic meaning of the Hebrew by using an imperative mood for a future tense, drawing attention to the way in which in the speech of the boys of Friesland (the native province of his university, Franeker) similar alterations to correct syntax occurred. Drusius thus demonstrated that he regarded the Septuagint as a living translation, rather than a work of literary craft, which provided a further explanation of the tendency of the ordinary people who had written the New Testament to make use of its versions. On the other hand, although Drusius was clear that the New Testament authors frequently followed the Septuagint, particularly in their citations from the Psalter (which were thus influenced by the practices of the synagogue), he did not regard the Greek translation of the Old Testament as the only source of the Hebraisms and other peculiarities in the language of the Gospels and Epistles (Drusius 1588; Korteweg 2006). Drusius and his contemporaries were beneficiaries of a significant expansion in contemporary knowledge both of Greek and of Hebrew and its application to historical questions. Over the course of the seventeenth century, the linguistic ambition of scholars widened still further, as did their ability to locate relevant sources both in the libraries of western Europe and through exchange or purchase from the Middle East. Progress in this regard was not always uniform and discovery did not always lead to the successful incorporation of new texts into scholarly understanding. Two examples may be thought representative in this regard. First, the reception of the work of Andreas Masius (1514– 73) demonstrates that although early modern scholars quickly recognized the value of the Syro-Hexaplaric text, they also quickly lost track of the sources that Masius had used in its partial reconstruction. Secondly, the determination to make both confessional and intellectual capital out of the acquisition of Codex Alexandrinus in the second quarter of the seventeenth century in England at no stage succeeded in producing an accurate critical representation of the text based on the tradition found in that manuscript, despite the enormous effort expended on it by scholars of the stature and capabilities of Patrick Young and Johann Ernst Grabe (1666–1711). Their relative failure and its repercussions explain why later scholars have regarded the collations eventually undertaken by Robert Holmes (1748–1805) from the 1780s as foundational of a new approach. In this respect, and despite the remarks made above, early modern scholarship on the Septuagint often promised more than it was able to deliver. The assumptions that underlay the endeavors of early modern editors and commentators in part accounted for this. The most significant of those assumptions was the belief, shared across Christian confessions, that the Bible was itself an inspired text. For some, particularly in the Lutheran tradition, this provided a basis for skepticism about the value of the Septuagint in general and a straightforward argument to prefer the witness of the Hebrew Bible. For many, whether Catholic or Reformed, however, it opened up an extremely complex search for historical authority and meaning that directed attention at what had been known to the Church Fathers, which in turn generated debate about the status of the Septuagint in the early

306

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Church. For example, the 1588 Latin translation of the Septuagint was edited by Flaminio Nobili (1532–90) and built on the efforts of Pedro Chacon (1526–81) and Fulvio Orsini (1529–1600) to collate early Latin citations with the Greek manuscript tradition, in an attempt to reconstruct a Latin version (the “Vetus latina”) that predated the Vulgate. The editors adduced passages quoted by the Fathers to justify choices of words that departed from the Vulgate. They believed that, taken together, the Septuagint and patristic usage revealed the text honored by the practice of the Church, before Jerome had improved it through the composition of his own, more elegant Latin version (Mandelbrote 2016b). Debate over the question of whether the Hexapla or the Vulgate best represented the biblical knowledge of the Fathers divided Catholic as well as Protestant commentators, flaring for example at the end of the seventeenth century in the controversy between the Cistercian Paul-Yves Pezron (1639–1706) and his opponents the Benedictine editor of the Vulgate, Jean Martianay (1647–1717), and the Dominican, Michel Le Quien (1661– 1733). For both Martianay and Le Quien, the work of Jerome represented the apogee of the engagement of Christian editors with the Hebrew of the Old Testament, with which they also saw Origen as being primarily concerned. To Pezron and others, however, the witness of Josephus indicated that the original Hebrew Bible had been lost and was best preserved in some version of the Septuagint. Behind this debate lay centuries of repetition and contrast of the various opinions of Jerome, Augustine, Justin Martyr, and Eusebius (to cite the principal authorities) about the status of the Septuagint, for which the testimony concerning its translation provided by the so-called Letter of Aristeas proved critical. The Letter of Aristeas circulated widely in Latin and subsequently in vernacular translations in both manuscript and print from the 1460s. It was first published in the original Greek in 1561. It provided a justification both for princely patronage of libraries and for the translation of Scripture that proved influential first in Italy and then across Europe. Yet it also raised long-running questions about the relationship between the Old Testament in Greek and Jewish worship, about the true extent of the original Septuagint, and about the nature of biblical inspiration that underpinned much early modern debate. To give a formative example: the humanist Juan Luis Vives (1492–1540) put the arguments against some of the miraculous accretions to the story given in the Letter of Aristeas, and set Jerome to trap Augustine on the question of the relative authority of the surviving witness of the Greek and Hebrew Old Testaments (Augustine 1555: 1105–10). John Fisher (c. 1469), Bishop of Rochester, went through the whole gamut of available evidence when prompted by Richard Pace (c. 1482–1536) in 1527 or 1528, before the controversy over Henry VIII’s divorce fractured the English Church. The Letter of Aristeas, Fisher noticed, did not contain many of the claims made for the providential nature of the Septuagint translation. Those claims were untrue, both because the consent of the Fathers was that the Hebrew contained nothing that was contrary to the teachings of the Holy Spirit, and because it was easy to make mistakes when copying and translating Hebrew, which it was evident the translators of the Septuagint had done. Those translators could not have anticipated the use that Christians would make of the prophecies of the Old Testament and were no more reliable therefore than other Jewish authors. Despite all this, Fisher thought it worthwhile to praise the learning and ingenuity of Ptolemy II Philadelphus, the Hellenistic ruler who had commissioned the translation for the library of Alexandria. The Septuagint was not the product of deliberate fraud, but the circumstances of its translation probably meant that its original text extended no further than the Pentateuch (Rex 1992). Post-Reformation debates, however, encouraged

THE SEPTUAGINT IN EARLY MODERN EUROPE 307

people to place greater weight both on the Letter of Aristeas and on surviving texts of the Septuagint. The editors of the Roman Septuagint of 1587, therefore, who had identified Codex Vaticanus as a manuscript of particular antiquity argued that “[it] was not only older but also better than all others, and in particular that it approximated most closely, if not throughout then certainly for the most part, to the Septuagint translation that we were looking for” and that their edition had restored the text to the pristine splendor in which it had existed before the time of Jerome (quoted in Mandelbrote 2016a: 42). These beliefs were not entirely dismantled by the exposure of the Letter of Aristeas as a forgery first by Humphrey Hody (1659–1707) in the mid-1680s and then by Anthonie van Dale (1638–1708) in 1705, the year in which Hody published his remarkable De Bibliorum textibus originalibus (Wasserstein and Wasserstein 2006; Mandelbrote 2017). From the very different perspectives of the Church of England and of the Dutch Mennonite community, both Hody and van Dale undermined textual traditions that had been brought to bear in order to buttress the antiquity and the authority of the Septuagint. They did not, however, immediately convince all of their readers or potential readers. Hody’s intervention, which showed, for example, that the Greek Old Testament displayed regional and temporal variations in the pronunciation and orthography of Hebrew names, ought to have been decisive in demonstrating that the surviving text of the Septuagint (even in the Pentateuch) could not have been the product of a single moment of translation. Hody presented his readers with a human history of the gradual composition of the Septuagint, but those readers were quite capable of adapting his arguments (or those of Richard Simon) to support a providential story of the creation and eventual adoption for purposes of worship of what (for them) became, by the time of Christ, a singular text of the Greek Old Testament (e.g., Owen 1769).

ONGOING RESEARCH QUESTIONS Henry Owen (1716–95) was a client of one of the foremost patrons of biblical translation of the late eighteenth century, Shute Barrington (1734–1826), bishop successively of Llandaff, Salisbury, and Durham. Barrington was the initial dedicatee of the work of Robert Holmes. Both Owen and Holmes recognized the shortcomings of earlier editions of the Septuagint, but they did not reject the assumptions that lay behind early modern engagement with the Greek Old Testament. Holmes set out the reasons for his work in 1789. It was important because the Septuagint was “made long before the coming of Christ”: for this reason, he still hoped to recover a “pure and perfect” text. The usage of the “the Jewish Church” justified the value of that text, “as a true and faithful version” of Scripture. Moreover, once recovered, the Septuagint could be used to correct “the Hebrew Original,” the preservation of whose text, particularly in the prophetic books, was in doubt. This was because neither the Hebrew text of the Old Testament nor even the New Testament, which could also benefit from comparison with the Septuagint, was free from corruption (Holmes 1789: 1–3). The ongoing research questions that face interpreters of early modern research on the Septuagint, therefore, cannot be as simple as to identify the origins of modern Septuagint research. This is surely a moving target if ever there was one. Instead, there remains a need to understand how and why a comparatively limited set of assumptions about the Septuagint remained so dominant for so long. That project cannot be completed in isolation from consideration of early modern attitudes to the Hebrew Bible and to the canon, both of which are now the subject of more informed scrutiny (e.g., Twining 2020;

308

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Levitin forthcoming). Such activity addresses the broader problem of conceptualizing the world in which early modern scholars placed the Septuagint. That world was shaped by contemporary knowledge of the Bible and of patristic literature, by the understanding of the history of the Christian Church and its relationship to Judaism, but also by the rediscovery of a large number of sources related to the ancient world that had not been previously known or had not been used as a prism through which to examine the authority and witness of the Bible. Such sources included real and imagined pagan histories, such as those of Manetho, Berossus, or Sanchoniatho, as well as pseudo-Aristeas and Josephus. The recovery of early modern attitudes to the Septuagint has already revised assumptions about the trajectory of critical scholarship and about the chronology that one should apply to the history of biblical criticism. It thus raises questions about the significance and achievement of scholarship both in the past and the present. Important figures have been added to the canon of those who should be considered as significant contributors to the growth of historical knowledge of the Bible, but the idea of such a canon is itself flawed. The use of confessionalization as a route to understanding disputes and fault lines in the history of biblical criticism brings out the extent to which early modern writers conceptualized the role of criticism itself differently from modern readers (Hardy 2017). Early modern writers were not our precursors and they did not have our aims in mind, however much we may find to admire or to criticize in what they did. As such, it is hard to imagine that early modern readers of the Septuagint have much to teach modern critics about method. Their experience, however, underlines the difficulty of both reaching and disseminating firm conclusions. Twenty years ago, there was no history of early modern Septuagint studies to speak of. Now, the importance of the Septuagint to the early modern mind is widely recognized and a whole generation of scholars have contributed to its understanding. There remains much to do. In particular, the pursuit of debates in intellectual history has tended to obscure the importance of readership of the Septuagint as a tool in translation and above all in education (cf. van Miert 2018). It has not yet exposed the extent to which knowledge of the Septuagint, like knowledge of other original texts of the Bible, informed pastoral practice, most obviously in the writing of sermons, but also in the giving of counsel (cf. Steiger 2008). Early modern writers and readers, after all, were not primarily concerned with historical questions or even questions of authority: their interest instead was in human and moral issues for which the richness and copiousness of the Bible (including the meanings provided by the Septuagint) offered hope and enlightenment (Mandelbrote 2013). Biblical criticism, insofar as it existed, was one aspect of polemical and doctrinal scholarship and of the educational experience of the higher clergy: it was not, strictly speaking, an end in itself. There are also questions of scope. Some of these are chronological. Far more attention has been paid to the active controversies of the seventeenth century than to those of the sixteenth or the eighteenth centuries. The timing of any transition from early modern to modern attitudes of criticism, moreover, remains deeply contested: it is far too easy to project the world of Wellhausen back onto the attitudes of nineteenthand even eighteenth-century critics whose linguistic and historical assumptions were in fact quite different (Nicholson 1998). More serious, however, has been the tendency of early modern intellectual historians to privilege western European perspectives on scholarship. Although there has been interest in Jewish attitudes to the Septuagint, the most significant study in that area relates to a Jewish scholar in northern Italy who undoubtedly read Christian commentary on the Septuagint and was in turn widely

THE SEPTUAGINT IN EARLY MODERN EUROPE 309

read by contemporary Christian intellectuals (De’ Rossi 2001). Very little has been written about the role of the Septuagint in Orthodox Christianity or even in Orthodox intellectual life in this period (see Seleznev in this volume). This neglect has occurred despite the fact that exchange with Orthodox communities underpinned significant turns in western scholarship and interest in the Septuagint and that Venetian printers in turn produced editions of the Greek Old Testament, based on the scholarship of western European critics, for distribution to Orthodox Christians. Orthodox scholars studied at both Leiden and Oxford during the seventeenth century. The knowledge that they gained of Western attitudes to the Septuagint informed their own debates about the canon, the usefulness of vernacular Scripture, and the role of the Bible in the ordering of the Church (Ware 1964; Tsourkas 1967).

RECOMMENDED RESOURCES 1. Cameron (2016) provides in various chapters introductory material on the nature of biblical publishing and editing, including discussion of the editing of the Septuagint, and on the controversies involving the text and authority of the Bible. 2. De’ Rossi (2001) is an edition of The Light of the Eyes by the Manutan scholar Azariah de’ Rossi (1511–78) that offers an essential study of Jewish attitudes to the Letter of Aristeas. 3. Wasserstein and Wasserstein (2006) follows, among other things, the later, Christian history of the significance of the Letter of Aristeas. 4. Mandelbrote (2016a) considers the manuscript tradition of the Letter of Aristeas and its implications. 5. Höpfl (1913) remains fundamental for those who wish to explore the attitude of the Roman editors of the Septuagint in the immediate aftermath of the Council of Trent and the identification of the importance of Codex Vaticanus. 6. Mandelbrote (2016b) provides a broader context for understanding the work of the Roman editors. 7. Hardy (2017) gives the fullest treatment of mid-seventeenth-century debates over the significance of the Septuagint, building on the legacy of Scaliger. 8. Van Miert (2018) is not always in agreement with Hardy (2017), but is fundamental for setting Septuagint scholarship within a broader tradition of biblical editing and translation in early modern Dutch philology. 9. Twining (2020 and forthcoming work) can be expected to broaden and deepen our understanding of Catholic criticism. 10. Mandelbrote (2006) treats English scholarship of the seventeenth century, with a glance to the context provided by the Orthodox world. 11. Mandelbrote (2016c and 2017) explore to some extent the nature of eighteenth-century scholarship, both professional and lay.

310

PART FIVE

Theology, Translation, and Commentary

312

CHAPTER 21

The Septuagint and Theology W. EDWARD GLENNY

INTRODUCTION AND HISTORY OF SCHOLARSHIP One of the reasons people study the Septuagint (LXX) is to understand better the theological and philosophical beliefs of its translators and their communities. This endeavor is grounded on the fact that the Septuagint comprises translations of the various books of the Jewish Scriptures. Those translations, like all other translations, are, by the nature of the translation process from one language to another, inexact representations of the original source texts (see Joosten 2000: 31–2), and the source documents are fundamentally religious documents. Thus, the resulting translations contain some unique theological elements, differing to various degrees from their Semitic sources (Ausloos 2017: 236–8). Although in the nineteenth and early twentieth centuries scholars were motivated to study the Septuagint because it was thought to be helpful for tracing the evolution of the theology of the Hebrew Bible, the discovery of the Dead Sea Scrolls lessened that motivation. They provided evidence that there was a plurality of text forms of the Jewish Scriptures in the time the Septuagint was translated and that the text that is represented by the Septuagint is in some places earlier than the text found in the Masoretic Text (MT) (Tov 2015a: 201–23; Ausloos and Lemmelijn 2020: 21–2). At present, the main incentive for studying the theology of the Septuagint is for its contribution to the study of Hellenistic Judaism and Christian theology, and it is primarily faith communities that see the greatest value in this undertaking. Tov (2020b: 25) explains, “The main benefit of consulting a theology of the LXX and the Hebrew Bible is for those who take these literatures as completed units, which in the case of the LXX are the church fathers and the believing communities.” The increasing interest in the theology of the Septuagint can also be explained simply as one aspect of the growing interest in the Septuagint in general (Ausloos and Lemmelijn 2020: 19–20). This burgeoning interest has been accompanied by an ever-increasing number of resources that make it more possible than ever before to study this corpus of literature (Glenny 2016: 263–4). It is helpful to consider the work that has laid the foundation for the present interest in Septuagint theology. Prior to the nineteenth century, there were many who recognized differences between the Septuagint and the Hebrew Bible (e.g., Origen, Jerome, Augustine), and there were many, most notably in the Church, who based their theology on the Septuagint, considering the Greek translations to be the Word of God. Most

314

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

early Christians would have thought that the Greek matched the Hebrew and thus not have considered that the Septuagint might have a distinct theology. At any rate, there was little interest in writing a theology of the Septuagint or developing systematically a work demonstrating where the Greek translation evidenced theological development or differences from its Hebrew/Aramaic Vorlage. Origen’s textual work might be the exception, but his work was focused as much on the establishment of the Greek text as on its theology.1 Apparently, early readers understood these documents to be “theological texts,” as interpreters do today (see Ausloos and Lemmelijn 2020: 23, 30; see also Tov 1999e: 257), and they treated them as such, often drawing theology from them (such as the reference to a “virgin” in Isa. 7:14). But there is no developed discussion of a distinct theology of the Septuagint in comparison to the Hebrew Bible. Thus, during the long period from the translation of the Septuagint until the nineteenth century, Greek texts were either considered to be Scripture along with the Hebrew (Augustine) or one of the two traditions was preferred over the other, but the differences between the two were not developed with regard to their theology. The real foundation for the contemporary discussion concerning a theology of the Septuagint can be traced back to the first half of the nineteenth century, when scholars began to critically study the differences between the Hebrew Bible and the Septuagint translations of it (Rösel 2018c is especially helpful; for historical overviews see Tov 1999e: 257–61; Kreuzer 2017; Rösel 2018a: 273–5; Ausloos and Lemmelijn 2020: 19– 23). In 1841, Frankel set out to address introductory issues concerning the Septuagint in his first work, Vorstudien zu der Septuaginta, and this work included in it a section on “Hermeneutic and Exegesis of the Septuagint” (163–203; it contained examples of “free exegesis” in §29–30). In 1851, he wrote a critical introduction to the books of the Greek Pentateuch, treating each individually, including evidence of the characteristics of each translation and references to philosophical and religious exegesis in each translation. His basic methodological perspective was “to provide information through Palestinian exegesis on Alexandrian hermeneutic,” and he saw the translators as “more or less exegetes” (1851: 3, 21). Geiger (1857) published another important work discussing the differences between the Hebrew and Septuagint in which he compiled many important observations on the theologically motivated translations (see esp. 321–423; he calls the section “Tendentious Change”). And after considering the influences upon the translators, Freudenthal (1890: 221–2) concluded that it was not the influence of Greek philosophy that triggered the Septuagint translators’ renderings, but rather that their thoughts were Jewish. At the beginning of the twentieth century, Deissmann (1903: 174), writing about the Hellenization of Semitic religion, famously asserted that “the Bible whose God is Yahweh is a national Bible; the Bible whose God is κύριος is a universal Bible.” He pointed out this change in the main title of God and other shifts of theological emphasis that took place in the Septuagint, like the rendering of ‫ ברית‬as διαθήκη. About the same time, Flashar’s studies in the Psalter focused on “exegetical investigations” and the “peculiarities of the translator” of that book (1912: 83–4). In the 1920s, Kaminka (1928: 267–73) concluded his Studies on the Septuagint Minor Prophets by pointing out some of the ways the Septuagint differs from the Hebrew in the Twelve, including some theological issues. In the 1930s, Dodd (1935), using material he had presented in the Grinfield Lectures on the

On the Hexapla see Gentry in the present volume.

1

THE SEPTUAGINT AND THEOLOGY 315

Septuagint, suggested that characteristic ideas of the Old Testament were changed through its translation and that the Septuagint contains a distinctive, self-contained theology. In this same time period, Bertram described the Septuagint as a testimony of a changed religion (1936: 161), containing a “Septuagint piety” (1939: 101) that differs from the Hebrew text, and later he wrote about the Septuagint being a “preparation for the gospel” (praeparatio evangelica) (1957).2 By this point, there were several scholars who saw the differences between the Hebrew and Greek bibles as evidence of a changed or new religion. Two important works contributing to the topic of theology in the Septuagint appeared in 1948. The first was the published dissertation of Prijs, Jewish Tradition in the Septuagint, in which he developed the title by arguing that the variations between the Hebrew and the Septuagint were the result of Jewish exegesis and give evidence of a developing Jewish worldview. In the same year, Seeligmann published his important work, The Septuagint Version of Isaiah, in which he discussed Septuagint Isaiah as a “Document of JewishAlexandrian Theology” (1948: 95–121). According to Seeligmann, the translator of Isaiah understood many of the prophecies of the Greek text of Isaiah to be fulfilled in his time (second century BCE). Seeligmann suggested theological study in the Septuagint should be based on the differences between the Hebrew and Greek texts and that it should begin with study of the themes of God, Torah, and Israel, and then develop these themes with further subthemes. The beginning of another stage in the discussion of theology in the Septuagint is marked by the invitation of Ziegler (1962: 28). He stated that the “legacy and task” (Erbe und Auftrag) of Septuagint research is “the rewarding and fruitful work of carrying out the preparatory research necessary so that the long awaited theology of the LXX can finally be written.”3 Ziegler’s vision was not realized in the immediately following years, being overshadowed by interest in textual questions and the possibility of variant Hebrew Vorlagen owing to the discoveries in the Judean desert. In addition, about this time Hanhart (1967) raised objections to writing a Septuagint theology on the grounds that the Septuagint should only be understood as a copy of its archetype and represented a decline from it; thus it should be integrated into a theology of the Old Testament, a suggestion that has not received support. In a 1987 essay, Tov affirmed Ziegler’s belief that a Septuagint theology could be written. But he went on in the article to explain that what he had in mind was not a comprehensive theology of the Septuagint, but what he calls dimensions of “theological exegesis” (see esp. 244–60), and that there is a relatively small amount of it after the problems of unclear Vorlagen are considered. In the 1990s, there was interest in the theology of different books in the Septuagint, as evidenced by the published dissertations of Olofsson (1990) and Schaper (1995) on the Psalms and Rösel (1994) on Genesis. The first volumes of the Bible d’Alexandrie commentary and Wevers’ notes on the Greek Pentateuch (1990, 1993b, 1995, 1997, 1998), all focusing on individual books, were also published in that decade. At the turn of the century, Joosten (2000) reflected on the methodological issues involved with crafting a theology of the Septuagint. He emphasized that the theology of

For a short summary and cautions about Bertram’s sometimes ideologically motivated anti-Jewish interpretations, see Rösel (2018a: 274) and Tov (1999e: 269). 3 The original reads: “Es ist eine dankbare, fruchtbare Aufgabe, solche Vorarbeiten zu machen, damit endlich auch einmal eine längst ersehnte Theology der Septuaginta gerschrieben kann.” The English translation is from Rösel (2018a: 273); see also Tov (1999e: 257). 2

316

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the Septuagint involves the places where it differs from its Hebrew Vorlage, but not all differences are theological. Differences that are the result of a different Vorlage, changes made in textual transmission, misunderstandings of the translator, or changes consistent with the translator’s normal translation technique would not be theological in nature. Only theological changes intended by the translator give evidence of his theology.4 Joosten also understands the characteristic literal translation of the Hebrew to contribute to the theology of the Septuagint. This aspect of theology, which indicates that the translators understood the text to be inspired and justified their harmonization of passages in it, is what he calls the “theology of the Word” (2000: 46). In a 2002 essay, Dafni wrote about what she calls “The Theology of the Language of the Septuagint” (“Theologie der Sprache der Septuaginta”) that can be found in the misinterpretations and specific word choices of the translator. For Aejmelaeus (2007b) the identification of theology in the Septuagint requires the addition of interpretive elements that are found in the differences between the Hebrew and Greek texts. To the present day, there has been no attempt to write a comprehensive theology of the Septuagint, but Rösel (see 2018c, 2018d), and Ausloos and Lemmelijn (2020) are editing collections of studies of the theology of different portions of it. There is clearly a debate among scholars concerning the extent of distinct theology in the Septuagint, and Pietersma sees the debate as “something of a crisis on the hermeneutical front of the discipline” (2006c: 3). Pietersma (2006a) divides work on the theology of the Septuagint into two main approaches, and he describes proponents of the two main approaches as the “minimalists” and the “maximalists.” He is a representative of what he calls “minimalists,” those who see a minimum of exegesis and theology in the Septuagint. Others who have been identified with this position are the scholars who comprise the Finnish school and some members of the NETS translation team. At the opposite end of the spectrum are the “maximalists,” whom Pietersma suggests are represented by scholars like Schaper and Rösel. Minimalists see the translator as a “medium,” creating the potential for exegesis, while maximalists understand the translator to be more an “author” who is doing exegesis on the source text (Pietersma 2006a: 35–6). Pietersma’s methodology is decidedly focused on the source text, and he emphasizes the translator’s role in communicating what is in the Vorlage. Tov (2020b) offers another perspective on the differences of opinion concerning the study of the Septuagint and its theology. He understands the theological and the textcritical to be two different approaches to the Septuagint. The two approaches are not mutually exclusive, but refer to general methods, emphases, or paradigms for interpreting the Septuagint. For Tov (2020b: 23–4), the recognition of theology in a translation differs from the recognition of textual differences because the former involves an “element of interpretation” and is a “subjective recognition of a way of understanding elements in the translation,” while the latter is a more objective description of what is seen in the translation. For these reasons, he is more open to explaining differences between the Septuagint translation and its Vorlage by text-critical factors, while someone who would favor the theological approach would be more open to understanding and explaining differences by using theological factors. Tov feels scholars are inclined to look at texts through one of these two paradigms and use one approach or the other, and the approach they use will often determine the way they interpret the differences from the Vorlage

Joosten also says that “unconscious inferences” in the translation are a source of its theology (2000: 35).

4

THE SEPTUAGINT AND THEOLOGY 317

in the Septuagint. Tov (2020b: 42) concludes, “It is very important to realize that the decision to ascribe a deviation to a theological or textual factor depends on a scholar’s personal inclination,” and he emphasizes “the role of intuition” in this decision-making process.

RESEARCH METHODOLOGY Scholars differ on the amount of unique theology in the Septuagint and the degree to which it is possible to systematize the theology found there. Foundational to these differences are disagreements concerning the methodology one should employ in seeking to determine the theology of the Septuagint, and it is to that subject that we now turn our attention.5

What Do We Mean by “Theology of the Septuagint”? The Term “Septuagint” Every term in the phrase “theology of the Septuagint” is disputed and needs to be defined by those who use it. The most basic term is “Septuagint” (LXX), and technically speaking no such thing as “the Septuagint” exists (Williams 2011). The term has been used to refer to various collections, recensions, and versions of the Greek Jewish Scriptures, starting from the Greek translation of the Pentateuch until modern editions. My use of the term “Septuagint” in this essay, unless otherwise noted, is a general use of the term, referring to the Greek Jewish Scriptures, consisting primarily of the books of the Hebrew Bible and contained in modern editions, like Rahlfs and Hanhart Septuaginta (2006) or the “Göttingen editions” where they are available (see Tov 2015a: 12, 36; Glenny 2016: 263–5). When studying theology in the Septuagint, scholars often limit their discussion to the Greek translation of the books in the Hebrew canon.6 The Term “Theology” The term “theology” is no less difficult than the term “Septuagint.” Even scholars who affirm that such a thing can be found in this corpus use a diversity of terms to refer to it. Tov (1999e, 2020b) calls this theological content “theological exegesis” or “exegetical elements” in the translation. Pietersma (2006a: 33, 39, 45) speaks of “exegesis.” Cook (2005: 65) suggests the terms “theological and/or ideological tendencies.” Dafni (2002: esp. 327) refers to the theology of the “language of the Septuagint.” Ausloos and Lemmelijn (2020: 29; Ausloos 2017: 242) use the term “theological accents.” Joosten (2000: 32, 46) and Rösel (2018a: 279–80) refer simply to “theology.” Rösel (2018a: 279–80) attempts to define precisely what “theology of the Septuagint” means, and his description is a good place to begin. He distinguishes between understanding the term “Septuagint” in the phrase “theology of the Septuagint” as a Interestingly most of those differences are related to basic questions like the nature of the so-called Septuagint, the identity of the translators, and the methods and intentions of the translators. 6 See the Introduction to this volume for further discussion of the term “Septuagint” (LXX) and the heterogeneous nature of the collection of writings to which it refers. As a point of clarification, most specialists use the term Old Greek (OG) to designate a (critical) text that, in their judgment, represents the original translation of books other than the Pentateuch (McLay 2003: 6), and some use the abbreviation LXX/OG, when referring to the initial translations of the Hebrew Bible as a whole, as a reminder of the diversity that characterizes the corpus (Jobes and Silva 2015: 16). 5

318

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

subjective genitive and an objective genitive. As a subjective genitive (i.e., the Septuagint’s theology), it signifies “that the focus of investigation is the implicit theology that emerges from the intentional decisions of the translators.” Thus, it is the theological truth implicit in the Septuagint that becomes “the locus of theological inquiry.” By contrast, as an objective genitive, “theology of the Septuagint” refers to theology that emerges from the Septuagint, that is, “theological systematization resulting from the translated text.” He explains that only some Septuagint books, like portions of Wisdom of Solomon, contain this type of “doctrinal exposition or apology.” When Rösel refers to the “theology of the Septuagint” he means an “implicit theology” (subjective genitive) that involves theologically motivated and influenced translations and theological statements intended by the translator either consciously or unconsciously.7 This kind of theology normally involves some sort of theological reflection on the part of the translators or their communities. The systematization of terms and ideas in the Septuagint translation, like the terms for the altar or the names for God, demonstrates that this implicit theology (subjective genitive) can involve intellectual reflection (Rösel 2018a: 280).8 Rösel (2018a: 280) mentions that the idea that such implicit theology “emerges from the intentional decisions of the translators” is an area of disagreement. In his later work, he qualifies this requirement, using the description of God in Hab. 3:5 as an example of how hard it is to know if a distinct reading in the LXX has arisen with theological intent. In that verse the translator renders ‫“( דבר‬plague”) as “word,” which changes the context of the passage; he then adapted the second part of the passage to describe God anthropomorphically, avoiding any connection of God with sickness or a foreign god, making the determination of theological intent difficult to discern (2018d). Aspects of Rösel’s theological method also suggest he does not always limit the theology of the Septuagint to distinct readings that the translator intended to be theological. For instance, he proposes that the focus on the translator as the creative personality in the translation process be switched instead to the reflection and systematization that took place in the translator’s community (2018a: 281–2). This would mean that at least some theological reflection could take place before the translation process and the theology or theological reading of the text could already be established in the mind of the translator before the act of translation, resulting sometimes in a theological intention that is unconscious at the time of translation (see Glenny forthcoming). The word “of” is also important in the phrase “theology of the Septuagint,” as was implied in the discussion above about “Septuagint” being a subjective or objective genitive. McLay (2010: 608) suggests replacing “of” with “in,” resulting in the phrase “theology in the Septuagint.” He sets forth three basic points on which he proposes one should build a “theology of [or in] the Septuagint”: (1) “a theology of—or more accurately, a theology in—the Septuagint, is not limited to the Old Greek (OG) text,” but it also includes later versions and manuscripts; (2) it “is not limited to the differences one might isolate between the Greek texts and the presumed Semitic source text,” but it also includes passages where the Greek text agrees with its Vorlage; and (3) it “may be described with the same legitimacy and using the same basic principles as a theology of the Old Testament/Hebrew Bible or New Testament” (2010: 608). Thus, for McLay, Theology that is intended unconsciously might be theology in the translation that is the result of the influence of the translator’s community. 8 For example, the Hebrew Bible uses ‫ מזבח‬for pagan altars and for legitimate Israelite altars, but the LXX uses θυσιαστήριον for legitimate altars and βωμός for pagan altars. 7

THE SEPTUAGINT AND THEOLOGY 319

Septuagint theology involves all the theology “in” the Septuagint, like a New Testament or Old Testament theology. He feels this is only reasonable, since the Septuagint was the Bible of the early Church and therefore it has the same authority as the Hebrew text, especially for Christians.9

How Do We Establish the Theology of the Septuagint? This section is a summary of eight suggested guidelines that one might use when seeking to determine the theology of the Septuagint. 1) As Far as Possible, Theology of the Septuagint Should Be Based on the Old Greek This does not mean that other versions have nothing to add or should not be considered in theological discussion as the theological process develops, especially if one is considering the Septuagint translations and versions in their historical contexts. However, the reception history of the Septuagint is a different aspect of study than its theology. Ausloos (2017: 242–3) makes a helpful distinction between the theology of the Septuagint and the theological interpretation of it in later texts. This distinction is especially true in the Christian tradition, since Christian writers, including the authors of the NT, often interpreted Septuagint texts in Christological or messianic ways that may not have been intended by the translator. An example is the Christological actualization of Isa. 7:14 in the gospel of Matthew. 2) Each Translation Unit Should Be Studied Individually before Attempting to Construct a Larger Theology of the Septuagint The Septuagint contains many different translation units that were translated by different translators over several centuries (Ausloos and Lemmelijn 2020: 25–6). It is important to consider the separate books in their historical and cultural contexts and not apply conclusions from one book or translation unit to others. Translation units may include several books, as with the Twelve, but normally they are limited to one book or occasionally a portion of a book. If one applies the theological conclusions from one translation unit to others, there is danger of leveling out the theological differences between the books. 3) Distinct Septuagint Theology Results from the Translation Process and Appears Primarily in the Theological Discrepancies and Variations between the Source and Target Texts According to Tov (2020b: 24), “Theological exegesis of the LXX may be defined as any theological element added to the source text by the translator” (see also Joosten 2000: 33).10 These theological elements could be “added” by the additions, changes, and sometimes even omissions in the translation.11 If one follows McLay’s approach, including the agreements between the Vorlage and the translation, Septuagint theology risks being a weak counterpart to the theology of the Hebrew Bible (see Joosten 2000: 33). See the definition of the Old Greek (OG) in the Introduction. 10 Tov (2020b: 24–5) writes that “Theological exegesis relates to the description of God, and his acts, the Messiah, the exile, as well as the whole spectrum of religious experience”; he also mentions Seeligmann’s (1990: 224) narrower definition: “God, Israel, comprising the Messianic idea as a national redemptive force, and the Torah,” which Joosten (2000: 32, n. 3) thinks is a little more systematic than Tov’s definition. Ausloos and Lemmelijn (2020) divide their discussion of theology into nine broad topics: God, Law, Cult, Prophecy, Man before God, Wisdom, People and Covenant, and Promise. 11 The MT is usually the default “source text” employed when studying discrepancies between the LXX and its source; however, the interpreter must also consider the possibility of other source texts and readings that differ from the MT (see 4a below). 9

320

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

There are some who would not follow this principle. As mentioned above, McLay (2010: 25) believes that the theology of the Septuagint need not be limited to the passages where there is a difference between the Semitic Vorlage and the Greek, since all the texts in the Septuagint contribute to its theology and the Greek texts were often read without constant reference to their Semitic sources. Rösel (2018a: 283) also feels that, when one is considering the theology of the Septuagint, there should be some emphasis on the passages in the Septuagint that agree with the Vorlage, since they also have implications concerning the rich theology in the Septuagint. However, he does not subscribe as fully to this method as McLay does, since Rösel’s approach is more historically focused than McLay’s. Implicit in this guideline is the importance of the Greek text for determining Septuagint theology. Such theology is not based on the sociohistorical context of the translation but upon the theological divergences from the Vorlage found in the Greek text (Ausloos 2017: 249). 4) Theological Statements in the Septuagint Are Statements That the Translator Intended to Be Theological Although most believe the distinct theology of the Septuagint is to be grasped in the differences between the translation and its Vorlage, many differences are not an indication of the theology of the translator. In the words of Ausloos and Lemmelijn (2020: 34), “Observing a difference is one thing, interpreting it takes another step.” Differences that are based on a different Vorlage, a scribal error, the consistent translation technique of the translator, a mistake of the translator, or linguistic or contextual exegesis (Tov 2020b: 30) are to varying degrees not accepted as theological interpretations and sources for Septuagint theology, because the translator’s motivation for these types of renderings was not theological. In order to determine that the reason for the differences between the Vorlage and translation is theological, the theological interpreter must study the Semitic and Greek texts, considering text-critical and translation technique issues. Only on the basis of such thorough study can one make decisions concerning the reasons for discrepancies between the source text and its translation. There are several difficulties with the implementation of this guideline, and not all follow it to the same degree (see Glenny forthcoming). For instance, it is difficult to determine the translator’s intent and motivation, and it is also questionable whether one can completely separate the translator’s theological ideology from changes that are made for contextual and linguistic reasons. It is also questionable whether the translators could work so independently of their communities and the ideology of those communities that these things would not affect their work.12 In spite of the difficulties with the implementation of this principle, all follow it to some degree, especially where it is clear the discrepancy in the Septuagint is the result of a different Vorlage or an error in the transmission of the Greek text. Some examples will hopefully help clarify the guideline and the difficulty in implementing it. a. Different Vorlage The first example involves determining whether the source of the difference might be a different Vorlage. In Exod. 15:3 the Septuagint renders MT ‫“( יהוה איש מלחמה‬Yahweh is a man of war”) with κύριος συντρίβων πολέμους (“The Lord who shatters wars”).

See Seligmann (1948: 96), who feels even errors, which are unconscious changes, can reflect a degree of the theological thinking of the translator.

12

THE SEPTUAGINT AND THEOLOGY 321

Rösel (2018a: 281) contends that since there is no Hebrew textual witness that supports this Septuagint reading, there is no reason to attribute the difference to a different Vorlage and so it should be understood as a theological statement on the part of the translator. In support of Rösel’s understanding, the Greek version of the verse is quoted and used to make a theological assertion in Jdt. 9:7 and 16:2 (Perkins 2007: 135–6). Douglas (2012) reads the evidence differently and argues that a different Hebrew Vorlage is the cause of this difference. Although the exact proposed Vorlage reading does not occur elsewhere, he bases his argument on Scripture parallels in Hos. 2:20 and Ps. 76:4 that could support the assumption of a reconstructed Septuagint Vorlage, ‫יהוה שבר‬ ‫“( מלחמה‬Yahweh shattered war”). Tov (2020b: 26–7) agrees with Douglas and sees further evidence for a different Vorlage in the Samaritan Pentateuch text of Exod. 15:3, ‫יהוה גיבור‬ ‫“( במלחמה‬Yahweh is a man for war”). This variant differs from the presumed reconstructed 13 Vorlage of the Septuagint in only one consonant (compare ‫ שבר‬and ‫)גיבור‬.  Therefore, since we cannot be certain the Septuagint reading is a result of the default translator’s creative thinking, Tov feels the burden of proof should be on those who see evidence of Septuagint theology, not on those who assume a Hebrew variant.14 He admits that the textual evidence for this decision “is not manuscript evidence, but [it] suffices to support the assumption of a variant.” Tov wants “certainty” that the Septuagint rendering was a result of the translator’s “creative thinking” before he bases theology on the rendering in the LXX, while Rösel requires more textual evidence than Tov does before he will attribute a different reading in the translation to a different Hebrew Vorlage. They differ on where the burden of proof lies in such decisions. In connection with this issue, it is worth mentioning that Rösel (2018a: 281) questions whether a reading that is based on a different Vorlage should be considered to be “theologically insignificant.” He feels that our theological study of the Septuagint needs to move away from focusing on the translator as a lone creative personality, and instead we should understand the theology of the Septuagint to be a result of the “process of reflection and systematization of Jewish Hellenistic communities whose religious beliefs influenced to various degrees the translation of the biblical texts.” He also feels such interpretation could take place equally by the translator at the time of translation or by a scribe at the time of transmission of the text; what is decisive about such theologically significant material is that “it can be found in the Greek Bible” (2018a: 281–2). b. Scribal Error Another exception to the general principle that differences between the Septuagint and its Vorlage reflect the theology of the Septuagint involves scribal error. If differences in meaning between the Vorlage and the translation are the result of scribal errors, especially in the Greek tradition, the value of the Greek reading for theology is generally rejected.15 In this case, the reading does not reflect the translator’s theology but only a textual variant. It should also be noted that scribal errors in the transmission of a text are very difficult to distinguish from textual updating by a copyist or so-called redactor.

In later Hebrew ‫ גיבור‬replaced ‫ איש‬as the word for “man.” The difference actually seems to involve two consonants, since the reading in the Samaritan Pentateuch also has a ‫ ב‬prefix on the third word in the phrase, which is not found in the MT, but it does not change the meaning much. 14 Compare Tov’s (2015a: 44, 48) emphasis on “inner-translational factors” elsewhere. See also the discussion of this passage in Glenny (forthcoming). 15 If the difference is a result of scribal error in the Hebrew textual tradition, it will likely fall under the previous category involving a different Hebrew Vorlage. 13

322

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

c. Changes Following the Typical Translation Technique Not only is the study of textual issues important for discerning the theology in a Septuagint text, but the study of the translation technique is also important.16 The principle that applies here is that if differences between the Vorlage and the translation are the result of the consistent application of the translation technique of the translator, then those differences do not reflect his theological ideas but are rather his normal, default approach to the text. In such cases the translator was following his developed techniques, and he was not attempting to express theology but simply to render the text well. Joosten (2000: 36–8) illustrates this principle from the translation of Amos 9:11-12. The MT text of 9:12 reads ‫“( למען יירשו את שארית אדום‬that they may possess the remnant of Edom”), and the corresponding translation is ὅπως ἐκζητήσωσιν οἱ κατάλοιποι τῶν ἀνθρώπων (“that the remnant of mankind may seek [me]”). It is generally thought that the translator confused or changed yod and dalet in the main Hebrew verb in 9:12, resulting in the rendering “seek” (apparently reading ‫ )ידרׁשו‬rather than “possess” (‫ ;)יירׁשו‬also he apparently revocalized the noun “Edom” (‫ )אדום‬so it read “Adam” (‫)אדם‬. In Joosten’s view, because the translator was attempting to render the Vorlage well, sticking closely to the words in the text and simply applying his normal technique, these choices do not reflect theological exegesis.17 For him, in order for a difference from the Vorlage in the Septuagint to count as theology in the Septuagint it must be influenced and caused by the theological intentions of the translator (see also Tov 2020b).18 d. Changes Resulting from the Translators’ Mistakes If the difference between the Vorlage and the Septuagint translation is the result of a translator’s mistake, the theological value of the rendering should be questioned. It is clear that often the translators did not understand the Semitic texts they were working with (Tov 1999a). This is likely the case in the rendering of the clause ‫“( מה שחו‬what is his thought”) as χριστὸν αὐτοῦ (“his anointed”) in Amos 4:13. Apparently the translator did not know the hapax legomenon ‫“( שח‬thought”) in Amos 4:13, and in his attempt to make sense of the words, he rendered them ἀπαγγέλλων εἰς ἀνθρώπους τὸν χριστὸν αὐτοῦ (“announcing his anointed to humans”). This rendering makes sense as a translation of a Hebrew text similar to the MT (Glenny 2018) and may indicate a “messianic perspective on the part of the translator” (Jobes and Silva 2015: 338). However, it is also possible that the Septuagint rendering was the result of a mistake on the part of the translator. If so, the translator was not intending to make a theological statement, and the reading is not theological in nature. According to this view, this apparently mistaken rendering and

For different views and approaches to studying translation technique, see Aejmelaeus (2007b), Ausloos and Lemmelijn (2020), Tov (1999d), and Glenny (2009). See also the chapter in this volume by Dhont. 17 Not all agree with Joosten’s treatment of Amos 9:11-12 or his application of this principle. Other factors besides the use of the use of the translator’s normal translation techniques might influence our understanding of theological intent. In Amos 9:11-12, the facts that the words in which the consonants were read differently are common words and that there is a pattern of interest in Gentiles in this translation (see also 4:13 and Glenny [2009: 216–28]) suggest there may be more going on here than the normal application of a translation technique common elsewhere in this translation unit. See Glenny (2009 and 2012) for discussion. 18 This requirement greatly limits the amount of theology one might find in the Septuagint. For example, Joosten (2000: 38) writes that in their research his study group on the book of Hosea found only three passages that “seemed to be influenced by the theology of the translator.” 16

THE SEPTUAGINT AND THEOLOGY 323

the variation of first and third person pronouns referring to the Lord in this verse are evidence of a translator who was not highly skilled.19 e. Changes Resulting from Linguistics or Contextual Factors Tov (2015a: 49–50, 2020b: 30) distinguishes between theological renderings or, as he sometimes calls them, theologoumena, and translations motivated by “linguistic” or “contextual” factors. His discussion of the renderings of Gen. 3:17 and 8:21 is an example of this distinction. In 3:17 the phrase ‫“( ארורה האדמה בעבורך‬cursed be the ground because of you”) is rendered ἐπικατάρατος ἡ γῆ ἐν τοῖς ἔργοις σου (“cursed is the earth in your labors”; NETS). Similarly, in 8:21 the phrase ‫“( לא אסף לקלל עוד את האדמה בעבור האדם‬I will never again curse the ground because of mankind”) is rendered Οὐ προσθήσω ἔτι τοῦ καταράσασθαι τὴν γῆν διὰ τὰ ἔργα τῶν ἀνθρώπων (“I will not proceed hereafter to curse the earth because of the deeds of humans”; NETS). In both of these translations, the deeds of mankind become the cause of God’s judgment rather than simply “Adam” or “mankind,” as in the Hebrew. Harl (1986: 139) suggested there was a theological viewpoint in LXX Gen. 8:21, and that the translation indicates that “God takes not into account the nature of man but the actions of everyone.”20 Tov (2020b: 35–6) instead suggests the best explanation of the verses is textual, and that the discrepancy between the MT and LXX is based on a misinterpretation of the prepositional phrase ‫“( בעבור‬because of”), which occurs for the first time in Scripture in these two verses and was probably not known by the translator, who sometimes recognizes its correct meaning later in the book (12:13, 16; 18:26, 29, 31; 26:24). In Gen. 3:17 and 8:21, the translator perhaps interpreted the final resh as a dalet, and then further adapted the word to fit the context, resulting in something like ‫בעבודתיך‬ (“in [or because of] your works”). Because the problem the translator faces in this verse is linguistic, there is no theological intention on his part, so this is not theological exegesis. Another related application of this principle, employed by some, is that differences that can be traced to necessities of translation to a new language are not theological. 5) Other Evidence of Theology beyond Discrepancy between the Vorlage and Translation Is Possible Other phenomena may provide evidence of Septuagint theology. One of these is the translator’s choice of vocabulary. Ausloos and Lemmelijn (2020: 39–42; see also Ausloos 2017: 243–7) contend that even if at first sight the vocabulary is a faithful translation of the Vorlage in the sense that it “quantitatively and qualitatively” renders the words in an adequate way, the resulting translation may not be what the translator intended and can result in a different theology in the translation. They illustrate this possibility from Gen. 1:2, where the Hebrew ‫ ורוח אלהים‬is rendered καὶ πνεῦμα θεοῦ (“and the Spirit of God”). They suggest the translator may not have understood the meaning of the Hebrew in this context describing the initial chaos of creation, and that the Hebrew actually means something like “and a wind from God” or “and a mighty wind,” which involves translating ‫ אלהים‬in the superlative sense (i.e., “mighty”). This understanding fits well with the

One might counter this argument with evidence of the same patterns of translation elsewhere in the translation unit. In that regard, note the translator’s manipulation/change of Hebrew radicals above in 9:11-12 and the change of persons in LXX Amos 5:1; see also the discussion of this passage in Glenny (forthcoming) and Glenny (2018). 20 The original French reads, “Dieu tient compte non pas de la nature de l’homme mais des actes de chacun.” 19

324

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Hebrew context describing the earth as “without form and void” and in total “darkness,” and it is possible that the translator read later theological developments into the Hebrew words and his rendering became the accepted understanding of them. However, although such an explanation of the text is possible, it is difficult to validate Septuagint theology that is based on an adequate and faithful rendering of the Vorlage, and in this case counter examples could be provided against their proposed understanding of the meaning of the Hebrew. Joosten (2000: 42–6) suggests that evidence of the translator’s theology can also be found in more formal elements of the translation. For example, he contends that the translator’s view of Scripture is revealed in things like the degree of literalism and harmonization of different passages in it. He suggests that by the literal translation of parts of the Twelve and other portions of the biblical texts, especially the sometimes incomprehensible rendering of difficult passages in the Vorlage, the translators signal that they hold the biblical texts to be the divine verbal revelation that must be transmitted as faithfully as possible (e.g., Hos. 8:1). Furthermore, the impact of some passages on the translation of other more or less parallel ones and the influence of the Pentateuch on later translations indicates that the translators understood the divine word to be unified and that its passages could be harmonized (e.g., the influence of Exod. 3:14 on Jer. 1:6). 6) Theological Exegesis Is Not Necessarily Tied to Literal or Free Translation Tov (2020b: 27) suggests theological exegesis in the Septuagint is found primarily in freer translation units, like Proverbs and Isaiah. In contrast, others, like Rösel (2018a: 282), are more open to theological content and expect to find theological renderings in more literal translation units as well as in freer ones. Rösel supports his view with studies in Ecclesiastes and Ruth by Aitken (2005) and Ziegert (2008) that have shown how translations that correspond closely to their Vorlagen can contain theologically significant material. We can conclude that theological content and exegesis are found throughout the Septuagint, regardless of translation technique, although not in the same proportion in all books. 7) Patterns or Repetition of Theological Themes in the Discrepancies with the Vorlage Suggest Septuagint Theology If there are a few discrepancies in a literal translation and several of them are related to the same subject, it is more likely those discrepancies reflect the translator’s theology than if the discrepancies concern different topics and issues. In the free translation of Proverbs, Johann Cook (2010b: 630–5) has shown that there is a pattern in the discrepancies in the Septuagint, indicating that the translator has a distinct theology. However, such patterns or repetitions are not to be expected or necessarily required to discern theology in the translation. The work of the translator is by nature ad hoc, working with the text and issues that confront him word-by-word and verse-by-verse. 8) Consider Whether Theological Exegesis Can Be Systematized across Translation Units Although study of the theology of the Septuagint must begin with the analysis of the theology in individual translation units, and normally this is ad hoc and does not reflect any overall systematic approach, it is helpful to consider if the theology in such units and even beyond such units in larger corpora can be systematized to any degree. For example, Cook (2010b: 630–5) has shown convincingly that it is possible to construct a theology of the book of Proverbs, arguing that “the person(s) responsible for this translated unit

THE SEPTUAGINT AND THEOLOGY 325

had a fairly systematic approach toward specific religious matters,” including “the law of Moses.” He also finds evidence of “anti-Hellenistic inclinations,” and determines the translator understood the strange woman as “a metaphor for strange, foreign wisdom” (634). Thus, he concludes that the translator of Proverbs had “a systematic approach towards his parent text” (635). However, he contends that “the more faithfully rendered versions, such as Ecclesiastes, will hardly qualify as candidates for a full-fledged theology” (636). One might also go beyond the consideration of the theology of individual units by using Rösel’s double comparison method (2018a: 255, 264–5; see also Joosten 2000: 33), which involves comparing the Vorlage and translation in a unit to seek to identify theology in that unit and then comparing any such theology of that unit with a related unit. For example, after studying the theological content of Exodus, one could compare it with the theological content of another book in the Pentateuch, like Numbers (see Rösel 2018d). Also, Joosten (2000: 38–40) suggests the study and tracing of theological themes through a translation unit or a larger body of literature. As mentioned above, Joosten (2000: 42–6) suggests the theology concerning Scripture in a translation unit might be determined by looking at the formal aspects of that translation, things like the translator’s attitude toward the Vorlage and the use of parallel passages in the translation. Finally, we need to remember, as Ausloos and Lemmelijn caution (2020: 43), “it is practically impossible to analyze specific theological concepts within ‘the’ Septuagint with a view to provide systematic answers to general questions like: What does the Septuagint state on this or that subject?” However, Rösel (2018a: 280–1) has demonstrated that there are some themes and theological ideologies that can be traced through the Septuagint, whether by the systematization of “terms and ideas” or the systematization of “translation equivalents.” He mentions for example the rendering of ‫“( מזבח‬altar”), which the translators rendered with two different words (θυσιαστήριον and βωμός) to distinguish between Israelite and pagan altars. He mentions also rendering Yahweh (‫ )יהוה‬with κύριος, rendering “Sheol” (‫ )שאול‬with “Hades” (ᾅδης), denoting pagan gods using “idol” (εἴδωλον), and calque translations like νόμος for the Hebrew “Torah” (‫ )תורה‬and διαθήκη for the Hebrew “covenant” (‫ )ברית‬.21

ONGOING RESEARCH QUESTIONS There are many questions and disagreements about the method one should use to determine the theology of the Septuagint. The most basic question is whether one should read the Septuagint from a theological approach, or should one read it from a different perspective, most significantly a textual approach? This question relates to the burden of proof and degree of evidence that is necessary to determine whether a different Vorlage is the source of a unique Septuagint rendering and the priority one gives to different possible explanations for potential unique theological renderings in the Septuagint. The issue of theological method in the Septuagint is also related to basic questions of Septuagint origins: Did the translators understand what they were writing to be Scripture, theology, canonical, or simply important Jewish writings? Did they understand it to be a “standalone” document in its own right, or did they intend it to lead the reader to a better understanding of its source? Another question is whether agreements between the See Ausloos and Lemmelijn (2020) and Rösel (2018c and 2018d) for examples of how one might organize a book on Septuagint theology.

21

326

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Vorlage and the Septuagint translation should count as contributions to the theology in the Septuagint and, if they should, then to what degree? A vital issue is the importance of the theological intent of the translator. Must one demonstrate that the translator intended to make a theological statement and that such theological intentions motivated his translation in order for a potentially theological statement to count as a contribution toward Septuagint theology? And related to that question, are theological and linguistic intentions mutually exclusive? If a translator makes changes in a text because of linguistic or contextual difficulties, is his change not also influenced by his community’s beliefs, as well as his own ideology and worldview? How do we neatly separate the two? Can any translator operate in a theological vacuum? Does evidence that a unique theological reading in the Septuagint is the result of a different Vorlage mean that it has no importance for the theology of the Septuagint? If the reading is part of the tradition and existed at the time of the translation of the Septuagint (and was read in the Septuagint), does it not have importance for understanding the theology of the translator and his community? Should the LXX translators be understood as lone creative personalities, or were the translators conscious of each other’s work (see Tov 2020b: 25 and Rösel 2018d)? There is no shortage of issues and questions concerning the possibility, definition, and methodology one should employ in studying the theology of the Septuagint. Scholars disagree to some extent on almost all of them. Some interpreters are minimalists, and some are maximalists; some use a theological approach to the Septuagint, and others use a textual approach. Hopefully there will be continued progress on all fronts. But in order to do that there must be continued study and discussion of the theological relevance of specific texts and translation units, and sustained effort to clarify and refine the methodology employed. Because of its inherent relationship to the field of Septuagint research in general, progress in the study of the theology of the Septuagint must coincide with progress in the study of other areas of the discipline.

RECOMMENDED RESOURCES 1. Ausloos (2017). Ausloos summarizes some of the most important challenges of doing theology in the Septuagint. More than half the article addresses the question: How does one detect the theology of the Septuagint? He encapsulates his views with seven valuable concluding points. 2. Ausloos and Lemmelijn (2020). This book contains an introduction to the topic of Septuagint Theology and an attempt to fashion one. The up-to-date introduction addresses the question: Is there such a thing as “the theology of the Septuagint?” And then the theology itself is divided into nine broad topics (God, Law, Cult, Prophecy, Man before God, Wisdom, People and Covenant, and Promise), which are each introduced and then discussed according to sections and books of the Septuagint. 3. Cook (2010b). Cook discusses the definition of “theology” when speaking about Septuagint theology, a paradigm suitable for interpreting the Septuagint, the relationship of the “free” translation technique in Proverbs to the theology of the book, and a summary of the theology of LXX Proverbs. He concludes that we are still at the stage of writing theologies of individual books, rather than the whole Septuagint. 4. Douglas (2012). Douglas warns that our lack of knowledge of the translator’s canon limits our understanding of what he believed, and our lack of understanding concerning his intentions, which is caused by our lack of knowledge of the Semitic Vorlage and the mind

THE SEPTUAGINT AND THEOLOGY 327

5.

6.

7.

8.

9.

and world of the translator, limits our ability to know what caused differences between the Vorlage and translation. Glenny (2016). This article addresses the question: How does the Septuagint relate to the Christian Old Testament, and more specifically, what role does the Septuagint play in Christian biblical theology? Following an overview of five approaches to the role of the Septuagint in a Christian whole-Bible biblical theology, four suggestions are given for the importance of the Septuagint in Christian biblical theology. McLay (2010). McLay argues that because the Septuagint was considered Scripture, like the Hebrew Bible or New Testament, its theology should be determined in the same way as these other Scriptures. Thus, Septuagint theology includes agreements between it and its Vorlage as well as disagreements, translator’s theological intent is not required for theology, and Septuagint theology is not limited to the Old Greek text. Rösel (2018a). After a brief summary of the history and value of Septuagint theology, Rösel develops some of the differences in theology between the Hebrew and Jewish Greek Scriptures, demonstrating that “the translators had their own theological and hermeneutical ideas” (264). Then he suggests a four-step preliminary process for completing a theology of the Septuagint. Rösel (2018b). This paper is a response to some critiques of Rösel’s proposal in Cook (2010b) for writing a theology of the Septuagint. First, he summarizes a suggested fourstep approach to writing a theology; next he defines further what he means by implicit theology in the Septuagint. Then he addresses important issues and areas of disagreement for doing Septuagint theology. Finally, he stresses the importance of the historical context of the Septuagint text employed and different perspectives possible for interpreting the Septuagint. Tov (2020b). This article was written as a response to Rösel’s presentation at the 2018 Stellenbosch Congress on Septuagint theology. In it, Tov especially focuses on the interaction between theological and text-critical approaches to the Septuagint, and he concludes that often decisions to ascribe differences to one approach or another depend on a scholar’s personal inclination.

328

CHAPTER 22

The Septuagint and Modern Language Translations WILLIAM A. ROSS

INTRODUCTION A basic metric for the growing interest in study of the Septuagint over the last half century is the increasing number of projects to translate the corpus into modern languages. Central to this disciplinary growth overall is the International Organization for Septuagint and Cognate Studies (IOSCS), which for more than fifty years has been a venue for much of the scholarship related to the Greek version of the Old Testament. Shortly after it was founded in 1968, scholars in the IOSCS began discussing the need for a fresh translation of the Septuagint into English (Fritsch 1970a; similarly also Filson 1972). At the time, only two English translations existed (Thomson 1808; Brenton 1844), but both were well out of date in terms of their English language and textual basis. Despite this clear need within biblical scholarship, it would be decades before serious progress occurred not only in English, but in any language. In time, however, translation projects began to proliferate. First to the grindstone was a group of French scholars who in the mid-1980s began a translation project known as La Bible d’Alexandrie (BdA). Around this same time an Italian project also began (Cacciari and Tampellini 1998), now available in three volumes (Lucca et al. 2012–16; cf. Brunello 1960, Mortari 1999). Then, about a decade later, Pietersma (1997) cast his vision for a project to be known as the New English Translation of the Septuagint (NETS), and one year later German scholars began the Septuaginta Deutsch project (LXX.D). Soon after, in the early 2000s, a Spanish project also began called La Biblia griega. Septuaginta (LBGS). Most recently to the press is the Lexham English Septuagint (Penner and Brannan et al. 2019). Still other modern language translations have appeared or are still under way. Aside from reports of projects in Japanese, Modern Greek, Hungarian, Serbian, and Russian, published translations of the Septuagint in whole or in part have emerged over the last decade in Romanian (Anania 2001; Bădiliţă et al. 2004–8), Modern Hebrew (Zipor 2005), Polish (Popowski 2013), and Portuguese (Lourenço 2017, 2018, 2019). There have also been recent translations of the Septuagint into English as part of a larger Bible translation, including the Orthodox Study Bible (Maximos et al. 2008) and the translation produced by King (2013).1 For a helpful discussion of King’s work, see Scanlin (2016). Also worthy of mention—but also beyond the scope of this chapter—are the English translations of individual books produced by contributors to the Brill Septuagint Commentary Series (SEPT), on which see Porter in the present volume.

1

330

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Although each of these projects could be discussed at length, the English, French, German, and Spanish projects have been selected for analysis here. Aside from representing the most pervasive modern languages used within the academy of biblical studies, these projects in particular have occurred in constant conversation with Septuagint scholarship. Moreover, the approach taken in each of these translation projects was developed and applied by teams of scholars aiming at methodological coherence and consistency. Yet, when set in comparison, each manifests a distinct perspective upon the nature of the Septuagint itself—often deliberately in contrast with the other projects—particularly dealing with its origin, purpose, and language. The differences among the methods and results of each project thus illustrate broader differences and disagreements within the discipline of Septuagint studies as a whole.

DEBATES INVOLVED WITH TRANSLATING A TRANSLATION Translation is a multifaceted process that is difficult to characterize without diminishing. This reality is in part responsible for the divergence among methods in the modern language translation projects of the Septuagint discussed here. However, the difficulty of describing a translation is all the more complex when the base text is itself a translation that was produced by numerous translators. As Ottley (1920: 172) observed a century ago, the problems involved with translating the Septuagint “are not entirely simple, nor quite like those of translation in general; and something can be said on behalf of more than one method.” As a source text, the Septuagint presents complicating factors insofar as its origins, purpose, and even its linguistic character remain disputed (see Wright 2011; Kreuzer 2015a). Given that the Greek translations of the Hebrew scriptures came about at the hands of many people in many places over many years, whether and how “the” Septuagint may be treated as a coherent corpus in either language or scope is debatable. Current Septuagint scholarship enjoys only limited consensus on any of these aspects of its object of study. Nevertheless, Septuagint scholars certainly agree that “translating a translation is not simply translation” (Fernández Marcos 2008: 288; cf. Cook 2006b: 32). To deal with the complexities involved in the task, each of the modern language Septuagint translation projects discussed here has a unifying method for the activity of diverse translators. To describe the differences among these projects, it is useful to situate the modern translations along a spectrum to create a rubric for subsequent discussion in more detail (Figure 22.1). Of course, any visual representation of a translation will fail to do it proper justice, and the comments here about contrast in method are for illustrative purposes only. However,

FIGURE 22.1  Understanding of Relationship of Target Texts on Source Texts in Modern Language Translations of the Septuagint.

THE SEPTUAGINT AND MODERN LANGUAGE TRANSLATIONS 331

it is fair to say that the primary issue underlying the differences among modern language translations of the Septuagint is how they conceptualize and explain the relationship between the Semitic source text and the Greek target text (Kraus 2006: 64; Harl 1992). That basic relationship can be situated along a spectrum in terms of the degree to which the Septuagint is regarded as a text independent from its Semitic source text and how that independence is understood. Or, stated conversely, in terms of the degree to which one prioritizes the status of the Septuagint as a derivative text and the ways in which that derivation is understood. Supposing a higher degree of dependence leads to a modern language translation of the Septuagint that permits deferring to the Hebrew source text in certain situations to understand and translate the Greek. Alternatively, supposing a higher degree of independence leads to a modern language translation of the Septuagint that tends to disregard the Hebrew source text in its understanding and translation of the Greek. In reality, however, things are not that simple, and scholarly opinions fall at different points along this spectrum owing in large measure to the peculiarities of the language of the Septuagint qua Greek. In the century since the discovery of documentary papyri and inscriptions, the notion of a Jewish Greek dialect—which was supposed to account for these peculiarities—has been thoroughly disproven (see Horsley 1989; Lee 2016). This erroneous theory arose for a variety of reasons, not least of all owing to insufficient understanding of the post-classical (Koine) phase of Greek in general, as well as a prevailing prescriptivist linguistic attitude that expected conformity with classical varieties of the language. Scholarship is still laboring to improve the former and combat the latter. Recent work has turned for its understanding of the linguistic peculiarities of the Septuagint to analysis of translation technique and the social context in which it arose.2 For the present topic, understanding how the ancient translator approached his task, the reasoning goes, can and should inform how the modern translator approaches hers. But in practice it is not so easy. Although there are notable exceptions, ordinarily the ancient Septuagint translators followed a working method that prioritized close adherence to the word order of their Hebrew source text. Scholars agree on this point, yet widely disagree on its significance, specifically regarding whether and how that method entailed producing a translation that departed from linguistic conventions in post-classical Greek. For example, frequent parataxis, reproduction of Hebrew idioms, omission of the copula, transcriptions, specialized religious vocabulary, compound verbs, and standardized equivalents all characterize the Septuagint translation (Aitken 2014c: 121–3). Yet caution is necessary in passing judgment on whether such features were conventional in the language. In many cases (but not all), when compared with contemporary post-classical usage, the peculiar qualities of the language of the Septuagint subsist in degree rather than kind. That is, linguistic features that might be considered peculiar in some way only seem so because they distinguish the Septuagint in terms of their frequency rather than their novelty or comprehensibility per se within the language. Additionally, it is often difficult to know whether linguistic features of the Septuagint that seem unusual to us and are poorly attested—or totally unattested—in contemporary sources would have been considered conventional in the language and, if so, by whom.

On which see Dhont in the present volume.

2

332

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

So despite the general agreement upon the value of analyzing translation technique and social context, complications like those described above have given rise to disagreement over how such analysis should contribute to translating the Septuagint into modern languages. Are the peculiar phenomena in the language of the Septuagint taken to be the exception or the rule? In the absence of compositional Greek sources from the same period attesting the same linguistic features, should we assume that they would have been conventional or unconventional to native speakers when the translation was produced? Should we presume that the ancient translators were striving to represent the meaning of the source text at every level of language, or only some levels? How one answers such questions will significantly inform one's approach to translating a translation. On the one hand, seeing a high degree of dependence of the Septuagint upon its Semitic source text tends to correspond with greater willingness to categorize apparent linguistic peculiarities as unconventional Greek and to translate them as such. From this perspective, the ancient translators are thought to have sacrificed comprehensibility or style in service of preserving some more important features of their source text, and therefore in such cases either the Hebrew may help disambiguate or the stylistic peculiarity may be brought into the modern translation of the Septuagint. On the other hand, seeing a high degree of independence of the Septuagint from its Hebrew source text tends to correspond with greater willingness to grant that apparent linguistic peculiarities represent shortcomings in current scholarly understanding of or gaps in the evidence for conventional Greek and to translate it as such, often using the broader context as a guide. From this perspective, the ancient translators are thought to have employed comprehensible language that neither presumes nor requires knowledge of the source text in order to be understood as Greek, and therefore the Hebrew need not assume any particular role in a modern translation of the Septuagint. The difference between these two ends of the spectrum is largely one of perspective: Is the language of the Septuagint Hebrew in substance with Greek characteristics, or is it Greek in substance with Hebrew characteristics? Did the translators use Greek to encode the religious traditions that were in their source texts, or did they use the source texts to communicate their religious traditions in Greek? How does one answer such questions? Few would respond to such questions simplistically, and for that reason they provide many avenues for refinement and further research. But how they are answered informs whether the principles of a modern language translation are oriented more toward the process of the ancient translation of the Septuagint or rather the product. The following section addresses these issues by describing each of the modern language translation projects in terms of Figure 22.1. A brief description of the scope, content, layout, and intended audience is also provided, along with questions for future research.

COMPARING MODERN TRANSLATION METHODS In order to demonstrate clearly the differences in translation methods among the projects, and given their more polarized approaches, NETS and BdA are considered first and in more detail.

The New English Translation of the Septuagint The New English Translation of the Septuagint (NETS) was published over a century after the next most recent English translation of the Septuagint appeared (Pietersma and Wright 2007). This publication was preceded by a standalone translation of the Psalms

THE SEPTUAGINT AND MODERN LANGUAGE TRANSLATIONS 333

in 2000, for which Pietersma originally developed the NETS translation procedures and statement of principles. A second printing of NETS (Pietersma and Wright 2009) appeared soon after that included corrections and emendations, and more changes have since been recorded by the translation committee in 2014.3 NETS is printed in two columns in a single hardback volume, with sparse and brief notes on translation issues. It provides translations of all the books included in Rahlfs (1935) except for Odes—although Odes 12 (Prayer of Manasseh) is appended to the Psalter—and with the addition of the AlphaText of Esther. Brief general introductions and limited bibliography accompany the translation of each book. As a translation, NETS is meant to target readership that is “biblically well-educated” (Pietersma and Wright 2009: xiv).4 The matter of the base text used for NETS is not straightforward, but the approach reflects in important ways the stated purpose of NETS as well as its translation principles. For their Greek text, translators worked from the eclectic text of the Göttingen edition of the Septuagint (1931–). In cases where a given book of the Septuagint was not yet available in that series, such as Ecclesiastes or Judges, translators employed the edition of Rahlfs and Hanhart (2006).5 It is important to note, however, that NETS translators did not, in fact, produce English translations of the Greek texts de novo. They instead used the 1989 text of the New Revised Standard Version as their point of departure for revision against the Greek. The decision to use an English translation of the Hebrew Bible as both necessary and sufficient for an English translation of the Greek Septuagint was, in part, practically motivated. Although many readers may be unaware of this purpose, from an early stage NETS was intended to function as a synoptic tool for “comparative study of the Hebrew and Greek texts, albeit in English translation.” The NRSV thus commended itself for such a task, given its widespread popularity among Bible readers, as well as its own translation method that was itself “reasonably well suited to NETS’ purposes” (Pietersma and Wright 2009: xv).6 The choice to revise the NRSV to produce NETS was deemed all the more appropriate owing to the underlying theory about the original nature and purpose of the Septuagint translation itself. Simply put, “the paradigm basic to NETS is that of the Septuagint as an interlinear text” (Pietersma and Wright 2009: xviii). That is, NETS maintains that the Septuagint as it was produced had a “dependent and subservient linguistic relationship to its Semitic parent” that “can be best conceptualized as a Greek interlinear translation of a Hebrew original within a Hebrew-Greek diglot.” Evidence for this linguistic relationship is found in the “strict, often rigid, quantitative equivalence [of the translation] to the Hebrew,” an approach that in fact produced not Greek but “translationese” that conveys something other than conventional linguistic meaning (Pietersma and Wright 2009: xiv; cf. Boyd-Taylor 2004: 55; Boyd-Taylor 2011: 40). The resulting linguistic “unintelligibility”—purported to be one of the “inherent characteristics” of the translation—bespeaks a Hebrew-oriented pedagogical function (Pietersma and Wright 2009: xv; cf. Boyd-Taylor 2008a). The most recent changes are available on the NETS website (http://ccat.sas.upenn.edu/nets/edition/errata-2014. pdf ) and will appear in a forthcoming second edition. 4 Note that a related commentary series is in production, on which see Büchner (2017) and Hiebert in the present volume. 5 Translators also had license to “improve upon” their Greek base text, changes that are included in the textual notes of NETS (Pietersma and Wright 2009: xix). 6 When a book of the Septuagint lacks a Semitic source text for comparison, the NRSV served “only optionally” as the base text for NETS translators (Pietersma and Wright 2009: xix). 3

334

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

From the NETS perspective, then, the Septuagint was in fact intentionally produced as “a Greek translation which aimed at bringing the Greek reader to the Hebrew original rather than bringing the Hebrew original to the Greek reader” (Pietersma and Wright 2009: xiv). This function is taken to underlie or otherwise characterize the production of the Septuagint and to constitute the fundamental impetus for its “linguistic strangeness” as Greek (Pietersma and Wright 2009: xv). Any literary features in the Greek are thus considered “the exception that proves the rule,” since the translators “often used unidiomatic Greek.” Accordingly, in the production of NETS, the nature of the language of the Septuagint was considered such that, not only would an English translation with any literary elegance have required “a different Greek” entirely, but even “a fully idiomatic translation into English [could] scarcely be justified” (Pietersma and Wright 2009: xviii). Some corollary discussion of the interlinear paradigm will help clarify some of the logic underlying the translation principles discussed below, specifically pertaining to Greek semantics. Pietersma (2002: 342–3) argues that the “unidiomatic” aspect of the language of the Septuagint is not always a consequence of its vocabulary or grammar per se, which are acknowledged to represent conventional post-classical Greek often or even usually in some respect (Pietersma 2001: 225). Rather, the unintelligibility is a result of the “literal, formal-correspondence type of translation,” which led to the “excessive use” of otherwise conventional features (Pietersma 2002: 342–3, cf. Boyd-Taylor 2004: 151).7 Whatever degree of unintelligibility is present in its language, then, is the product of positive transfer from the source text whereby the translator, in keeping with his word-for-word approach (often called isomorphism), was induced to employ conventional linguistic features with unconventional frequency. Boyd-Taylor (2011, 2013), who played a seminal role in developing the interlinear paradigm, explains that such transfer appears most clearly “in the company a word keeps. Within a translation, a word will often occur without its typical collocates, and be found instead with dubious companions in unfamiliar haunts. Its distribution will prove to be unconventional; the word occurs either more or less frequently than expected” (2004: 57). For Boyd-Taylor, the positive interference from such transfer is considered so pervasive and significant that the text of the Septuagint actually presents, not Greek, but an “interlanguage.” Despite such claims, advocates of this paradigm simultaneously argue that the Septuagint interlanguage does not everywhere require Hebrew to understand, nor does interlinearity entail that no refined or sophisticated usage appears in the Septuagint (Pietersma 2002: 350–1, 357). Nevertheless, the NETS translation method does not consider the Greek or its own linguistic context in or outside the Septuagint to be a reliable arbiter of meaning in every instance, and thus licenses using the Semitic source text to adjudicate where necessary (Wright 2008: 107).8 It is not necessary to review the NETS translation principles in full here, but some salient points should be described in light of the above (see Pietersma 1996; cf. Pietersma and Wright 2009: xvii–xviii).9 Most notably, translators had step-by-step instructions for how to determine word meaning using the following diagram of a semantic “spectrum” (Figure 22.2).

Pietersma (2002: 340) originally affirmed that the paradigm specifically pertains to “the birth of the Septuagint,” that is, its original purpose. In later years, however, interlinearity was promoted as a “heuristic tool” rather than a theory of origins (Pietersma 2010; cf. Wright 2008: 107–10). On Septuagint origins, see Aitken in the present volume. 8 Of course, the interlinear paradigm has not gone uncritiqued (e.g., Mulroney 2016: 51–78). 9 Available online: http://homes.chass.utoronto.ca/~pietersm/manual.html (accessed January 20, 2020). 7

THE SEPTUAGINT AND MODERN LANGUAGE TRANSLATIONS 335

FIGURE 22.2  The NETS Semantic “Spectrum”

A given Greek word is situated on the spectrum according to the NETS translator’s quantification of how frequently that word corresponds to a single Hebrew word. The translator thus formulaically ascertains whether occurrences of “Hebrew X are equated with Greek Y” in all (contextual rendering), most (stereotype), some (calque), or few (isolate rendering) instances. In practice, however, this is actually closer to a binary system, since words that fall on the left of the vertical line in Figure 22.2 are assigned their “normal” Greek meanings while those that fall on the right “are effectively assigned the meaning of their Hebrew counterparts” (Pietersma 1996: 14). Although they were not in fact rigidly applied in the production of NETS, in principle these guidelines were considered sufficient to arbitrate the meaning of the text as the ancient translator himself understood it at the point of its production (Wright 2008: 106; see also Hiebert 2004: esp. 74–5).

La Bible d’Alexandrie The translation project known as La Bible d’Alexandrie (BdA) saw its first publication in 1986 through the efforts of a group of scholars led by Marguerite Harl, Dominique Barthélemy, Olivier Munnich, and Gilles Dorival. Although it has been over thirty years since the flagship volume appeared, the series still remains incomplete.10 However, BdA is not only a translation. Issued in paperback fascicles, each volume also includes an extended introduction to each book, along with generous notes throughout the text in an apparatus. These notes cover philological and linguistic matters in the Greek text and its French translation, significant divergences between the Hebrew and Greek texts, and interpretive notes with observations upon the reception of the text in Philo, the New Testament, and other early Christian literature (Harl 2001: 182; Dorival 2008: 70). While the approach to translating the Septuagint has remained largely consistent in BdA over the years, external factors have changed significantly with the growth of the discipline of Septuagint studies in general. These factors include new introductions to the field, specialized monographs, studies and reference works for Septuagint grammar, syntax, and lexicon, but most importantly, new textual editions. When the project began, not only was the Göttingen edition incomplete (as it still remains), but Hanhart’s revised version of the edition by Rahlfs (1935) would not appear for another two decades. Thus, Rahlfs’s edition provides the textual basis of BdA, with Göttingen consulted where available, and as the discipline has grown each installment in the series has taken new developments into account (Dorival 2008: 71–5). Thus far, each volume of BdA contains only one book of the Septuagint, except for the Twelve Prophets—of which ten are now translated in four volumes—and the publication of Baruch, Lamentations, and the Epistle of Jeremiah together.

For full bibliography, see Jones in this volume. On the prehistory of BdA, see Harl (2004: 151–84).

10

336

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

Methodologically speaking, BdA falls on the opposite end of the spectrum from NETS. Its distinction in this regard becomes apparent even in the scholarly profile of the original project collaborators, who were primarily specialists in the study of Greek, along with some experts in Jewish-Hellenistic and Patristic literature (Dorival 2008: 75–6). Perhaps unsurprisingly, then, BdA has always aimed to “give a precise and complete translation of the Greek text as it is” (Dorival 2008: 68). The translators who contribute to the series have agreed to regard the Septuagint an “œuvre autonome, détachée de son modèle” (“autonomous work, detached from its [Semitic] model”), and thus believe it is best understood as a Hellenistic “œuvre littéraire au sens plein du terme” (“literary work in the full sense of the term”) (Harl 1992: 36). Put differently, the goal of BdA is to render the Septuagint into French strictly “according to the Greek” (Harl 2001: 182–3), producing a translation that is “as literary as possible on the basis of syntactical and lexical usages of the Greek language current at the translators’ epoch.” This first step in the BdA translation method totally disregards the Hebrew source text (Harl 2001: 183). The meaning of Septuagint vocabulary is sought primarily in the immediate context, along with any other text in which a given word appears, “without projecting any supposed Hebrew signification onto it” (Dorival 2008: 69). Similarly, Harl (2001: 186) explains that “the criterion for determining the words in the LXX is not the meaning of their counterparts in the Hebrew. It is their meaning in the Koinè [sic], or more precisely, the sense they acquire in the context of the LXX sentences, according to the use the translators make of them, following their choices and habits” (cf. Harl 1992: 37). At this stage in their work, BdA translators closely follow Septuagint syntax and attend to its textual divisions and stylistic devices. Linguistic phenomena that appear awkward or incorrect—at least, from the perspective of classical Greek conventions—are rendered into French as accurately as possible, without respect to whether they arose from Hebrew influence or simply reflect new (and perhaps poorly documented or under-studied) developments in post-classical Greek. In this respect, BdA does not shy away from acknowledging what might be “un-greek usages” (Harl 2001: 186) or identifying “the strangenesses [sic] and the oddities of the Greek language” employed by the Septuagint translators (Dorival 2008: 70). But in order to evaluate such phenomena, the BdA approach to translating the Septuagint “according to the Greek” entails referring to other Hellenistic literature and the Ptolemaic documentary papyri to help understand and translate the text. Even in cases where such features do prove to have been unconventional within contemporary post-classical Greek usage, they are interpreted in the context in which they appear and, in cases that recur frequently, are assumed to have become (more) comprehensible to ancient readers as they repeatedly encountered them in the Septuagint (Harl 2001: 186). BdA thus aims as far as possible to render the Septuagint text into French as an ancient Greek reader would have understood it, whether by linguistic convention or contextual inference (Harl 1992: 39–40). It is important to recognize, however, that the BdA translation method does not totally ignore the Hebrew. But the Hebrew is only consulted after the initial French translation according to the Greek. Still, the aim in consulting the Hebrew is primarily to understand the divergences, not to allow the Hebrew to adjudicate ambiguities in the Greek. The operative assumption here is that Septuagint translators, while not totally innocent of genuine mistakes, were ordinarily “competent and conscientious” in their work. They proceeded in their typical word-for-word approach while simultaneously remaining “concerned with logic and clarity as well as with expression and stylistic device.” Accordingly, divergences between the Greek and Hebrew texts are of interest

THE SEPTUAGINT AND MODERN LANGUAGE TRANSLATIONS 337

for understanding the Greek text itself since they often reflect “a certain degree of liberty” and deliberate linguistic creativity, and may even indicate interpretive tendencies (Harl 2001: 187, 189). In this connection, by prioritizing understanding and translating the language of the Septuagint in its Greek textual and linguistic context, BdA remains alive to numerous variables pertaining to the source text: the reality that the Masoretic text is not identical to the Septuagint Vorlage(n); the ambiguities that would be inevitable in such an unpointed text; and the grammatical flexibility and obscurities in Hebrew in general. Indeed, for BdA it is precisely the number of “incongruities” between the Greek and Hebrew texts that “show clearly that a [modern] translation of the LXX wishing to present the meaning of the Greek faithfully cannot use MT as its phraseological and lexical foundation” (Harl 2001: 193). The third step of the BdA translation process exists to help resolve remaining difficulties  in the Greek text. At this point, BdA translators examine the reception of the Septuagint by ancient Jewish and Christian commentators. The rationale behind this approach is that, while ancient readers were not always correct in their understanding, they were for the most part native speakers of post-classical Greek, if usually a later stage of the language. Moreover, ancient readers were also less likely to project the meaning of the Hebrew text onto the Greek text owing to their general lack of knowledge of the Hebrew language (Dorival 2008: 69). The aim for BdA is thus to give interpretive priority to the ancient understanding of the Greek target text, rather than to a modern understanding of a supposed Hebrew source text (see Harl 1999). Finally, it is worth pointing out that, while BdA has four broad stages in its translation method, translators do not have a prescribed approach as detailed as that of NETS and also differ slightly in their emphases.11 Even as the number of contributors has grown, with diversified specialties and opinions, Dorival (2008: 78) clarifies that BdA does not “force one position” on any given issue, giving rise to certain perspectival “differences between the volumes” in the series. Along these lines, Kraus (2006: 68–9) has suggested that more recent volumes of BdA seem to pay more attention to the Hebrew source text than the earlier ones.

The Lexham English Septuagint The recently published Lexham English Septuagint (LES) has a unique history that is not directly connected with the IOSCS and its intramural debates over Septuagint translation. But LES represents a significant development for the discipline in that it provides scholars with another English translation option, yet one produced using principles very similar to those of BdA. As such, this project, brought together by Penner, Brannan et al. (2019), certainly merits attention. Perhaps ironically, LES originated as an “interlinear edition of the Lexham GreekEnglish Interlinear Septuagint,” itself a digital product produced by a team of translators and available only on Logos Bible Software. Brannan then took “material” from that interlinear to use “as the initial basis for a translation,” writing a program “to reassemble, as much as possible, the interlinear lines into English word order.” It was this digitally produced English text that was then edited by a team of scholars who converted it into

The fourth and final stage, after initial translation, textual analysis, and investigation of reception history, is revision for purposes of style in French (Harl 2001: 196–7).

11

338

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

readable English “in consultation with the Greek text.” In certain cases, this procedure “required an almost entirely new translation by the LES translator.” The results then underwent various reviews and edits to produce “a transparent, literal translation of the Septuagint” that was originally available only on the Logos platform. In time, Penner “was invited to reedit the LES” in full in order to produce a printed volume (Penner and Brannan 2019: xii). Behind all this activity, the textual basis of LES was the diplomatic edition produced by Swete (1887–94). Consequently, it differs slightly from NETS in arrangement and presents slightly more content, with LES including all of the Odes as well as Enoch. The English text is printed in a single column, with marginal subheadings and extremely sparse translation footnotes. In contrast to both BdA and NETS, LES contains no further information about its content, aside from one page of discussion of the history of the Septuagint in the volume introduction. The editors state that their desire in producing LES was not to supplant existing English translations of the Septuagint, but rather to provide a resource to enhance insight and understanding of a given passage by comparison (Penner and Brannan 2019: x). As already mentioned, the LES translation principles have important points of overlap with those used in the BdA project. On the one hand, given its origins as an interlinear, LES has a “strong transparent connection to the Greek.” On the other hand, in their revision of the text the editors attempted to “focus on the text as received rather than as produced,” an approach stated in terms clearly meant to contrast with NETS. In this connection, “[e]very effort was made to render the Greek in its own right, with no eye to the Hebrew at all” (Penner and Brannan 2019: xiii). In this sense, it is correct to say that the LES project falls farthest to the right-hand side of Figure 22.1, since every other modern translation consults the Hebrew text under certain circumstances, even if only very selectively. At this point it is important to note that the use of Swete’s edition as a textual basis is connected with the translation principles adopted in LES. The primary witness used in that edition was Vaticanus, a fourth-century Christian codex. Consequently, it was that historical and social context that provided the theoretical framework for understanding and translating the Greek in LES. Penner and Brannan (2019: xiii) state that LES “is an attempt to answer the question, ‘How would this [Greek] text have been read— understood and experienced—by a fourth century, Greek-speaking gentile Christian?’ … Ideally, the translation should be as rough or as smooth as the Greek would have seemed to a Greek reader [in that context] who knew no Hebrew.” Further, they state their guiding principle that “if the Greek is smooth and represents good Greek style, then the English equivalent should convey that style, and that the English should be awkward if the Greek is awkward” (Penner and Brannan 2019: xiv). Some may reasonably wish to understand in more detail the criteria and evidence used for such judgment. Unfortunately, the editors say little about resources used in the process, except to mention the recent Syntax by Muraoka (2016). More could be said about this new English translation, and surely will be elsewhere. One matter for discussion ought to be whether and how LES qualifies as a translation at all. It is worth noting that both NETS and LES—while on opposite sides of the spectrum in Figure 22.1—present English versions of the Septuagint that are in reality revisions of a preexisting English text that in itself would not exactly qualify as a translation of the Greek. In this connection, the LES editors correctly point out that NETS is properly categorized as a “modified NRSV,” rather than as a translation “made directly from the

THE SEPTUAGINT AND MODERN LANGUAGE TRANSLATIONS 339

Greek” (Penner and Brannan 2019: x). Yet it may prove unclear to some whether it is equally correct to claim that LES itself represents a direct translation in that sense. Is it accurate to refer to someone as a translator, as Penner and Brannan (2019: xii) apparently do in regards to their own project, when in fact what they have done is rearrange into more conventional English word order an English text that was computer-generated from an interlinear, while “consulting” the Greek? While it is true that LES is the only modern translation examined here that makes no reference to the Hebrew at all, it may nevertheless be that LES is more properly described as a translation made indirectly from the Greek. Several years ago, Law (2013: 8) pointed out that, while NETS is “indispensable for the scholarly study of the Septuagint” in English, a very different endeavor would be required to appeal to non-specialist readers of the Bible. Initial soundings suggest that LES may indeed have that kind of broader appeal. Regardless, it is certainly the case that, given its position on the spectrum in Figure 22.1, LES deserves to be consulted alongside NETS for scholarly research purposes.

Septuaginta Deutsch The Septuaginta Deutsch (LXX.D) project began in 1999 and was finished after ten years of work that was carried out by a team of scholars at several universities, becoming the first full translation of the Septuagint in German. Produced under the general editorship of the New Testament scholars Wolfgang Kraus and Martin Karrer (2009b), LXX.D was published by the German Bible Society in one volume. The main text is printed in two columns with German subheadings, and each book or group of books (e.g., 1-4 Kingdoms) is prefaced by a brief, one- or two-page introduction. Double-text books like Judges or Tobit are printed with the two versions in the columns (so NETS), rather than on facing pages (so BdA). In most cases, the translation of each book of the Septuagint was produced by two scholars, often one specialist in Hebrew Bible and one in New Testament (Kreuzer 2001: 41). Kreuzer (2004: 107–8) notes that LXX.D translators also made use of a word list to strive for “eine konkordante Übersetzung” (“a concordant translation”). For the textual basis, these translators employed the Göttingen edition wherever available, otherwise using the Rahlfs-Hanhart (2006) edition.12 Concise notes are provided at the bottom of the page pertaining to matters of text and translation for books also available in other critical editions, including Rahlfs (1935) and others (Kraus and Karrer 2009b: xvii–xviii). Two years after the appearance of LXX.D, two accompanying volumes were published containing more detailed book introductions along with more expansive “Erläuterungen und Kommentare” (“explanatory notes and comments”) produced by the translators (Karrer and Kraus 2011a, 2011b). In terms of the delimitations of the project, LXX.D presents the same corpus of books as Rahlfs (1935), including the Psalms of Solomon and Odes. Kreuzer (2001: 43) gives a sense of the general aim for the German translation itself and the target audience, stating that LXX.D is intended to be “philologically reliable, easily readable, and transparent in a scholarly sense.” Elsewhere, Kraus and Karrer (2005: n.p.) state that LXX.D is “intended for readers in the universities and in the German-speaking public. It is written in a German style as near as possible to

But the editions by Fernández Marcos and Busto Sáiz (1989, 1992, 1996) were used in the Historical Books, in which the Antiochene text of Kaige sections in 3–4 Reigns (1-2 Kings) is given separately (Karrer 2008: 108).

12

340

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the Greek and in this way encourages ongoing scientific work on the Septuagint” (cf. Kraus 2006: 79; 2010: 248). Methodologically speaking, LXX.D does not represent a totally distinct approach, but rather it is rightly portrayed—and was consciously undertaken—as something of a middle way between NETS and BdA (Kraus 2006: 67–70). Indeed, Kraus and Karrer (2009a: xix) view the Septuagint translation itself as situated somewhere “between two languages,” and thus aim to carry over all its resulting “Eigentümlichkeiten” (“peculiarities”) into German. In this sense, LXX.D simultaneously upholds perspectives on the Septuagint  that otherwise distinguish the English and French projects. For example, Kreuzer (2001: 43) states that (much like BdA) LXX.D aims to understand the Septuagint “as a document of Hellenistic Judaism in its own right,” which (unlike LES) entails using the “oldest accessible form of the Greek text.” At the same time, the German translation directs much attention to the Semitic source text (much like NETS), even identifying differences by typographical means, such as placing the translation of plusses in square brackets and various divergences in italics (Kraus and Karrer 2009a: xviii, xxi). This mediating approach has led the scholars involved in the LXX.D project to adopt the methodological catchphrase that was coined by Utzschneider (2001): “auf Augenhöhe mit dem Text” (“at eye-level with the [Greek] text”). Perhaps unsurprisingly, the LXX.D approach requires a milder formulation of the opposing perspectives represented in the spectrum of Figure 22.1 in order to hold them together. Kraus (2010: 247) argues that “[o]n the one hand, the Septuagint became an independent work of literature. On the other hand, it cannot be seen in isolation from its Hebrew Vorlage. These aspects are not mutually exclusive; they complement one another … The Septuagint mediates between Hebrew tradition and a Greek-speaking audience.” The German approach bears similarity to the NETS approach insofar as consultation of the source text is viewed as appropriate. Yet LXX.D does not go quite so far as NETS in its conceptualization of the degree of dependence as what is entailed by the interlinear paradigm, nor does the German translation consult the Semitic source text for the same reasons (Kraus 2010: 247). At the same time, LXX.D also bears similarities to BdA and LES in that it sees the Septuagint as a “literary work” that can and indeed should be read independently, since such a reading repays analysis of the narrative, sociohistorical, and theological aspects of the Greek text “on its own” (Kraus 2006: 83). Within the LXX.D approach, however, those aspects of the text render the Septuagint an independent literary work, “not although but rather because” it is a translation of a Hebrew text (Kraus 2006: 71, emphasis original). In other words—and somewhat paradoxically—the dependence of the Septuagint translators on their source text is itself the sine qua non of the independence of their target text. Along these lines, Kraus (2010: 245) states that the Septuagint itself is a “transfer of religious tradition from a cultural context shaped by the Hebrew language into the context of Greekspeaking people living in a Ptolemaic world. How this transfer is achieved varies book to book.” Thus, the method of the LXX.D project intentionally “looks both ways” along the spectrum in Figure 22.1 (Johnson 2016: 29).

La Biblia griega. Septuaginta The recently completed Spanish translation of the Septuagint is less discussed in the related secondary literature despite the fact that one of its editors is well respected for a general introduction to the discipline of Septuagint scholarship (Fernández Marcos 2000b). Despite a rich tradition of biblical scholarship and translation, the Septuagint was in fact

THE SEPTUAGINT AND MODERN LANGUAGE TRANSLATIONS 341

mostly ignored in Spanish scholarship until the twentieth century (Fernández Marcos 2008: 284–5). But that situation has changed drastically in recent decades, as illustrated by the publication of La Biblia griega. Septuaginta (LBGS), a project that was managed by two scholarly research groups based at the Consejo Superior de Investigaciones Científicas in Madrid. LBGS is published in four volumes, completed under the supervision of Fernández Marcos, Spottorno Díaz-Caro, and Cañas Reíllo (2008, 2011, 2013, 2015).13 The volumes are organized by the familiar genre categories of Pentateuch, Historical Books, Poetry and Wisdom, and Prophetical books, respectively. Each volume contains a brief general introduction, as well as lengthier and specific introductions to each book of the Septuagint included, touching on various matters such as origin, language, text, and reception, along with select bibliography. The text of the Spanish translation itself is presented in a single column with generous margins, except double-texts, which are printed in two columns. In terms of its scope, the corpus presented in the four LBGS volumes includes all the books printed in the edition by Rahlfs and Hanhart (2006), including the Psalms of Solomon and Odes, much like LXX.D. The edition of Rahlfs and Hanhart also served as the base text of the translation wherever the relevant volume of the Göttingen series was not yet complete. The exception to that approach was the books of 1-4 Kingdoms, where translators used the critical editions of the Antiochene text, which were produced by some of the same scholars involved in LBGS (Fernández Marcos and Busto Sáiz 1989, 1992, 1996). As indicated in Figure 22.1, the translation method of LBGS falls somewhere between LXX.D on one side and BdA and LES on the other. As Fernández Marcos and Spottorno Díaz-Caro (2008: 25) state, the Spanish translation is intended to provide “una version literal, fiel al original griego, porque únicamente de esta manera se podrán percibir los rasgos específicos y novedosos de la Biblia griega” (“a literal version, faithful to the Greek original, because only in this way will it be possible to perceive the specific and novel features of the Greek Bible”). Notably, the kinds of features in view here are not linguistic per se, but rather are more textual, one might even say literary in nature, such as the well-known differences between the Septuagint Vorlagen from the Masoretic text (e.g., Jeremiah). The attitude of the editors of LBGS toward the language of the Septuagint becomes visible by virtue of their approach to producing the Spanish translation. Fernández Marcos and Spottorno Díaz-Caro (2008: 26) explain that the modern translators strove to produce a readable Spanish text with “un estilo literario bueno en la medida de lo posible … y reproducir en la lengua término lo que etsá formulado en la lengua origen” (“good literary style as far as possible … and to reproduce in the target language what is formulated in the source language”). The LBGS project thus attempts to keep in view the meaning of the Greek text, the variation among the ancient translators’ approaches, and their sociohistorical context. The explicit aim of the Spanish translation was “to assure that the cultivated reader who has no knowledge of the Greek language may have access to the Greek Bible not only in the content but, as far as possible, even in the form and style” (Fernández Marcos 2008: 288). The rationale behind this approach to the modern translation is the position of the LBGS project that the Septuagint is rightly construed as “una obra literaria independiente” (“an independent literary work”), although with certain qualifications (Fernández Marcos The first volume, El Pentateuco, is now in its second edition (Fernández Marcos, Díaz-Caro and Cañas Reíllo 2016).

13

342

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

and Spottorno Díaz-Caro 2008: 26). The Septuagint is considered independent in the sense that it was produced to replace the Hebrew Bible for Greek-speaking Jews and arose from an unpointed text that was to a certain degree open to different readings and interpretations (cf. Fernández Marcos 2001: 235). As such, the Septuagint “must be translated from the Greek text which we have in front of us, not from the Hebrew text that is behind it … In other words, it is the meaning of the Greek text which should be translated in its Jewish-Hellenistic context for a Greek-speaking audience” (Fernández Marcos 2008: 289). The LBGS translators do acknowledge certain linguistic peculiarities in Greek that were occasioned by the Semitic source text, yet they affirm that in most cases these could be “absorbed without difficulty by the context and rendered intelligibly” (Fernández Marcos 2008: 290 n. 14). At the same time, however, the editors affirm that “the Septuagint is a translation and that the Hebrew can be, and indeed should be, consulted as part of the context” in especially difficult passages (Fernández Marcos 2008: 289; cf. Fernández Marcos 2001: 239–40). As van der Louw (2012: 136) has pointed out, doing so proved more necessary for some books than others. Yet even while acknowledging the necessity of occasionally consulting the Hebrew for difficult texts, the LBGS translators remain alert to the possibility that ambiguities in the Greek might have been deliberate and, if so, should be transferred into the modern translation itself (Fernández Marcos 2001: 236). Broadly speaking, then, the approach toward translating is correctly positioned—and was deliberately staked out—as a “middle road between the English and French projects” that is positioned “closer to the German project” (Fernández Marcos 2008: 288).

ONGOING RESEARCH QUESTIONS The fundamental issue involved in modern translations of the Septuagint has to do with how the modern translators determine whether any given Septuagint rendering constitutes conventional Greek or not. Only after making that decision does the broader and more theoretical matter of dependence, as portrayed in Figure 22.1, come into play. This qualification helps explain why modern translations on different sides of the spectrum can end up with what might otherwise be unexpected results. The example from 2 Kgdms 11:7 provided in Table 22.1 will help bear this out and return discussion to some of the complications raised at the outset of the chapter.14 While there are some text-critical issues involved in this passage, what is of relevance here is the translation of ‫ ׁשלום‬with εἰρήνη. While the former is often used to refer to “welfare,” it is less clear whether the latter was used the same way, although many scholars are doubtful. For example, Rajak (2009: 128) says the choice of εἰρήνη here “could not be less appropriate.” Similarly, Lust (2003: xix) states that this verse “must have been hard to understand for native Greek speakers.” Whatever else might be said about such conclusions, they provide important background for the modern translation choices for this verse, which may seem somewhat counterintuitive in light of the foregoing discussion of method. In the table below, notice that it is only NETS that renders εἰρήνη as “peace,” resulting in an unconventional English translation. All others render the word closer to the meaning of ‫“( ׁשלום‬welfare,” “in Ordnung sei,” “gut gehe,” “va bien”). But if Rajak, Lust, and others are correct that

Note that there is no 2 Kingdoms volume yet published in either BdA or in the Göttingen series.

14

THE SEPTUAGINT AND MODERN LANGUAGE TRANSLATIONS 343

Table 22.1 Translation Comparison of 2 Kgdms 11:7 MT

‫ויבא אוריה אליו ויׁשאל דוד לׁשלום יואב ולׁשלום העם ולׁשלום המלחמה‬

NRSV

When Uriah came to him, David asked how Joab and the people fared, and how the war was going

LXX

καὶ παραγίνεται Ουριας καὶ εἰσῆλθεν πρὸς αὐτόν, καὶ ἐπηρώτησεν Δαυιδ εἰς εἰρήνην Ιωαβ καὶ εἰς εἰρήνην τοῦ λαοῦ καὶ εἰς εἰρήνην τοῦ πολέμου.15 καὶ παραγίνεται Οὐρίας πρὸς Δαυίδ, καὶ ἐπηρώτησεν αὐτὸν Δαυὶδ εἰ ὑγιαίνει Ἰωὰβ καὶ εἰ ὑγιαίνει ὁ λαὸς καὶ εἰ ὑγιαίνει ὁ πόλεμος. καὶ εἶπεν Ὑγιαίνει16

NETS

And Ourias came and went in to him, and Dauid inquired after the peace of Ioab and after the peace of the people and after the peace of the war (Pietersma and Wright 2009: 283)

LES

And Uriah arrived and went in to him. And David inquired with regard to the welfare of Joab and to the welfare of the people and to the welfare of the battle (Penner and Brannan 2019: 372)

BdA

n/a

LXX.D

Und Urias fand sich ein und ging hinein zu ihm, und David fragte, ob es Joab gut gehe und dem Volk und dem Kreig.Rahlfs-Hanhart Und Urias fand sich ein bei David, und David fragte ihn, ob mid Joab und mit dem Volk und mit dem Krieg alles in Ordnung sei. Und er sprach: Es ist alles in Ordnung.Antiochene Text (Kraus and Karrer 2009b: 345)

LBGS

Y Ourías se presenta a donde David y David le preguntó si se encuentra bien Ioab, si se encuentra bien el pueblo y si va bien la guerra. Y dijo: «Va bien» (Fernández Marcos et al. 2011: 276)

εἰρήνη never means anything of the sort, one would expect the opposite choices to have been made in these modern translations. After all, NETS is ostensibly the more likely to consult the meaning of the Hebrew when the Greek is ambiguous and thus be able to make sense of this “topsy-turvy” use of εἰρήνη and translate it into conventional English as “welfare” or the like, yet it does not (Rajak 2009: 128). Likewise, LES, LBGS, and LXX.D are all ostensibly less likely to consult the meaning of the Hebrew—if to varying degrees—and thus would presumably be compelled to render εἰρήνη into English as awkward as the Greek is in the eyes of some, yet they do not. Why? The reason for the somewhat surprising inversion of results in these modern translations is the somewhat paradoxical position that scholars of the Septuagint are placed into by the lexicographical evidence for ancient Greek. As of yet, there are no other known attestations of εἰρήνη used to mean “welfare” in extant Greek sources. Therefore, on the one hand, the NETS translators worked with the assumption that the use of εἰρήνη in 2 Kgdms 11:7 would have been unconventional to a Greek reader, and thus translated it into unconventional English using “peace.” No recourse was taken to the Hebrew in this case, since no adjudication was considered necessary for this particular issue. The current Rahlfs and Hanhart (2006), which does not differ from Rahlfs (1935) or Swete (1887) except in transcription of proper names. There is no Göttingen volume for 2 Kingdoms. 16 The Antiochene text edited by Fernández Marcos and Busto Sáiz (1989: 116). 15

344

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

state of Greek lexicography is decisive given the “stereotypical” rendering (see Figure 22.2 and discussion above). On the other hand, the other modern translations, while certainly aware of the lexicographical situation, do not assume that absence of evidence entails that such a use of εἰρήνη was necessarily unconventional or incomprehensible for an ancient reader. The broader context of the verse in Greek takes on greater importance in clarifying the sense of εἰρήνη here and thus justifies translating into more conventional modern language. Ironically, then, those who hold to opposing side of the methodological spectrum in Figure 22.1 could accuse the other of illegitimately importing meaning specifically from the Semitic source text into the modern translation, although those accusations would be made on opposite theoretical grounds. Thus, in a major way, the real issue lying behind Figure 22.1 is how ancient linguistic conventions are viewed in light of the evidence (or absence thereof ). As this discussion indicates, it is difficult to think of a task within Septuagint scholarship that is broader or more wide-ranging than translating the corpus into a modern language. In a very real sense, all ongoing research questions in the discipline are relevant to it, from the finest details of grammar and textual criticism all the way to matters of culture, history, and theories of communication. To be sure, as each smaller area of research related to Septuagint studies progresses, all modern translations can and should be examined, compared, and improved. While the scholars involved in producing the modern translations discussed here have sometimes stridently upheld the superiority of their methodological framework over the alternatives, few if any would claim to have resolved all the complexities, much less produced a perfect translation, if such a thing exists. In this sense, the research questions connected to translating the Septuagint into a modern language will always remain open-ended.

RECOMMENDED RESOURCES 1. Taylor (2001). This volume contains several important essays by Harl, Dogniez, Pietersma, van der Kooij, and Fernández Marcos discussing and debating the translation method of BdA and NETS, with examples. 2. Wright (2008). A useful overview and comparison of NETS, BdA, and LXX.D. 3. Muraoka (2015). An exemplary exploration and discussion of the translation decisions in numerous passages from NETS, often with an eye to other modern translations. 4. Kraus (2006). An overview of the LXX.D translation method and comparison with NETS and BdA, with examples. 5. Johnson (2016). A useful discussion of the hermeneutical complexities involved in a translated text with reference to NETS, BdA, and LXX.D. 6. Hiebert (2010). This edited volume presents a number of essays interacting with the translation methods of the various modern projects discussed in this chapter. Of particular interest are the essays by Wright, Pietersma, and Kraus.

CHAPTER 23

The Septuagint: The Text as Produced ROBERT J. V. HIEBERT

INTRODUCTION In 2005, the International Organization for Septuagint and Cognate Studies (IOSCS) entered into an agreement with the Society of Biblical Literature Research and Publications Committee to launch the Society of Biblical Literature Commentary on the Septuagint (SBLCS). This occurred as another IOSCS-sponsored project was nearing completion, namely, A New English Translation of the Septuagint, commonly referred to by the acronym NETS (Pietersma and Wright 2009).1 The publication of NETS marked the appearance of the first English translation of the Septuagint in over 160 years since the versions of Charles Thomson (1808) and Lancelot C. L. Brenton (1844) had been introduced to readers. The decision to proceed with a Septuagint commentary series in English based on the same principles that guided the NETS project was not only the next logical step in a process that involves bringing this often-overlooked stage in biblical transmission history to readers, but also represented the first step in the realization of a long-standing dream within the IOSCS that a project of this sort should be inaugurated. The “Preamble to the Guidelines for the Contributors to the SBL Commentary on the Septuagint” articulates the essence of the vision that undergirds both NETS and SBLCS: The objective of SBLCS is “to elucidate the meaning of the text-as-produced in distinction from the text-as-received. Meaning, however, is neither to be presupposed nor to be superimposed from either the source text or the text-as-received.”2 This statement articulates the distinction that is made between meaning as it would have been understood by the Septuagint translators, who would have been conversant with both their Semitic source texts and the Greek target language, and meaning attributed to the product of the translators’ efforts by subsequent readers, most of whom would not have had a working knowledge of Hebrew and Aramaic. Other current translation and commentary projects dealing with the Septuagint evince different approaches to elucidating meaning. The de facto focus of La Bible d’Alexandrie (BdA) is on the Septuagint’s reception history (Harl 1998: 31–5). Septuaginta Deutsch (LXX.D) seeks to chart a course between SBLCS and BdA (Kraus 2006: 63–83). Volumes

Unless otherwise indicated, in the present chapter English translations of Septuagint words or passages are those of NETS. 2 Throughout this chapter, citations of the SBLCS Guidelines come from Büchner 2017: 257–9. 1

346

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

in Brill’s Septuagint Commentary Series (SEPT) are limited to the text of one of the three great Bible codices—fourth-century Sinaiticus and Vaticanus and fifth-century Alexandrinus.3 The SBLCS project is grounded on four principles that are regarded to be inherent in the statement concerning the elucidation of the meaning of the text-asproduced cited above. In the present chapter, I propose to articulate and to flesh out some of the implications of those principles as they apply to my work on a forthcoming commentary on Septuagint Genesis.4

FUNDAMENTAL PRINCIPLES OF THE SBLCS SERIES Principle 1 The first of the basic principles undergirding the SBLCS project is stated as follows: 1.0 The commentary is genetic, in the sense that it seeks to trace the translation process that results in the product, i.e., the so-called original text of the Old Greek. This implies, obviously, that the object of textual commentary will be a critically reconstructed eclectic text. In practical terms, that means reliance on the Göttingen Septuagint for the sections of the Septuagint corpus that have been published in that series. This commitment to critical editions is predicated on the conviction that it is generally feasible to reconstruct the Old Greek (OG) translation of the Hebrew Bible because of the wealth of extant textual evidence, whether in Greek manuscripts of Genesis or in the ancient secondary (“daughter”) versions and the citations of the Church Fathers and other early commentators. In the case of the book of Genesis, I can, therefore, be reasonably assured that John Wevers’ edition is a fair representation of the OG text (Wevers 1974a).5 That said, as Wevers himself demonstrated in the appendix of proposed changes to the critical text in his Notes on the Greek Text of Genesis (Wevers 1993b: 855–6), the quest for the pristine OG text may well continue even after a critical edition has been published. The first sub-point to the first foundational principle states the following: 1.1. The text-as-produced is conceptualized as a dependent entity, derived from its source text. That is to say, it is perceived to be compositionally dependent on its source, though not semantically dependent. That the OG is derived from, and dependent on, its Hebrew source text would seem to be a logical assertion, inasmuch as any translation exists in such a relationship with its parent text. In the field of Septuagint studies, however, the idea that the product of an original Greek translator’s efforts is dependent on the Semitic Vorlage has met with some resistance from those who regard the Septuagint to be an independent entity that, for its originally intended Hellenistic Jewish readership, would in effect have served as a replacement for the Hebrew text (Harl 1992: 33–42, 1998: 31–5; Dogniez and Harl Online: https://brill.com/view/serial/SEPT (accessed March 6, 2020). See the chapter by Porter in the present volume. 4 I hereby express my gratitude to the Social Sciences and Humanities Research Council of Canada for the funding to support my research for this commentary volume. 5 In the case of books for which there is not yet a full-fledged Göttingen critical edition (Joshua, Judges, Ruth, 1-2 Samuel and 1-2 Kings [1-4 Kingdoms], 1 Chronicles, 4 Maccabees, Proverbs, Ecclesiastes, Song of Songs, and Psalms of Solomon), the Septuagint edition by Alfred Rahlfs will serve as the default text. 3

THE SEPTUAGINT: THE TEXT AS PRODUCED 347

2001: 14; Muraoka 2008: 221–35).6 To be sure, this would have been the case during the course of the reception history of the Septuagint for those who considered the translators’ work to be the product of divine inspiration and to be revered as Scripture. In his treatise On the Life of Moses, Philo of Alexandria, who embellishes the account of the translation of the “Law” or Pentateuch that is found in a second-century BCE composition known as the Letter of Aristeas (§§302, 307–19), articulates this perspective: [The translators] became as it were possessed, and, under inspiration, wrote, not each several scribe something different, but the same word for word, as though dictated to each by an invisible prompter … [T]hese writers, as it clearly appears, arrived at a wording which corresponded with the matter, and alone, or better than any other, would bring out clearly what was meant. The clearest proof of this is that, if Chaldeans have learned Greek, or Greeks Chaldean, and read both versions, the Chaldean and the translation, they regard them with awe and reverence as sisters, or rather one and the same. (Vit. Mos. 2.37, 39-40 in Colson 1935) An even greater reverence for the Greek version is expressed in this description by Epiphanius (On Weights and Measures 76–83, 143–53, 155–6): They were seventy-two in number and were shut up from morn till eve in the Pharian island … over against Alexandria, in six and thirty cells, one pair to each cell … They slept in pairs in thirty-six bed-chambers. Such was the life they led, to the end that they should have no collusion with each other and should produce an unadulterated translation … [T]he books were distributed periodically to each pair of translators, passing from the first pair to the second, and again from the second to the third in rotation. Thus each book was translated thirty-six times … There were also twentytwo of the apocryphal books. When the work was completed, the king took his seat on a lofty throne, and thirty-six readers sat at his feet having thirty-six reproductions of each book, while one held a copy of the Hebrew volume. Then one reader recited and the rest diligently attended; and there was found no discrepancy … where they added a word they all added it in common, and where they omitted the omission was made by all alike. (Hadas [1951] 2007: 76–7) This miraculous achievement could mean nothing less than that those who produced the whole Septuagint corpus, Apocrypha included, were divinely inspired. But the question is whether the translated corpus—and for our purposes the Pentateuch and the book of Genesis in particular—would have been so regarded at the point of production. It is the contention of the participants in the NETS and SBLCS projects that the textual-linguistic make-up of these translations suggests otherwise, because of what those translations reveal about the ways in which the translators went about their work and because of the nature of the relationship between their products and their source texts. The assertion in the first sub-point that a relationship of compositional dependence, but not semantic dependence, exists between the text-as-produced and its source text also

See A. Pietersma’s response to Muraoka, available online: http://ccat.sas.upenn.edu/nets/discussion/pietersma-remuraoka.pdf (accessed March 6, 2020).

6

348

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

warrants explication. Despite the insistence by Philo and Epiphanius that there was exact semantic correspondence between the source text and the products of the translators, it is clear when one makes careful comparisons that this is often simply not the case. This is so even in instances in which the two are quantitatively equivalent, as in the following example: Gen. 11:1

‫‏ ויהי כל הארץ ׂשפה אחת‬ Now the whole earth had one language (NRSV) Καὶ ἦν πᾶσα ἡ γῆ χεῖλος ἕν And the whole earth was one lip (NETS)

The Hebrew and Greek texts consist of seven morphemes each: conjunction, finite verb, attributive modifier/qualifier, article, noun, noun, adjective (cardinal number). While each Greek constituent corresponds to its Hebrew counterpart, there is semantic dissonance in the Greek version in comparison to its Hebrew source text, primarily because nowhere in contemporaneous or antecedent Greek compositional literature is χεῖλος attested to mean “language,” unlike the case for Hebrew ‫ׂשפה‬. Both terms frequently denote “lip,” but the semantic range of ‫ ׂשפה‬also includes “language.” The preceding example also illustrates the significance of the second sub-point: 1.2. The aim is to uncover the strategies and norms7 by means of which the text came into being. Therefore, the commentator will analyze the relationship between the target text and the source text, attempting to account for the process underlying the derivation of the Greek version from its Semitic parent. It is from this analysis that the commentator will formulate his or her principles of interpretation and procedural methodology. The choice of χεῖλος as the counterpart for ‫ ׂשפה‬in this context constitutes a case of what Gideon Toury, a specialist in the field of descriptive translation studies, would call a form of interference from the source text, whereby “phenomena pertaining to the makeup of the source text tend to be transferred to the target text.” This particular kind of interference is known as negative transfer, which entails “deviation from normal, codified practices of the target system” (Toury 1995: 275). Thus it can be concluded that the strategy the translator adopted in the present instance was to choose the default equivalence employed in Genesis despite the semantic tension this created: ‫ = ׂשפה‬χεῖλος (11:1, 6, 9; 22:17; 41:3, 17); ‫ = ׂשפה‬γλῶσσα (11:7[1°]); ‫ = ׂשפה‬φωνή (11:7[2°]). The result, employing Toury’s terminology once again, is an adequate translation, one that accords with “the norms of the source text, and through them also to the norms of the source language and culture” (Toury 1995: 56; see also Even-Zohar 1975: 43). Of course, interference does occur not only when lexical equivalences are involved but also in matters of syntax. For example, one of the syntactical phenomena that occurs in biblical Hebrew is the infinitive absolute + cognate finite verb construction, which is typically employed to signify emphasis of a verbal idea (GKC §113 l-n). This construction is rendered in a number of ways in Genesis, one of which is by means of a Greek dative

“Translational norms may be conceived of as culturally sanctioned guidelines for resolving the tensions inherent in the process of translation” (Boyd-Taylor 2006b: 31 n. 11).

7

THE SEPTUAGINT: THE TEXT AS PRODUCED 349

noun + cognate finite verb—a construction that is attested in compositional Greek literature that predates the Septuagint, though with not nearly as great a frequency as in the Septuagint (Conybeare and Stock [1905] 1995: §61; Thackeray 1908: 597–8; Sollamo 1985: 105–7; cf. Plato, Crat. 385b, 10 and Phaedr. 265c, 8-9). Gen. 2:17

‫מות תמות‬ You shall die (NRSV) θανάτῳ ἀποθανεῖσθε You shall die by death (NETS)

Gen. 40:15

‫כי גנב גנבתי‬ For in fact I was stolen (NRSV) ὅτι κλοπῇ ἐκλάπην For by stealth I was stolen (NETS)

In these cases, there is once again quantitative equivalence in regard to the number of lexical constituents in the Hebrew and Greek texts. And although there is precedent in Greek compositional literature for the kind of equivalent chosen by Septuagint translators to render the Hebrew syntactical construction, the frequency with which that translation approach occurs in the Septuagint makes for another kind of interference from the source text. Toury calls this positive transfer because it involves the strategy of selecting equivalents that are consistent with the norms of the language of the target text but whose frequency of occurrence in the target text is occasioned by the number of times that its counterpart appears in the source text (Toury 1995: 275). The translator of Genesis has, however, produced not only adequate translations that reflect source text norms, but also acceptable translations for which “norm systems of the target culture are triggered and set into motion” (Toury 1995: 56). These two types of translations can be illustrated by contrasting different strategies for rendering the Hebrew construction that involves the combination of the verb ‫ יסף‬+ another verb to signify either the repetition or the continuation of the action of the second verb. In ten of eleven occurrences of this construction in Genesis, the Greek translator has employed the verb προστίθημι (“put to,” “add,” etc. [LSJ, ad loc.]) for ‫“( יסף‬add,” “do again,” “continue to do,” etc. [DCH, ad loc., both Qal and Hiphil]), presumably because of their partially overlapping semantic ranges. This equivalence produces stilted Greek—that is, adequate translations— due to negative transfer from the source text8 as the following example illustrates: Gen. 4:12

‫‏כי תעבד את האדמה לא תסף תת כחה לך‬ When you till the ground, it will no longer yield to you its strength (NRSV) ὅτι ἐργᾷ τὴν γῆν, καὶ οὐ προσθήσει τὴν ἰσχὺν αὐτῆς δοῦναί σοι For you will till the earth, and it will not add 9 to yield its strength to you (NETS)

Gen. 4:2, 12; 8:12, 21 (2×); 18:29; 25:1; 37:8; 38:5, 26; 44:23. In NETS, the rendering is “continue.”

8 9

350

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

In this case, the Hebrew construction involves ‫ יסף‬plus an infinitive verb, and the Greek follows suit with προστίθημι plus an infinitive. A more acceptable translation, by Toury’s standards, of the ‫ יסף‬+ infinitive construction is found in a single occurrence of the Greek translator’s rendering of that configuration in Genesis. Gen. 8:10

‫ויחל עוד ׁשבעת ימים אחרים ויסף ׁשלח את היונה מן התבה‬ He waited another seven days, and again he sent out the dove from the ark (NRSV) καὶ ἐπισχὼν ἔτι ἡμέρας ἑπτὰ ἑτέρας πάλιν ἐξαπέστειλεν τὴν περιστερὰν ἐκ τῆς κιβωτοῦ And when he had waited yet another seven days, again he sent forth the dove from the ark (NETS)

In this case, the adverb πάλιν serves as a fitting counterpart to ‫ויסף‬, and the infinitive ‫שלח‬ is rendered aptly by the finite verb ἐξαπέστειλεν.

Principle 2 The second basic principle upon which the SBLCS project is founded states: 2.0 The primary focus of the commentary is the verbal make-up of the translation, understood in terms of conventional linguistic usage (i.e., the grammar and lexicon of the target language) rather than in terms of what may be encountered in translation Greek. This principle derives from the assumption that, although the Septuagint is a translation of a Semitic source text, its role and function in the culture in which it was produced can be determined only in relation to how it would have compared to compositional, rather than translational, writing at that time and place. The implications of this assumption with respect to understanding the meaning of this translated text are further explicated in the first sub-point. 2.1. The text-as-produced can be said to have semantic autonomy because it means what it means in terms of the grammar and lexicon of the Greek language at the time of the Septuagint’s production. In arguing for the existence of a distinctive vernacular of Greek employed by Jews in the Hellenistic period, Henry Gehman says the following: It certainly would be too bold to speak of a “Jewish-Greek jargon”, and yet we can hardly avoid speaking of a Jewish-Greek, which was in use in the synagogues and in religious circles. If the Jews who read the LXX did not understand Hebrew, we may infer at least that the translation made sense to them and that it was intelligible when it was read in the synagogue. (1951: 81; see also Turner 1955: 208–13 and Turner 1965) In his critique of Gehman’s assertion, John Lee points out that “no one would suppose that because we can understand the English of the AV [Authorized Version] its idioms must be a normal feature of the English we speak” (Lee 1983: 19). Gehman proceeds to discuss a series of syntactical and lexical examples in regard to which he concludes that,

THE SEPTUAGINT: THE TEXT AS PRODUCED 351

in this alleged Jewish-Greek dialect, the Greek renderings are to be understood essentially as semantic equivalents of the Hebrew source text. Thus in certain situations, he claims, καί and ‫ ו‬can mean “that,” “while,” “then,” or “so”; ὅτι and ‫ כי‬can mean “when”; θάλασσα and ‫ ים‬can mean “west”; and χεῖλος and ‫ ׂשפה‬can mean “language.” (Gehman 1951: 81– 90). However, careful philological research would suggest otherwise with respect to the cases he considers. Gehman’s conclusions are based on the flawed a priori assumption that the Greek translation means what the Hebrew source text does, a notion that is often not supported by evidence from contemporaneous non-translation literature. In fact, the more that scholars have studied Greek papyri and inscriptions, the clearer it has become that the language of the Septuagint is, by and large, that of the vernacular of the Hellenistic period. This has been documented already by Adolf Deissmann, who maintains, quite rightly, that the “language, spoken and written, of the Seventy Interpreters was the Egyptian Greek of the period of the Ptolemies” (Deissmann 1901: 70). He goes on to state that “it would be a serious mistake to suppose that the LXX everywhere used each particular word in the sense of its corresponding Hebrew. Very frequently the LXX did not translate the original at all, but made a substitution for it, and the actual meaning of the word substituted is, of course, to be ascertained only from Egyptian Greek” (1901: 74). There are any number of examples in Genesis of semantic differences between the Hebrew source text and its Septuagint counterpart. One such case involves the ‫= מרע‬ νυμφαγωγός pair. A ‫ מרע‬is a close friend or companion of one sort or another (BDB, ad loc., ‫ ;מֵ ֵר ַע‬HALOT, ad loc., I ‫)מֵ ֵר ַע‬. Various renderings of this term are found in the Septuagint: ἑταῖρος “companion” (Judg.A 14:11, 20); συνέταιρος “companion” (Judg.A 15:2, 6); γνώριμος “acquaintance” (2 Sam./Kgdms 3:8); φίλος “friend” (Prov. 19:4); φιλία “friendship” (Prov. 19:7). Twice elsewhere it is rendered as νυμφαγωγός, denoting “leader of the bride … esp. one who leads her from her home to the bridegroom’s house” (LSJ, ad loc., I.1). In Judg.A 14:20, ‫ = מרע‬νυμφαγωγός is the equivalence used for the man to whom Samson’s bride goes after Samson’s falling out with the thirty ‫ = מרעים‬ἑταῖροι (14:11) who have extorted the answer to his riddle from her (14:12-20). In this context, νυμφαγωγός is rendered as “bridal escort” in NETS. The other place where the ‫= מרע‬ νυμφαγωγός equivalence occurs is in Gen. 26:26. Here the referent is Ahuzzath/Ochozath, one of King Abimelech’s men, and in NETS the translation of νυμφαγωγός is “groomsman.”10 John Wevers comments on the unexpectedness of this rendering: The word is ill-fitting, since it refers to the close friend of the bridegroom whose duty it was to escort the bride from her father’s house. It obviously does not mean that here … The Hebrew title is ‫“ מרע‬companion, confidential friend,” and Gen’s choice of νυμφαγωγός was probably due to the fact that the νυμφαγωγός was the closest friend of the bridegroom, and must then here mean “the confidential, personal advisor” of the king. (Wevers 1993b: 410–11) Wevers’ conclusion that νυμφαγωγός does not mean a bridegroom’s close friend and wedding attendant in Gen. 26:26 is in line with what is said about its meaning in the Septuagint in the LSJ lexicon’s entry: “generally, friend, LXX Ge. 21.22, Jd. 14.20” (LSJ,

Note that νυμφαγωγός occurs as well in Genesis 21:22, 32, where the referent is also Ochozath but where there is no Hebrew counterpart in the MT.

10

352

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

ad loc., I.2). G. B. Caird demurs in his critical appraisal of LSJ’s comments regarding denotations suggested for the Septuagint: In Judg. 14:20 Samson’s friend is explicitly described as one who had been his best man, and the A text of the LXX properly represents this by νυμφαγωγός. The translators of the Pentateuch were clearly familiar with this sense of ‫מֵ ֵר ַע‬, and they could see that in Gen. 21:22, 32; 26:26 it was used as an official title for Ahuzzath and took it to mean ‘King’s Groomsman’ (cf. ‘King’s Eye’ at the Persian court). It is therefore quite misleading to suggest that νυμφαγωγός was ever used in the general sense of ‘friend’. In Judges it means ‘groomsman’; and the entry for Genesis should read: ‘companion, of a king’s confidential adviser, LXX Ge. 21.22’. (Caird 1969: 24–5) Where I would diverge from Caird’s critique, however, is in his distinction between what νυμφαγωγός means in Judges and what it means in Genesis. Its usage in compositional Greek literature would suggest that “groomsman” would be an appropriate gloss in both places, which is what appears in NETS Gen. 26:26.11 The fact that there is no mention in the Genesis contexts of Ahuzzath assuming this role for Abimelech is not germane in the final analysis. The next sub-point of the second principle focuses on the reader for which the translated text would have been intended: 2.2. The “reader” of the text-as-produced is conceptualized as the prospective or implied reader, a construct based on the text itself, in distinction from any reader, actual or hypothetical, exterior to the text. The prospective reader is to be inferred from those features of the text’s make-up that are indicative of a specific linguistic, literary, or cultural aim. Given the paucity of documentary evidence regarding who exactly it was who decided to translate the Hebrew Pentateuch into Greek and who actually carried out this task, why it was undertaken in the first place, and for whom the result was intended, it would appear that the most reliable source of information that one might consult in order to arrive at some kind of answer to these questions is the translation product itself. As indicated above, sources such as the Letter of Aristeas date to at least a century after the Pentateuch would have been translated, and their testimony is colored by the particular apologetic interests of their authors with regard to asserting the authoritativeness of the translation. In that light, the search for the prospective or implied reader, then, must be based on an investigation of the linguistic and semantic features of the text. What one encounters in such a study is sometimes puzzling, in particular because of the variability of the translation technique that one encounters, even within a defined translation unit such as the book of Genesis that contains evidence of a generally conservative approach to the rendering of the source text. Deissmann speaks of “the marvelous variety of the linguistic elements of the Greek Bible” (1901: 65). The Septuagint translators took up their task despite the fact that “the idea of making the sacred book accessible in another language  was at that time unheard-of”—namely, the task “of turning Semitic into Greek”

In NETS Judg.A 14:20, the rendering is “bridal escort.” Note that, in same verse, ‫“( ֵרעָה‬be a special friend, be best man” [DCH, ad loc., II piel]) is translated as ἦν ἑταῖρος “was [his] companion.”

11

THE SEPTUAGINT: THE TEXT AS PRODUCED 353

(1901: 67)—which may be part of the reason for the unevenness of the result. What this might mean with respect to sketching a profile of the targeted reader is an interesting question. The variability in translation approach may be illustrated by both lexical and syntactical equivalences. Three times the name of the primeval garden ‫( גן‬which is consistently called a παράδεισος “orchard” [LSJ, ad loc., 2] in Septuagint Genesis) into which the Creator places the original human couple is a transcription of the Hebrew ‫עדן‬, namely, Ἔδεμ “Edem” (Gen. 2:8, 10; 4:16). But on two additional occasions the ‫“ גן עדן‬garden of Eden” is rendered as ὁ παράδεισος τοῦ τρυφῆς “the orchard of delight” (Gen. 3:23, 24). In these cases the translator evidently makes the link between the proper noun ‫“ עדן‬Eden” and its homonym, the common noun that denotes “delight, pleasure” (DCH, ad loc., 1.3 ‫)עֵדֶ ן‬. On five occasions in Septuagint Genesis, the region of Egypt in which Jacob and his family settle and which the Hebrew text calls ‫“ גׁשן‬Goshen” is transcribed as Γέσεμ “Gesem” (47:1, 4, 6[5], 27; 50:8), and in two additional instances the same Hebrew term is expanded to become Γέσεμ Ἀραβίας “Gesem of Arabia” (45:10; 46:34).12 But then twice ‫ גׁשן‬is rendered by the replacement name Ἡρώων πόλις “Heroonpolis” (Gen. 46:28[1°], 29), which is to be associated with Tell er-Retaba or biblical Pithom (Kitchen 2003: 256–9). Once the counterpart to ‫“ ארץ גׁשן‬the land of Goshen” is γῆ Ῥαμεσσή “the land of Ramesses.” (Gen. 46:28[2°]).13 The connection between ‫ גׁשן‬and Γέσεμ seems to represent a link to an individual mentioned in Neh. 2:19/2 Esd. 12:19 and 6:1-2/16:1-2 named ‫“ גׁשם הערבי‬Geshem the Arab”/Γῆσαμ ὁ Ἀραβί “Gesam the Arabi.” He was one of Nehemiah’s adversaries during his efforts to reconstruct the walls of Jerusalem and in fact a powerful North Arabian ruler in the mid-fifth century, nominally under Persian hegemony. As is indicated by the fact that his name is inscribed on a silver vessel that was found at Tell el-Maskhuta (Tjeku or biblical Succoth [Exod. 13:20]), he exercised control as far as the eastern Nile Delta, and it would appear that his renown was such that this whole region was known—even nearly two centuries later when the first readers of Greek Genesis lived—as the land of Geshem/Gesem/Gesam of Arabia (Rabinowitz 1956: 109; Williamson 1985: 192; Ward 1992: 2.1076–7; Kitchen 2003: 257–8). Greek literary evidence for the use of the name Βαβυλών “Babylon” dates back to the pre-classical and classical periods,14 so the decision by the Septuagint translator to use this replacement form as the counterpart to ‫“ בבל‬Babel” in Gen. 10:10 is not surprising. In Gen. 11:9, however, the translator undertakes to reproduce for the reader the wordplay in the Hebrew source text by introducing another type of replacement for ‫בבל‬:

Note that in the text of Wevers’ Göttingen edition, the second term is in the dative case Ἀραβίᾳ, but that in the appendix of his textual commentary volume containing a list of proposed changes to the critical text, he switches it to the genitive Ἀραβίας (Wevers 1993b: 856). 13 On one additional occasion in Genesis, γῆ Ῥαμεσσή “the land of Ramesses” is the equivalent for ‫“ ארץ רעמסס‬the land of Rameses” (Gen. 47:11). 14 For example, Lycaeus of Mytilene (seventh to sixth centuries BCE), Pindar, Aeschylus, Pythagoras (all sixth to fifth centuries BCE), and Herodotus (fifth century BCE), as a search in TLG reveals. 12

354

Gen. 11:9

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

‫על כן קרא ׁשמה בבל כי ׁשם בלל יהוה ׂשפת כל הארץ‬ Therefore it was called Babel, because there the Lo r d confused the language of all the earth (NRSV) διὰ τοῦτο ἐκλήθη τὸ ὄνομα αὐτῆς Σύγχυσις, ὅτι ἐκεῖ συνέχεεν κύριος τὰ χείλη πάσης τῆς γῆς Therefore its name was called Confusion, because there the Lord confused the lips of all the earth (NETS)

In eleven of the thirteen occurrences of the place name ‫“ בית־אל‬Bethel,” the Greek counterpart is the transcription βαιθήλ “Baithel.”15 In Gen. 28:19, the translator interprets the meaning of the Hebrew name for the reader: Gen. 28:19

‫ויקרא את ׁשם המקום ההוא בית אל‬ He called that place Bethel (NRSV) καὶ ἐκάλεσεν Ἰακὼβ τὸ ὄνομα τοῦ τόπου ἐκείνου Οἶκος θεοῦ And Iakob called the name of that place Divine–house (NETS)

In Gen. 31:13, it may be that the translator’s Hebrew Vorlage is different than that of the MT, or that the translator harmonizes this text with others (35:1, 7) that contain this name and also mention God’s appearance to Jacob there: Gen. 31:13

‫אנכי האל בית אל‬ I am the God of Bethel (NRSV) ἐγώ εἰμι ὁ θεὸς ὁ ὀφθείς σοι ἐν τόπῳ θεοῦ I am the God who appeared to you at a divine place (NETS)

The formula that is used in Hebrew for stating someone’s age involves the terms ‫בן‬ “son” or ‫“ בת‬daughter” plus the number and the term for year(s).16 This does not, of course, correspond to Greek idiom, and in the Septuagint such Hebrew expressions are typically replaced by idiomatic Greek constructions, as is illustrated in the following examples: Gen. 12:4

‫בן חמׁש ׁשנים וׁשבעים ׁשנה‬ Seventy-five years old (NRSV) ἐτῶν ἑβδομήκοντα πέντε Seventy-five years of age (NETS)

Gen. 17:17

‫בת תׁשעים ׁשנה‬ Ninety years old (NRSV) ἐνενήκοντα ἐτῶν Ninety years of age (NETS)

Gen. 12:8 (2×); 13:3 (2×); 35:1, 3, 6, 7, 8, 15, 16. Gen. 11:10; 12:4; 16:16; 17:1, 17 (2×); 21:5; 25:20, 26; 26:34; 41:46.

15 16

THE SEPTUAGINT: THE TEXT AS PRODUCED 355

In a single instance, however, the Hebrew age formula is replicated in Septuagint Genesis: Gen. 11:10

‫בן מאת ׁשנה‬ One hundred years old (NRSV) υἱὸς ἑκατὸν ἐτῶν A son of one hundred years (NETS)

The preceding examples illustrate the variability of the translator’s approach to rendering the source text. It is clear that this variability is not due to the translator’s lack of competence, but in fact in these cases it reveals that he understands the text perfectly well. Generally speaking, the interpretive alternatives to the transcriptions or to the isomorphic or Hebraizing renderings are quite evidently hermeneutically motivated to enhance the reader’s understanding of the text. Based on the evidence of the translation product, one may conclude that the prospective or implied reader of the Greek translation would have had little or no access to the underlying source text and, consequently, would need to have had Hebrew wordplays interpreted in Greek. Geographical and historical knowledge would seem to have been assumed in relation to the mention of names such as Γέσεμ Ἀραβίας “Gesem of Arabia,” Ἡρώων πόλις “Heroonpolis,” γῆ Ῥαμεσσή “the land of Ramesses,” and Βαβυλών “Babylon.”

Principle 3 The third basic principle of the SBLCS commentary series is as follows: 3.0 The text-as-produced represents a historical event and should be described with reference to the relevant features of its historical context. In the case of the Septuagint, the actual historical circumstances of this ancient translation project are, as indicated above, unknown. But this does not mean that nothing can be ascertained about the context out of which the translation emerged, something to which the first sub-point of this principle speaks: 3.1. The translation is to be viewed as a fact of the culture that produced it inasmuch as it is a specimen of discourse within that culture. The cultural milieu out of which the Septuagint emerged was a mixed one, and the translation itself evinces elements of Jewish, Egyptian, and Greek features. The most obvious evidence of the Jewish environment within which the translation was produced is the character of the translation itself, which, for the most part, reflects the Hebrew source text quite closely, not infrequently to the point of isomorphism and Hebraisms. The examples discussed above are indicative: ‫“ ויהי כל הארץ ׂשפה אחת‬Now the whole earth had one language” = Καὶ ἦν πᾶσα ἡ γῆ χεῖλος ἕν “And the whole earth was one lip” (Gen. 11:1); ‫“ בן מאת ׁשנה‬one hundred years old” = υἱὸς ἑκατὸν ἐτῶν “a son of one hundred years” (Gen. 11:10). Apart from such linguistic phenomena, however, the translation contains clues regarding the Jewish religious and cultural context in which it was produced. This may be illustrated, for example, with regard to some of the things that are said about circumcision and burial practices. The matter of circumcision comes up in the story in Genesis 34 concerning the revenge exacted by the brothers of Jacob’s daughter, Dinah,

356

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

following her violation by Shechem, the son of Hamor. In response to the request by Hamor that Dinah be given in marriage to Shechem and to the proposal that the Hamor and Jacob clans intermarry, Dinah’s brothers (in the OG, specifically Symeon and Leui) insist that this will only be allowed to happen if Hamor, Shechem, and their male compatriots agree to be circumcised. They do so, and Gen. 34:25 describes what ensues: Gen. 34:25

‫ויהי ביום הׁשליׁשי בהיותם כאבים ויקחו ׁשני בני יעקב ׁשמעון ולוי אחי דינה איׁש חרבו ויבאו‬ ‫על העיר בטח ויהרגו כל זכר‬ On the third day, when they were still in pain, two of the sons of Jacob, Simeon and Levi, Dinah’s brothers, took their swords and came against the city unawares, and killed all the males (NRSV) ἐγένετο δὲ ἐν τῇ ἡμέρᾳ τῇ τρίτῃ, ὅτε ἦσαν ἐν τῷ πόνῳ, ἔλαβον οἱ δύο υἱοὶ Ἰακὼβ Συμεὼν καὶ Λευὶ οἱ ἀδελφοὶ Δίνας ἕκαστος τὴν μάχαιραν αὐτοῦ, καὶ εἰσῆλθον εἰς τὴν πόλιν ἀσφαλῶς καὶ ἀπέκτειναν πᾶν ἀρσενικόν· Now it came about on the third day, when they were in pain, that the two sons of Iakob, Symeon and Leui, Dina’s brothers, took each one his dagger and entered into the city safely and killed every male (NETS)

The translator’s rendering of the phrase ‫ בהיותם כאבים‬as ὅτε ἦσαν ἐν τῷ πόνῳ is noteworthy, inasmuch as Hebrew ‫ ב‬+ infinitive construct configurations are variously translated in Genesis. The three most common ways of doing so are the following: ἐν + the articular infinitive (18×)17 ὅτε + finite verb (10×)18 ἡνίκα + finite verb (9×)19 The first of the three constructions above is used when attendant or accompanying circumstances in general (e.g., causal, concessive, temporal) are involved, whereas the latter two, by means of the adverbs of time that are employed, signify temporal context specifically. This would appear to indicate that, in Gen. 34:25, the Greek translator is marking the third day after the mass circumcision as being especially significant because of the degree of discomfort that the affected men would have been experiencing. That seems to be how other early versions understand the situation: Gen. 34:25

‫( והוה ביומא תליתאה כד תקיפו עליהון כיביהון‬Targum Onquelos) And on the third day when their pains grew strong against them ‫( והוה ביומא תליתאה כד הוו מתמקמקין מן כיב גוזרתהום‬Targum Pseudo-Jonathan) And on the third day when they were shaking with the pain of their circumcision

Gen. 4:8; 9:14; 11:2; 19:29 (2×), 33 (2×), 35 (2×); 28:6; 32:20(19), 26(25); 34:22; 35:1, 7, 17 (2°), 18; 38:28. 18 Gen. 2:4; 12:4; 25:20, 26; 33:18; 34:25; 35:9; 36:24; 41:46; 42:21. 19 Gen. 16:16; 17:24, 25; 21:5; 30:42; 35:22(21); 38:5; 45:1; 48:7. 17

THE SEPTUAGINT: THE TEXT AS PRODUCED

357

‫( והוה ביומא תליתיא כד הוון צעורין מן גזרתהון‬Targum Neofiti) And on the third day when they were suffering with the pain of their circumcision ‫( ܘܗܘܐ ܒܝܘܡܐ ܬܠܝܬܝܐ ܟܕ ܚܣܢܘ ܟܐܒܝܗܘܢ‬Peshitta) And on the third day when their pain grew strong et ecce die tertio, quando gravissimus vulnerum dolor est (Vulgate) And behold on the third day, when the pain of the wounds is most grievous

The fact that, in rabbinic tradition, the third day after circumcision was regarded to be when the individual was most vulnerable and the pain experienced was the most intense is indicated in various Talmudic tractates that appeal to the reference in Gen. 34:25 to the third day (b. Ned. 31b; b. Šabb. 134b; b. B. Meṣ. 86b). Nahum Sarna (1989: 387) comments: If circumcision has been postponed for medical reasons, it may not subsequently be carried out on a Sabbath or holy day. Once postponed, it is also not performed on a Thursday because this might lead unnecessarily to profanation of the Sabbath, which would be the third day of circumcision, when the pain is thought to be most intense and some special treatment might be called for. The Greek translator of Genesis treads carefully when describing the funeral rites for Jacob and Joseph. The Hebrew author uses the term ‫“ חנט‬embalm” (DCH, ad loc.) when depicting the preparation of their bodies (Gen. 50:2 [2×], 26), which of course is consistent with what was standard Egyptian practice in antiquity, especially for royals, nobles, and officials. Instead of employing an equivalent like ταριχεύω “preserve the body by artificial means, embalm” (LSJ, ad loc.) or σκελετεύω “dry or salt flesh … embalm a corpse” (LSJ, ad loc.), the translator chose denotations that do not explicitly signify this preservative process, namely, ἐνταφιάζω “prepare for burial, lay out” (50:2 [2×]; LSJ, ad loc.; cf. ἐντάφιον “shroud, winding-sheet”) and θάπτω “honor with funeral rites” (50:26; LSJ, ad loc.). It may well be that these translation choices were motivated by Jewish theological considerations, given the religious nature of the practice of preparing the dead for burial in ancient Egypt (Jones 1992: 2.493). The OG version of Genesis exhibits evidence, not only of sensitivity to Jewish concerns and practices, but also of the Egyptian setting of its translator. A case in point has to do with the rendering of the infrequently occurring term ‫אׁשל‬: Gen. 21:33

‫ויטע אׁשל בבאר ׁשבע ויקרא ׁשם בׁשם יהוה אל עולם‬ Abraham planted a tamarisk tree in Beer-sheba, and called there on the name of the lord, the Everlasting God (NRSV) καὶ ἐφύτευσεν Ἀβραὰμ ἄρουραν ἐπὶ τῷ φρέατι τοῦ ὅρκου, καὶ ἐπεκαλέσατο ἐκεῖ τὸ ὄνομα κυρίου Θεὸς αἰώνιος. And Abraam planted aa ploughed fielda by the well of the oath and there invoked the name of the Lord: Everlasting–god20 (NETS)

Or an aroura = a measure of land (nearly two-thirds of an acre) in Egypt.

20a

358

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

The ‫ אׁשל‬that Abraham planted is likely to be identified as Tamarix aphylla, which is common in the Negeb (Wenham 2000: 94).21 Whether this word was unknown to the Greek translator or he wished to avoid the possibility of conjuring up connotations of heterodox cultic activity associated with the planting of trees at a shrine,22 the choice of ἄρουρα as the counterpart constitutes a semantic replacement—a term that is attested as early as Homer’s Iliad, where it denotes the earth, earth or ground in general, tilled or arable land, or grain field, etc. (LSJ, ad loc.). Herodotus, however, reports that in Egypt an ἄρουρα is also a piece of land that measures 100 Egyptian cubits square (Hist. 2.168, 3-4). So this translation is indicative at least of the Greek world of the translator, and possibly also of his Egyptian context in particular. Another indication of the Egyptian context of the translator relates to the description of the sale of Joseph by his brothers: Gen. 37:28

‫וימכרו את יוסף ליׁשמעאלים בעׂשרים כסף‬ And [they] sold him to the Ishmaelites for twenty pieces of silver (NRSV) καὶ ἀπέδοντο τὸν Ἰωσὴφ τοῖς Ἰσμαηλίταις εἴκοσι χρυσῶν And [they] sold Ioseph to the Ismaelites for twenty gold pieces (NETS)

The difference in value regarding this transaction has to do no doubt with the fact that the going rate for slaves in the time of the Greek translator was not the same as it would have been in the time of the Hebrew author. The term χρυσοῦς was used from the third century BCE onward in Egypt and elsewhere to designate a measure of both weight and value, although in Egypt during the Ptolemaic era it appears as though no gold coin of this denomination was in circulation. If ‫“( כסף‬silver”) was considered to correspond to the silver δραχμή, a price of twenty silver drachmae for Joseph would have seemed unusually low in view of the fact that male slaves were typically sold for between 100 and 300 drachmae in third-century BCE Egypt (Westermann 1929: 60–1). Hence, given that a χρυσοῦς was equivalent to twenty silver drachmae, a price of 400 drachmae would have been a more realistic price for Joseph in that context (Lee 1983: 63–5). The term δραχμή does, in fact, also appear in Septuagint Genesis in the description of the gifts that Abraham’s servant, on a mission to find a suitable wife for Isaac, gives to Rebekah: Gen. 24:22

‫ויקח האיׁש נזם זהב ׁבקע מׁשקלו‬ The man took a gold nose-ring weighing a half shekel (NRSV) ἔλαβεν ὁ ἄνθρωπος ἐνώτια χρυσᾶ ἀνὰ δραχμὴν ὁλκῆς The man took gold earrings a drachma each in weight (NETS)

The conversion of a nose-ring to earrings and of a half-shekel weight to that of a drachma speaks to the differences in fashion jewelry and weight terminology in the respective contexts of the Hebrew author and the Greek translator in Egypt. It should be noted that the translator’s choice of δραχμή for ‫ בקע‬was appropriate, since a shekel and a didrachma were equivalent in weight (Wevers 1993b: 353; Hiebert 1994: 205–6, 213 n. The two other places in which the term ‫ אׁשל‬occurs are 1 Sam. 22:6 and 31:13. See, for example, Deut. 16:21.

21 22

THE SEPTUAGINT: THE TEXT AS PRODUCED 359

36). Thus the weight of each earring (drachma) would have been equivalent to that of the nose-ring (half shekel). The second sub-point of the third foundational principle addresses the matter of the kind of document that is produced by the process of translation. 3.2. The verbal make-up of the translation should be understood in relation to the cultural system in which it was produced, that is to say, the sort of text it is as a Greek document. The overall profile of the OG version of Genesis that has been sketched thus far indicates that, as a Greek document, it is a translated text that exhibits linguistic interference—involving both positive and negative transfer—from its source text. Toury delineates three modes of translation—linguistically motivated, textually dominated, and literary—which he describes as “a series which is hierarchically ordered in terms of the specificity of the conditions they impose on the act” of rendering a source text into a target language (Toury 1995: 170–1). One might distinguish these differing modalities in the following manner: (1) a linguistically motivated translation is grammatically and semantically adequate in terms of adhering to source text norms and its departures from the form of the source text are typically limited to those that are required by the target language (Boyd-Taylor 2006b: 37)—for example, word order; (2) a textually dominated translation, while still likely to exhibit “[i]nterference of the model underlying the source text … in terms of its literary-specific features,” is “well-formed in terms of general conventions of text formation pertinent to the target culture” (Toury 1995: 171) and manifests thematic coherence and formal cohesion that is characteristic of the kind of text it is structured to be (Boyd-Taylor 2006b: 38– 9)—for example, hypotaxis replaces parataxis; (3) a literary translation is well formed “from the point of view of the literary requirements of the recipient culture, at various possible costs in terms of the reconstruction of features of the source text” (Toury 1995: 171). Septuagint Genesis falls within the range of linguistically motivated and textually dominated modes of translation, and thus would not be characterized as literary in style, contrary to what the likes of Aristeas, Philo, and Epiphanius would have us believe. The question as to what slot within the recipient culture it would have occupied in terms of how it would have functioned remains open. The possibility that, because it “aimed at bringing the reader to the Hebrew original rather than bringing the Hebrew original to the reader,” it might have served as an educational tool for those whose knowledge of Hebrew would have been limited has been mooted (BoydTaylor 1998: 71–105; Pietersma 2002: 344–50; see also Brock 1972: 11–36). But without positive evidence, that remains only a possibility.

Principle 4 The fourth basic principle upon which the SBLCS commentary series has been established declares: 4.0. The text-as-produced is the act of a historical agent—the translator—and should be described with reference to the translator’s intentions, to the extent that these are evident. Once again, given the lack of solid evidence regarding the circumstances surrounding the translation of the Hebrew Pentateuch in general, and the book of Genesis in particular, one must undertake a careful and systematic analysis of the translation document itself

360

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

to ascertain what a translator’s intentions may have been with respect to the act of producing meaningful communication. The process of such an investigation is described in the following two sub-points. 4.1. The meaning of the text is best understood as encompassing both what the translator did and why. The OG translator of Genesis, as has been shown in the examples discussed above, adopted a variety of strategies in rendering the Hebrew source text. The strategies exhibited in these examples range from transcription of Hebrew terms, to adoption of stereotypical translation equivalences, to reflection of Hebrew idiom, to replication of Hebrew wordplay, to formulation of replacement renderings that conform to Greek idiom. Furthermore, there are clues throughout OG Genesis that point to the location and temporal context of the translator—an individual of Jewish ethnicity living in Ptolemaic Egypt. Why the translator adopted a variegated approach to rendering the Hebrew source text remains, as indicated above, something of a mystery. For the most part, the translator exhibits awareness of the meaning of his source text, though on rare occasions its meaning may have escaped him, as appears to be the case in the following example: Gen. 28:19

‫ואולם לוז ׁשם העיר לראׁשנה‬ But the name of the city was Luz at the first (NRSV) καὶ Οὐλαμλοὺζ ἦν ὄνομα τῇ πόλει τὸ πρότερον And the city’s name was formerly Oulamlouz (NETS)

In this case, the translator seems not to have been aware that ‫ אולם‬here is the adversative conjunction denoting “but” (DCH, ad loc. ‫ )אּולָם‬and to have assumed that it was part of the name of the city that subsequently came to be known as Bethel. Such exceptions aside, the translator was evidently quite intentional in the translation choices that he made. Consequently, one must conclude that, for whatever reason, sometimes the translator intended to bring the Greek reader to the Hebrew original and other times it was his desire to bring the Hebrew original to the Greek reader. 4.2. The commentator’s task thus includes the following: (a) to search out the intention of the translator insofar as this may be inferred from the transformation of the source text and the verbal make-up of the target text; (b) to describe the possibilities deliberately marked out by the language of the text. The second sub-point of the fourth principle speaks to the SBLCS commentator’s task in both determining the OG translator’s intention through an investigation of the process whereby the meaning of the source text was reconfigured in the target text, and doing so in the light of the range of possible outcomes for that process. The matter of discerning the translator’s intention has been addressed above, so I will focus on the second part of the commentator’s task having to do with identifying translation possibilities. One of the ways to illustrate this is by observing the different ways that the translator has rendered a given Hebrew lexeme or construction. Examples discussed above include the various renderings of the ‫ ב‬+ infinitive construct, such as ἐν + the articular infinitive, ὅτε + finite

THE SEPTUAGINT: THE TEXT AS PRODUCED 361

verb, and ἡνίκα + finite verb. An investigation of the various contexts in which these occur reveals that the first of these Greek constructions was employed by the translator when attendant or accompanying circumstances in general were described, whereas the latter two were used when temporal context specifically was being marked. Similarly, the translator decided to employ ἐνταφιάζω “prepare for burial” and θάπτω “honour with funeral rites” to render ‫“ טנח‬embalm” rather than terms such as ταριχεύω or σκελετεύω that would have been more explicit in terms of signifying the mummification of a corpse. These translation choices were, therefore, not arbitrary in the sense of creating different equivalences simply for the sake of stylistic variation. They then are indicators as to what the translator’s specific intentions appear to have been.

CONCLUSION The aim of the present chapter has been to discuss the foundational principles that guide the work of those involved in the Society of Biblical Literature Commentary on the Septuagint project. This undertaking will result in the publication of a series of volumes on the books that contain Old Greek version of the Jewish Scriptures. Commentators will seek to elucidate the meaning of the text-as-produced at the point of its inception, rather than the meanings that would have been perceived by subsequent readers, the great majority of whom would have had no working knowledge of the Semitic source texts of those books of the corpus that were translated. As has been shown above, the translator of the book of Genesis employed a range of interpretive strategies, an investigation of which provides insight into both the communicative intent at any given point in the translation and the context in which it was produced.

RECOMMENDED RESOURCES 1. Boyd-Taylor (2001). In this essay, it is argued that the creation of a comprehensive lexicon of the Septuagint (as opposed to specialized lexica for the Septuagint) is problematic because such an undertaking is likely to involve the category error of treating this corpus as though it were compositional literature, when in fact the prevalence of linguistic interference from the Semitic source text means that the product of this translation enterprise cannot be relied upon for lexicographical evidence as to conventional Greek usage of that period. 2. Boyd-Taylor (2011). Drawing upon the discipline of Descriptive Translation Studies, the author outlines a conceptual framework within which the so-called interlinear paradigm regarding the translation of the Septuagint is critically assessed, provides an analysis of selected Septuagint texts, and highlights the normative features of the work of the respective translators. 3. Hiebert (2000). The argument of this essay is that the interlinear model of translation satisfactorily accounts for the shifting dynamics of the Hebrew–Greek relationship that are reflected in the Septuagint text of Genesis, and that it provides a useful frame of reference for executing the NETS translation. 4. Hiebert (2010). This volume contains a collection of essays that address topics such as the nature and function of the Septuagint, the distinction between the Septuagint as produced and the Septuagint as it came to be received, and the issues involved in preparing translations of the Septuagint into English, French, and German that have recently been published.

362

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

5. Pietersma (2006c). This essay highlights the importance of distinguishing the Septuagint as produced from the Septuagint as received, and argues that the discipline of Descriptive Translation Studies provides a new Archimedean point for Septuagint research. 6. Pietersma (2008). As the title of this essay makes clear, the author’s purpose is to emphasize the distinction between text-production and text-reception, and he probes the exegetical and hermeneutical implications of such a distinction as it bears on the task of writing a commentary by providing a case study involving Psalm 8 and its interpretation in the New Testament book of Hebrews. 7. Wright (2006). In this essay, the author compares the claims about the origins and nature of the Septuagint text that are made in the Letter of Aristeas with the actual textual-linguistic makeup of the Septuagint, and concludes that the Letter provides little, if any, reliable information regarding these matters, but instead is a foundation myth created at a point when the Septuagint came to be regarded as sacred Scripture in order to function as an apologia for its authoritativeness. 8. Wright (2008). The author discusses and critically assesses the different perspectives and methodologies reflected in the recently published English, French, and German translations of the Septuagint.

CHAPTER 24

The Septuagint: A GreekText-Oriented Approach STANLEY E. PORTER

INTRODUCTION The recent growth in Septuagint studies inevitably raises hermeneutical questions. These hermeneutical questions have only relatively recently been brought to the fore within the wider field of biblical studies: first in the rise of various forms of literary theory and interpretation and then in appreciation of the questions raised by philosophical hermeneutics and even more recently in the perspectives of various methods within Translation Studies (see Porter and Robinson 2011). This order of their emergence within biblical studies does not necessarily reflect their significance, but each has bearing on Septuagint studies. Each of these areas has also brought awareness to biblical studies of the complexities of interpretation. Philosophical hermeneutics raises questions about the nature of human understanding, the role of the self in interpretation, and the place of presuppositions. Literary criticism asks questions about the focus of interpretation and differentiates between synchronic and diachronic understanding. Translation Studies expands the horizons beyond simply thinking in terms of literal versus interpretive translation, and asks about the goals, intentions, and practices of translation. As a result of these studies, major questions have emerged regarding the relationship between the author, the reader, and the text, the three major locations of interpretive authority. Changing relations among these entities have shifted the focus from the author to the text to the reader as the arbiter of meaning, with a resulting reassessment of the implications of such foci. Author-oriented readings typically emphasize the historical background behind the text; text-oriented readings usually emphasize the importance of a close reading of the text itself and its internal dynamics; and reader-oriented readings typically emphasize the role of the reader in the process of meaning making (Porter and Stovell 2012: 12–20). With the development of several commentary series for the Septuagint, hermeneutical questions must also be raised about the interpretive stances of such commentaries. This chapter defends one such hermeneutical approach, namely, a Greek-text-oriented Septuagint commentary series, currently in press as the Brill Septuagint Commentary Series (cf. Hess 2016). This approach to commentary writing, and hence interpretation of the Septuagint, is sometimes referred to as a commentary on the text as received.1 I do not wholeheartedly endorse this characterization, as it has been formulated by those who have opposed such an approach. See note 5 below.

1

364

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

In other words, in the triad of loci of authority mentioned above, such a commentary focuses upon the text of the Septuagint (not the preceding or underlying Hebrew text) as the primary object of interpretation.

THE HISTORY OF THE BRILL SEPTUAGINT COMMENTARY SERIES (SEPT) Septuagint studies have recently become a growth industry, reflected in numerous collections of essays, monographs, articles of various sorts, and now, some might say, at last, commentaries. Whereas the fields of Old Testament and certainly New Testament studies have been overwhelmed with commentaries, the field of Septuagint commentary writing is apparently still relatively sedate.2 The French have their La Bible d’Alexandrie (BdA), announced in 1980, instigated in 1986 and edited by Marguerite Harl, Gilles Dorival and Olivier Munnich in twenty-three volumes (and parts within these), with some books still to appear. The Germans have their Septuaginta Deutsch (LXX.D), a project completed in true German fashion from 1999 to 2011 and edited by Wolfgang Kraus and Martin Karrer, with a translation and a two-volume commentary. The English-speaking world also has a commentary series on the Septuagint, the Brill Septuagint Commentary Series (SEPT or sometimes BSCS), planned in the mid- to late-1990s, inaugurated in 2004 and edited by Stanley Porter, Richard Hess, and John Jarick, with fifteen volumes having appeared to date and more being written (announced first in Porter and Pearson 1997: esp. 546). This commentary series takes a Greek-text-oriented approach to commenting upon the Septuagint, a definition I will return to below. Sometime late last century, a group of scholars, often linked with the International Organization for Septuagint and Cognate Studies, and now in conjunction with the Society of Biblical Literature (as announced in 2005), announced a new commentary series.3 This same group published in 2007 the New English Translation of the Septuagint (NETS)4 based upon the New Revised Standard Version (Pietersma and Wright 2007: xiii–xx, esp. xiv–xvi, xix–xx [the Psalms first appeared in 2000]).5 No volume has yet been published in the series (for samples, see the chapters in Büchner 2017). Despite opposition by some now identified with the SBLCS, the Brill Septuagint Commentary Series continues to produce volumes at regular intervals.6 Much of this opposition was

Information on the following projects I have found on various websites and in Jobes and Silva (2015: 357–9, 362–3). I note that in their first edition (2000) the Brill Septuagint Commentary Series was included (though not by name) only in a footnote (312 n. 7), while the IOSCS project made it into the main text, even if a single commentary had not appeared. 3 This was around 1995/1996, earlier (I believe) than when the Brill Septuagint Commentary Series was first proposed. See Pietersma (2017: 1–2); for a formal announcement, see Pietersma (1998). 4 I note that Albert Pietersma published in 1996 his Translation Manual, where the foundations of the interlinear paradigm were laid (beginning in 1995). 5 The NETS translation distinguishes between the text as produced and the text as received (Brill Septuagint Commentary Series). See Pietersma and Wright (2007: xvi), cf. Pietersma (2017: 2–9). Pietersma has recently chopped logic over this matter. He wants the term LXX “to mean the LXX qua translation, that is to say, the LXX as an entity standing in an immediate relationship to its source text, in distinction from the LXX, cut loose from its historical moorings, and thus the LXX as a free-standing entity, in short, the LXX qua translation in distinction from the LXX qua text.” See Pietersma (2015: 165–76, quoting 165). 6 Nevertheless, only one Brill Septuagint Commentary Series author (the editor) is currently identifiable as a contributor to the recent Aitken (2015b), with many more of the contributors identified with the IOSCS/SBL/ NETS group (and a couple with La Bible d’Alexandrie). Only Clayton Croy’s Greek text of 3 Maccabees is cited of all of the editions and translations of the SEPT volumes (though many are included within the bibliographies). 2

THE SEPTUAGINT: A GREEK-TEXT-ORIENTED APPROACH 365

apparently motivated by the belief that there is only one proper approach to commenting upon the Septuagint, the one endorsed by the NETS project based upon the “interlinear paradigm,”7 which implies that the Septuagint translation was an attempt to provide an aid to the understanding of the Hebrew text through the Greek, even if such language is taken metaphorically (although one might wonder what that metaphor means).8 In many ways, the Brill Septuagint Commentary Series is closer to La Bible d’Alexandrie than it is to the Septuaginta Deutsch and the purported approach of the SBLCS commentary. The Brill Septuagint Commentary Series is a literary commentary on the Greek text of the Septuagint, with the Greek text of the individual biblical books treated in their own right and with their own integrity, structure, and compositional shape, that is, as each was “received” in Greek by its early users (at least as evidenced by the codex manuscript tradition) (see Hengel with Deines 2002: 57–74). The Hebrew text is to be referred to only where appropriate for understanding or clarification, because the commentary is not meant as a repository of bilingual data, and the Greek text is not seen as nonstandard Greek rendered in interlinear fashion (as it is in the proposed interlinear commentary; see Pietersma 1998: 44). Each commentator on a book selects one of the early codex editions of the Greek text for commentary, using a single manuscript text rather than an eclectic one, and thereby focuses upon a text that actually existed in an ancient reading community.

DESCRIPTION OF A GREEK-TEXT-ORIENTED SEPTUAGINT COMMENTARY SERIES The Purpose of a Textually Oriented Biblical Commentary I begin this description with a brief discussion of the larger question of the purpose of a biblical commentary. The biblical commentary has been transformed over the years into the current attempt at being a comprehensive repository of all theological and other knowledge. As a result, the modern biblical commentary has been unjustifiably

This is apparently a “paradigm” established by Albert Pietersma (1996), elucidated perhaps most fully in Pietersma (2002). The key features of this approach are that it assumes the “translationese” of the Septuagint (in fact, the position claims that the Septuagint is often not even Greek but Hebrew essentially in Greek characters), it uses the Hebrew text to interpret the Greek, and it defends the Greek as bad as it is as being rendered for a particular educational purpose (see Pietersma and Wright 2007: xiv–xv, for these characteristics). All of these assumptions are highly questionable and remind me of various errant views promoted early last century regarding the Greek of the New Testament. In fact, I believe that the assumptions are indefensible regarding the Septuagint as a translation (e.g., what about books written in Greek? was the approach literalistic throughout?), the nature of translation (e.g., its communicative purpose), the relationship between the Hebrew and Greek text, and its purpose (endorsing a narrow view of the notion of the Old Testament, that is, that it is Hebrew). Boyd-Taylor (1998: 71–105) has proposed the term “metaphrase” to describe the literalistic translation. More recently, BoydTaylor (2011) has attempted to ground the interlinear paradigm in Descriptive Translation Studies (Toury 1995). This strikes me as a practice in search of a theory. However, it has been used as the basis of Wagner (2013), who attempts to extend Boyd-Taylor’s type of analysis, but who also neglects the Brill Septuagint Commentary Series, despite his date of publication. 8 Though Dines (2004: 52–4) offers some defense of the interlinear approach, she also raises major questions, such as the priority of the Hebrew text, the nature of bilingualism in the third to second centuries BCE, and problems with the MT as point of comparison. For recent, significant critiques of the “interlinear paradigm,” see Joosten (2008a) and Schaper (2014: esp. 34–9). A volume in many ways dedicated to disputing the interlinear paradigm is Bons and Kraus (2011). 7

366

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

exalted to the position of being considered (at least by some) the pinnacle of scholarship. The commentary’s marketability, along with its exalted status that has threatened the monograph and even informed article, has compelled the development of what appears at first to be a variety of different commentary types, all created in an attempt to capture and satisfy various audiences. At the end of the day, however, there are essentially only two major types of commentaries: varied permutations on the historical-critical commentary and those that are variously represented as practical, applied, or narrowly theological in nature. As some recent study has shown, the modern commentary arguably has sacrificed its ability to comment trenchantly, using the best linguistic resources available,9 on the language of the text itself, and become an inflated repository of the observations of previous commentators (Porter 2013a, 2013c). The origin of the “modern” biblical commentary is closely tied to the classical commentary, as a product of the Renaissance and Reformation revival of learning.10 With the revival of interest in ancient and “foreign” languages came a need for commentators to clarify and explicate the language of the text. This approach usually took the form of direct commentary upon the lexical and syntactical features of the original language, supplemented by enlightening parallels from contemporary literature. This kind of commentary is seen in works by J. A. Bengel (1687–1752), whose Gnomon is a textfocused commentary in several volumes on the entire New Testament, and Jakob Wettstein (1693–1754), whose two volumes include numerous parallels from extrabiblical literature meant to illuminate the text (Baird 1992–2013: 1:74-80, 103-7). In many ways, the height of commentary achievement was reached in J. B. Lightfoot’s (1828–89) commentaries on the New Testament, especially on some Pauline epistles (Baird 1992–2013: 2:66-70).11 Commentaries were written to aid interpretation of the language of the text under consideration, a worthwhile goal that arguably should be retained in any commentary writing. What came to characterize the best examples of commentary writing were clarifying observations upon the language of the text being considered, based upon the best understanding of the language available at the time (which meant traditional grammar during the heyday of much commentary writing, and should mean linguistically informed language models being used today), with pertinent parallels observed in extra-biblical literature and occasional supplementary comments on particular issues of importance. Such an approach to commentary writing should be represented in contemporary commentary writing, including on the Septuagint. There are four reasons for this. The first reason for such an approach is related to the nature of commentary writing itself. Even though the modern biblical commentary has gone awry and has mired itself in an almost inextricable morass, the situation is not the same for Septuagint commentary writing. A Greek-text-oriented commentary on the Septuagint not only is one of several possible types of commentary (and should be valid simply because of this), but is more consonant with what a commentary is meant to accomplish. The Septuagint is a Greek text, and therefore merits comment on it as a Greek text. I address the question of This is certainly the case in those who represent the NETS approach (Pietersma and Wright 2007), who draw upon the very limited grammar of Conybeare and Stock (1905) to characterize the Greek of the Septuagint (a view not informed by the knowledge of the documentary papyri and reflecting a skewed view of the Greek of the Bible) and do not apparently know (or at least do not cite) the work of Evans 2001. 10 For my expanded comments on the origins and development of the modern commentary, see Porter 2016a. 11 Lightfoot wrote parts of other commentaries, including some on other Pauline letters, Acts, John’s Gospel, and 1 Peter. 9

THE SEPTUAGINT: A GREEK-TEXT-ORIENTED APPROACH 367

translation below, but wish here to address two further sub-issues. The first is whether commenting upon the Septuagint is like commenting upon the English text of the Bible. We may smirk when we consider this analogy. However, the vast majority of biblical commentary writing done today, even sometimes by those purportedly commenting upon the Hebrew or Greek text, is far more dependent upon English than many would want to admit. However, the Septuagint is different from simply the English translation of the Hebrew or Greek Bible, as is evidenced in its use in the New Testament and its role in the Orthodox church tradition—an issue I address further below. A second sub-issue is that one of the reasons that the interlinear paradigm of the SBLCS may have some attraction is because of what commentary writing has become. In many instances, in both Old Testament and New Testament studies, commentary writing has become less about the text as found within the book, and more about what is posited to exist behind the text of the book at hand.12 We have Old Testament and New Testament commentaries that are exercises in source criticism, at least as much if not more than they are commentaries on the text. For example, one can observe how George Foot Moore’s (1895) commentary is more about how the disparate sources of Judges are woven together in the narrative or how Willoughby Allen’s (1907) commentary is about how Matthew purportedly made use of his two primary sources, Mark and Q. The interlinear model becomes just such an exercise, especially as it has advertised itself as similar to the commentary style found in John Wevers’ Notes on the books of the Pentateuch (see Pietersma 1998: 43, referring to Wevers). The second reason for such a Greek-text-oriented approach is that it apparently reflects how the Septuagint was used within ancient Judaism, as well as within early Christianity. Whatever the historical origins of the Septuagint (and the stories are intriguing),13 the translation was probably instigated within the Egyptian Jewish community. The reasons for this may have been complex, but almost assuredly included the fact that there were significant numbers of Jews who knew Greek, the lingua franca of the Hellenistic and Greco-Roman world, and who were not as competent in Hebrew (as must be maintained by the interlinear paradigm, hence justifying the interlinear approach to give guidance). There is good reason to believe that what may have begun in Egypt spread throughout diaspora Jewish communities, so that the Septuagint effectively became the operative, even if not official (which process came later in any case), sacred text for Jews throughout the Mediterranean world (Archer and Chirichigno [1983] 2005: ix). The result was that a diaspora Jew such as Paul would have had knowledge of the Septuagint and used it as a, if not the, version of his Scriptures—at least he used it as one that was definitive enough for him to cite in his writings as having divine authority. The same situation was also the case even for some Jews within Palestine, as evidenced by use of the Septuagint by some of the New Testament writers who probably originated in Palestine (e.g., Matthew’s Gospel). I would even go so far as to posit that the evidence from the Gospels indicates that Jesus himself may well have used, at least on occasion, the Septuagint when he wished to cite Scripture.14

If we use the triad of behind, within, and before the text, the focus should be upon “within” the text, when it is often upon “behind” the text. 13 See Müller 1996: 46–67; Dines 2004: 27–39; Wasserstein and Wasserstein 2006: 19–50; Law 2013: 33–42; Jobes and Silva 2015: 17–24. On the origins of the Septuagint, see Aitken in the present volume. 14 On the use of the Septuagint in the New Testament, see Moyise in the present volume. 12

368

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

A third reason for such an approach is that a Greek-text-oriented commentary is the only one that does justice to the fact that the Septuagint is not simply a translated document, and certainly not simply an interlinear translated document. There are four sub-dimensions to this argument. The first is that some of the books of the Septuagint were originally written in Greek and hence cannot be considered translations. These books would probably include 2, 3, and 4 Maccabees, Wisdom of Solomon, and possibly the Letter of Jeremiah. (This is also a problem for the interlinear paradigm, when the book being commented upon is not an interlinear rendition at all, regardless of the metaphorical extensions of the terminology.) The second sub-dimension to this argument is that, whether we use the term “translation technique” or literal versus free or some other way of describing it (see Dines 2004: 117–28), there are recognizable differences in the language found within the books of the Septuagint. There are instances of what has variously been described as literary Greek, literary Atticistic Greek, good Greek, indifferent Greek, vernacular Greek, and literal Greek, and not all of these categories easily correlate with translated or originally Greek books.15 As a result, some of the translated books may be characterized as reflecting literary Greek, such as 1 Esdras, Daniel (Old Greek), Esther, Job, and Proverbs, while part of Baruch is even literary Atticistic Greek, and the translated book Tobit is vernacular. All of these translated books require a different interpretation of interlinearity than do those that reflect a different translational achievement. A third sub-dimension is that it might even be arguable that the category of interlinearity is only of relevance—if at all—for a select few Septuagint books, perhaps those seen as literal or even unintelligent Greek (such as parts of Jeremiah, a small part of Baruch, Judges (B text), Ruth, parts of Kingdoms, Song of Solomon, Lamentations, Theodotion Daniel, 2 Esdras, and Ecclesiastes, at most). Since the rest of the books might arguably be representative of various types of good, literary, vernacular, or even indifferent Greek, the interlinear model would not be appropriate, as the assumption of the literalness or unintelligence of the translation is not maintained. A fourth sub-dimension is that there are significant differences between the Hebrew and Greek versions of some of the books even where we think that they are translations (to say nothing of the use of the MT in this discussion). This variation characterizes Jeremiah, Job, and Daniel with its several additions not found in Hebrew and its other major version (Theodotion), and other books with several Greek versions such as Judges, Esther, Tobit, and portions of Chronicles. This raises not only the question of what is meant by a translation but the question of how helpful a nonexistent or radically different text can be for interpretation of the Greek text. In all of these instances a Greek-text-oriented commentary is able to address the question by commenting directly upon the Greek text as it is found in the particular book, without any assumption regarding the nature of the Greek or the relationship that it has to a real or imagined or never-existent Hebrew version. The fourth reason for such a Greek-text-oriented approach is that we do not know all that we wish we knew about the Greek of the Septuagint. There is sometimes the

I use the categories found in Thackeray (1909: 13). I realize that there has been further work on translation of the Septuagint (besides the interlinear paradigm), but have found Thackeray still the most helpful categorization, as he avoids dichotomizing the distinctions. See also Barr (1979) and van der Louw (2007).

15

THE SEPTUAGINT: A GREEK-TEXT-ORIENTED APPROACH 369

unfortunate assumption that, because we have had access to the Greek (and the assumption is less true of Hebrew) documents for so long, and have studied them using traditional grammar, we in fact know all that there is to know about them as Greek texts. We only need to remind ourselves of how much insight continues to be gained through study of the documentary papyri, or to realize the fundamental shift in perspective caused by the advent of modern linguistics, to realize that there have been significant developments and that we know far less than we could. We are indeed gaining new insights into ancient Greek all the time, because we are advancing in our understanding of linguistic frameworks for analyzing Greek and hence growing in our knowledge of the contribution that they make to our understanding. This development is reflected in the various theories regarding the nature of the Greek of the Septuagint. Similar in fashion to theories about the Greek of the New Testament, discussion of the type of Greek found in the Septuagint has had four major movements: (1) The Semitic Greek hypothesis, which reflected the truncated knowledge of Koine Greek (and is reflected in Conybeare and Stock’s (1905) grammar, and in some ways the interlinear paradigm); (2) The Hellenistic/Koine Greek hypothesis, which arose after the discovery of the documentary papyri and its broadening knowledge of the history and development of Greek (and is reflected in the work of Adolf Deissmann, James Hope Moulton, H. A. A. Kennedy and Albert Thumb, and the grammars of Robert Helbing, Henry St. John Thackeray and Félix-Marie Abel); (3) The Hebraic/Jewish-Greek hypothesis, which argued for a particular type of Jewish-Greek dialect (and is reflected in the work of Henry Gehman, Nigel Turner, Georg Walser, and in some ways Jan Joosten, in the interlinear paradigm, and Ilmari Soisalon-Soininen and his followers in the so-called Finnish School); and (4) The revival of the Koine Greek hypothesis, which has brought modern linguistic principles to bear on examination of the language and translation of the Septuagint (and is reflected in the work of Moisés Silva, John A. L. Lee, Harl, Dorval and Munnich, Julio Trebolle Barrera, Trevor Evans, James K. Aitken, and Takamitsu Muraoka, among others).16 After repeated aborted attempts to write a grammar of the Greek of the Septuagint by such luminaries as Thackeray (1909), Helbing (1907), Cignelli and Pierri (2003–) and others, we finally have such a product produced by Muraoka (2016).17 Whatever one thinks of this grammar, this is only a start. A Greek-text-oriented commentary presents a useful and, indeed, essential perspective that can help to encourage further scholarship on ancient Koine Greek.

This topic, with references to those who are cited in the above summary, is discussed in more detail in Porter 2016b: 15–38. Other matters of language in the Septuagint are discussed in Bons and Joosten (2016). 17 Despite many strengths, my major critique of Muraoka (2016) is that he does not seem to have an explicit linguistic theory and relies too much upon traditional grammars. He also is much too given simply to attempt to list uses rather than explain functions and semantics. There have been other important works along the way in the form of individualized studies. 16

370

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

The Implications of a Greek-Text-Oriented Septuagint Commentary and Its Reception A commentary on the Septuagint should ideally be no different from a commentary on any other biblical book. A commentary—or at least a commentary as it was originally conceived and as I have argued should remain—is to be an exposition of the text under consideration, with primary emphasis upon understanding the language in which this text is written. There are several implications of this definition in relation to the Septuagint. The first implication is that a Greek-text-oriented commentary should clearly focus upon commenting upon the Greek of the Septuagint text. As a result, there is a tremendous amount of latitude for debate about the relationship of the Greek text to the Hebrew text on such matters as types of translation, relationship between the textual versions, the purpose of the translation (whether for linguistic, educational, religious, or other purposes), and the like (e.g. McLay 2003: 116–33; Dines 2004: 47–61). But all of that is relatively incidental to performing the task of commenting upon the Greek text of the Septuagint. Rather than discussing the Greek text in relationship to a single paradigm (interlinear), or in relation to a set of categories (such as free, literary, indifferent, literal and the like), I believe that one should make observations upon the Greek text in relation to a number of other factors. These would primarily include factors related to where on a cline of standard to nonstandard Greek the particular text or passage or construction would fall. I realize that the assumption is that the original translator or author or reader understood the sense of the passage, at least to some extent and in some way, but this is an assumption shared by all interpreters of a text, until that point when no sense can be made of the text (see Wevers 1990: xv).18 The second implication is the recognition of the simple fact that the Hebrew Bible is not the Septuagint, so a commentary on the Septuagint must be upon the Greek text and not upon the Hebrew text, nor upon the Greek text as an aid to understanding the Hebrew text, nor as a window into the construction or reconstruction of the Hebrew text, nor as a guide to understanding the Hebrew text, nor as a reflection of and hence an object necessarily to be interpreted by the Hebrew text—some if not all of the potential things that an interlinear commentary might imply. There is very little point in labeling a commentary a Septuagint commentary if it is not primarily and essentially a commentary on the Greek text and not the Hebrew. Does that mean that there is no place for commentary upon the Hebrew text? First, I would not ever want to say that only one kind of commentary can be written. However, I do believe that some types of commentaries come closer to fulfilling the purpose for which a commentary is to be written than do others. Others can undertake other kinds of commentaries—and they will continue to do so, no doubt, in the wider field of biblical studies—and there is a place for them, so long as we do not forget that the Greek text of the collection of books that we call the Septuagint should remain central to a Septuagint commentary. However, we may well also agree that exercises in Greek and Hebrew textual comparison may not be appropriately called commentaries at all. I note that Wevers’ original works were not called commentaries but were called Notes, although perhaps mistakenly notes on the Greek Text of Exodus, Genesis, Deuteronomy, Wevers (1998: ix–x) claims to be hard-pressed to maintain the same perspective in his commentary on Numbers, but many of the problematic examples that he cites seem to be explainable if the Greek text is the point of focus.

18

THE SEPTUAGINT: A GREEK-TEXT-ORIENTED APPROACH 371

Leviticus or Numbers, when much of their notation is about the Hebrew text (and they are pretty much confined to lower linguistic levels of comment). I cannot help but think that these works have influenced the notion of the interlinear paradigm, as Wevers treats his examination of the Greek and Hebrew texts as a series of virtually word-for-word equivalents, with one informing the other. The third implication—and one that moves beyond much of what has been written previously insofar as commentary-related material is concerned—is that a Greek-textoriented Septuagint commentary should appreciate the literary elements of the text. Attention to the literary elements means that one must go beyond the interlinear features, that is, the word for word equivalency (or not), and deal with larger elements of the discourse. This potentially means attention to the type of text that one is commenting upon, its macro-structure and its micro-structure, the construction of its individual literary units, and how these units relate to each other, along with other standard types of literary questions (setting, character, point of view, etc.). As a result, there may well be a place for discussion of questions related to what is usually called genre or register (if they are distinguished), how texts are externally and internally organized, the various linguistic means by which such texts are created, and the more particular clausal features that are used to characterize the various participants, characters, and other elements of the text. When one speaks about a Greek-text-oriented commentary one must think in relation to a particular text or set of texts, in this instance the individual books of the Septuagint. This approach then by definition introduces consideration of various literary features of that text. I realize that some may call these features by varying names, some of them related to text linguistics or discourse analysis, others to rhetoric, and still others to something else altogether. However, they all have in common that the text is not simply a series of discrete elements that happen to be strung together (sometimes in a confusing way), but all of the individual elements function at various levels within the text to create the discourse, and all of these levels and their elements merit commentary. The fourth implication is that one must consider how the individual Septuagint texts were read and functioned and understood by their “original” audiences and users. I put the term “original” in quotation marks, because defining this term is a very difficult task for interpretation of any ancient text. This understanding is especially difficult for the Septuagint, with its obscure origins and purposes, its developing versions, and its lack of specific historical information regarding its use (see Jobes and Silva 2015: 308–25; Dines 2004: 81–107). Nevertheless, for all commentaries, there must be a fixed moment in the textual tradition that forms the historical basis of comment. This kind of problem is sometimes addressed by linguists in terms of register—typical situations that limit the kinds of choices that are made in the use of a language system. In this sense, it is perhaps easier for some approaches to the Septuagint to utilize a more lock-step or literalistic approach, as they do not need to consider such questions regarding the original function and understanding of the text, but one merely posits the correlation of individual elements within two relatively static texts. However, there is a hidden assumption in so doing that we understand how these individual elements, whether Hebrew at the time of composition (whenever that may have been) or at the time of translation or at another significant point, or whether Greek at the time of translation or composition or another point, were meant to function. Register is a powerful sociolinguistic concept to help to determine the parameters of such an understanding, and it is equally important in commenting upon the Septuagint. The history of the Septuagint is notoriously difficult to reconstruct, and along with it attempts to reconstruct the contexts of particular books. However, if we use the

372

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

notion of register as a way to typify a situational context, then we perhaps have a way of narrowing those situational parameters when we focus upon individual books. We may still wish to differentiate how the Septuagint, or at least individual books within it, was read early on by Jews or by Christians, but we may also posit that this would have been similar, as the Septuagint functioned as a sacred text for several religious communities. In other words, we are treating the Septuagint as a religious text that would have been meaningful to both Jews and Christians for a significant period of time. We may wish to distinguish further by positing how the Septuagint was read from the period of roughly the second century BCE to the beginning of the first century CE by Jews and, for some time, by Jews and Christians, and then how it was read by Christians up to the fourth or fifth centuries (or at least how it was read as we find it in the early codices). We probably cannot get more specific than that, but we perhaps do not need to in order to comment upon a text that functioned as a literary religious text for many devout followers of the God of Israel, whether in the synagogue or in the early church.

The Canonical Implications of a Greek-Text-Oriented Septuagint Commentary The focus upon the Greek text, as already mentioned above, has implications for the notion of the canon of the Old Testament. In the above comments, I raised the issue of the analogy of commenting upon the English text of the Bible as roughly equivalent to commenting upon the Septuagint as a canonical text. I also dismissed this analogy by deferring to subsequent discussion here. A possible equivalent to the “canonical” English text being a suitable parallel to the Septuagint is found in those who venerate the King James Bible. For some within those circles, they believe not only that the King James Bible preserves the truer text of the Bible than that found in more recent translations, or even than in earlier Greek and Hebrew manuscripts, but that there was a process of divine preservation of the text and superintendence over the translation. However, despite the use of this translation in some circles, and even effective canonization of it within some small groups of followers, there is no major worldwide religious group that believes that the King James Bible is canonical so far as I know (Carson 1979). There are at least three major arguments for the Septuagint that raise the possibility that the Greek text should be considered canonical or at least canonical-like and hence merit, on that basis alone, commentary on the Greek text as primary text (for treatment of some of the issues, see Crisp 2016; Glenny 2016).19 The first argument is that at the time the Greek Pentateuch was originally translated it was recognized as having canonical status. Commenting upon the translation of the Greek Pentateuch, Wevers (1990: xiv–xv) writes, “The translators realized that their product was itself God’s word; it was declared canonical, and presumably served as the synagogal Scriptures in Alexandria”20 (although we must recognize that there were other centers of Judaism in Egypt besides Alexandria). He notes that the Letter of Aristeas (§§308–11) states that, when the Greek Pentateuch was read out to them, those in authority among the Jews declared of the translation that it was “beautiful and holy” (καλῶς καὶ ὁσίως), along with having been done “accurately” (ἠκριβωμένως). They also pronounced a curse

On the Septuagint and the issue of canon, see Meade in the present volume. Cf. Dines (2004: 62), who does not seem to agree, but she traces the later reception (63–79). See Hengel with Deines (2002: 25–56) and Müller (1996: 68–78) for summary of the opinions of some early church fathers who apparently believed that the Septuagint was inspired in some sense.

19 20

THE SEPTUAGINT: A GREEK-TEXT-ORIENTED APPROACH 373

upon anyone who would alter it by either adding or changing anything from the entirety of what had been written or who would abbreviate it (εἴ τις διασκευάσει προστιθεὶς ἢ μεταφέρων τι τὸ σύνολον τῶν γεγραμμένων ἢ ποιούμενος ἀφαίρεσιν). This statement admittedly does not include the books translated later, and arguably under very different circumstances (whatever the original circumstances were, that is, whether by Greek or Jewish instigation and support). However, it does provide some encouragement for these later translations and their inclusion as having the same status as the earliest books, at least as this is evidenced by their later use along with the original books.21 The second argument for the Septuagint is that the Greek text was apparently widely used by the apostles in what have come to be canonical New Testament texts. In fact, Jennifer Dines (2004: 142–3) states, “It is clear that for all NT authors the Greek texts constitute authentic and authoritative Scripture.” There has been much discussion regarding whether the Hebrew or Greek texts, or some other texts altogether, were used by the New Testament authors in individual instances, especially as the Septuagint was in the process of revision during this time (Jobes and Silva 2015: 209). Nevertheless, it is generally conceded that the overwhelming majority of the quotations of the Old Testament in the New Testament are from some version of the Greek Old Testament (recognizing that there were versional differences in the first century).22 However, I believe that it has been very well established that Paul, whatever other text or texts he may have used, at least used the Greek Old Testament or a form of it in most instances. There is also widespread support for the fact that the Gospel authors used the Greek Old Testament, especially Matthew in his quotations of Jesus (along with the other citations of Jesus). The use of the Greek Old Testament in some of the Gospel accounts has even led some to speculate that Jesus himself, since he almost assuredly knew Greek and had occasion to use it in his teaching, may also have known and used the Greek Old Testament.23 One of the examples sometimes used to support this idea is Jesus’s reading from Isaiah in the Nazareth synagogue in Luke 4, when he cites a passage that departs from the Hebrew Old Testament and appears to create a composite citation with Isa 61:1–2 and 58:6, but includes the statement regarding the blind receiving sight from the Greek rather than Hebrew (Porter 2018: 63–8). It is hard to deny that the authors of the New Testament, and even quite possibly Jesus but certainly Paul, knew and used the Greek Old Testament and thereby confer a certain authority upon it by virtue of their use (no matter what other versions they may have used). The third and final argument is that a major branch of Christianity, the eastern or Orthodox church or churches, has used the Greek Old Testament with its additional books (even if itself translated into various vernaculars and with some variations, such as inclusion of 2 Esdras in the Slavonic Bible, and Psalm 151 and four books of Maccabees by the Greek Orthodox up until recently) as their Old Testament, along with also recognizing

On the Letter of Aristeas, see Wright in the present volume. See, e.g., Harl, Dorival, and Munnich (1988: 274–80, esp. 276), endorsed by Fernández Marcos (2000b: 265); cf. Jobes and Silva (2015: 213–20). For a list of such quotations, see Swete (1900: 381–405), with comment. This is confirmed by a number of individual studies on sub-corpora within the New Testament. See, e.g., Gundry (1967), for both Matthew and Mark, even if Matthew targumizes the Septuagint; New (1993); Freed (1965); Steyn (1995); Stanley (1992); Docherty (2009). 23 See Longenecker (1975: 57–66, esp. 60–1), although he does not fully appreciate the importance of the conclusion by not recognizing the strong possibility, if not probability, that Jesus taught in Greek. I believe that this is not just a possibility, but a virtual certainty. See Porter (2011), drawing upon a number of previously published works. 21 22

374

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

the canonical status of the thirty-nine books of the Hebrew Bible (not that some of these were not without controversy also, such as Esther; Hengel with Deines 2002: 64).24 I realize that there is much discussion and even some controversy among eastern Christians regarding the canon of the Old Testament, with tension over the status of the additional books based on Athanasius’s (296–373) divided opinion on them (Hengel with Deines 2002: 63–4, 125).25 However, it appears, at least to me, that much of that debate is over the canonical status of the books, not over the language in which they appear. Therefore, those circles that recognize the expanded canon of the Greek Bible accept the books found in the Hebrew Bible, but they also wish to accept as canonical the additional books. The issue of language seems to be assumed, with a tacit recognition that the Greek language was at least as canonical a language as the Hebrew, which was restricted to the thirty-nine books found in the Hebrew Bible. This reality is seen by virtue of the fact that the Greek canon has served as the basis of the eastern or Orthodox canons, even though they were early translated into other languages (such as Old Church Slavonic) and even later from the Masoretic text. There are several solutions to this canonical situation. One is to accept the Greek text as the canonical Old Testament, for the three reasons laid out above. The problem with this solution is that recognition of the Greek Old Testament as canonical did not mean that the Hebrew Bible was not also recognized. The Jews and early Christians did not universally use the Greek text, but continued to use the Hebrew as well, as did some of the early Christian writers. The western church, in distinction from the eastern or Orthodox, chose the Jewish canon of the Hebrew Bible as its Old Testament. The second solution is to reject the Greek text as the canonical Old Testament and recognize it simply as a translation and therefore as derivative. The problem with this solution is that it overlooks all of the major arguments presented above. Further, the Septuagint is not simply derivative, but some of the books were written in Greek from the start, and some of them were written in Palestine (1 Esdras and 2 Maccabees), and it continued to be used in Christian circles, if not in Jewish ones. The interlinear approach to the Septuagint could be interpreted as being structured in such a way as to push toward this solution, that is, to have a subtle anti-Greek predisposition that is oriented toward maintaining the Hebrew Bible as canonical. The third solution is to accept that there are two canonical and entwined Old Testaments, one in Hebrew and the other in Greek. These two canonical Old Testaments have the majority of their books in common, even if there are some internal differences, but also differ in overall composition. The eastern or Orthodox church has a larger number of canonical books within its Old Testament. The western tradition agrees with the Jewish canon of the Hebrew Bible, which the east accepts as well. The deciding factor would seem to be based, not simply on the language of the Old Testament, but on theological tradition.26 Therefore, it is not surprising that most western Christians,

On the Septuagint in the Orthodox tradition, see Seleznev in the present volume. Hengel notes that the eastern church continues to differentiate the nature of inspiration between the two groups of books but accepts the additional books as “part of Holy Scripture” (with Deines 2002: 125). 26 A fourth solution is proposed by Hengel that with the coming of the New Testament Christians do not need a closed Old Testament, since the Old Testament is fulfilled in the New. See Hengel with Deines 2002: 125–6. This solution is formulated along more theological lines than the three proposed above, which are based upon historical evidence. See also Glenny 2016. 24

25

THE SEPTUAGINT: A GREEK-TEXT-ORIENTED APPROACH 375

including perhaps those who argue for an interlinear paradigm for its understanding, tend to place priority upon the Hebrew Bible, even though there are those who maintain the Greek Bible as their Old Testament.

Reasons for Commenting upon a Single Greek Manuscript I have perhaps said enough to justify a Greek-text-oriented Septuagint commentary. However, I would go further and argue that such a commentary is perhaps best conceptualized as commentary upon a single manuscript and not an eclectic text. There is a major divide between Hebrew Bible and New Testament textual scholars over the use of single manuscripts or eclectic texts, with Hebrew Bible scholars traditionally opting for the former, and New Testament scholars traditionally opting for the latter. Even with the tendency in Old Testament studies to use a single manuscript tradition, the NETS commentary follows a “principle of original text.” It states that “though for any given book the best available critical edition [by which is meant the Göttingen version where available, and Rahlfs-Hanhart where it is not] will form the basis of interpretation, commentators shall improve upon that text where deemed necessary, and thus assist in the ongoing quest for the pristine Greek text” (Pietersma 1998: 44). I have to admit that I have no idea what the “pristine Greek text” means. However, our knowledge of the origins and circumstances of so many of the books of the Septuagint—such as their relationship to the Hebrew text where such a text exists, whether they were translated at Jewish or Greek initiation, where they were translated, and their development over the course of three centuries—makes such a text-critical claim well-nigh impossible to understand, and even less possible to implement. I believe that it creates both a false sense of what manuscripts are, how they were written, how they can and should be treated, and a false sense of the ease with which they can be interpreted and used in reconstructions of the ancient world. The result of using such categories is probably to convey a false sense of what has been achieved in the process of commenting upon such a manuscript. The tendency of modern editions of the Septuagint is, however, to stick much more closely to the single manuscript tradition—a feature followed by the Brill Septuagint Commentary Series, in which a single manuscript, taken from any one of the three early codex manuscripts, Sinaiticus, Vaticanus or Alexandrinus, is used. Examination of many editions of the Septuagint reveals focus upon these major codices. Swete’s (1887–94) edition of the Septuagint was based almost entirely upon codex Vaticanus; Rahlfs’s edition (1935), which has been a standard for nearly eighty years, used those same three codices (and so perhaps is more eclectic than others), and was revised by Hanhart (2006); and Brook and McLean’s (1906–40) so-called Larger Cambridge Septuagint was based upon Vaticanus, although also using some other manuscripts. The Göttingen edition, however, is truly eclectic, and provides the texts for the NETS translation. There are several reasons why using a single manuscript is to be preferred—or at least to be recognized as having intellectual and text-critical integrity, besides the fact that the history of Septuagint scholarship has recognized the validity of single manuscripts. These reasons are primarily two.27

For reasons to use a single-text in New Testament studies, see Porter (2013b: 72–5), a proposal that I have made previously (see reference on 74 n. 207). See also Evans and Porter (2017).

27

376

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

The first reason is that, whatever we think of eclectic texts, or even those that are based upon a limited number of codex manuscripts but supplemented by others, the single manuscript was an actual text used by an ancient theological community, whereas the standard modern eclectic text was not. As noted several times, we may not know about the origins of the Septuagint, but we do know from the manuscript tradition that the early manuscripts were used and copied within active theological communities. This establishes the use of the manuscript, if not in the earliest communities where such texts functioned, at least in an environment relatively close to it, and certainly much closer than a text created in the twentieth or twenty-first century—as high as our hopes are for this latter text being “pristine” or an original that never existed in any theological community. The second reason for a single manuscript is that their use offers insights into their original uses that cannot be gained through artificially created eclectic texts. Such insights may arise from the particular features of the text as manuscript that indicate its actual use within a theological and even liturgical and worshiping community. These indicators may include elements internal to the written text, such as additions or subtractions to the text, various spellings, the use of nomina sacra, and other text-immanent matters. However, these indicators may also include elements external to the written text, such as various colophons, word and line divisions, paragraphing, headings, and phonetic-rhetorical markings that guided how the text was understood, interpreted, and utilized within early religious communities (some of these factors are considered in Porter 2013d). Only single text manuscripts can provide access to these features for a particular book.

CONCLUSION A Greek-text-oriented Septuagint commentary series is a necessity for a number of reasons. These reasons include, not only the need to have diversity in the commentary market (although that reason alone should be sufficient to warrant such a commentary), but also fundamental observations regarding the nature of the Greek Old Testament, its use within Judaism and the early Christian church, and the role that it plays within the New Testament. Such a commentary focuses upon the distinctives of what makes the Septuagint important; not the fact that it is a translation designed as a guide to the Hebrew text, but the fact that it was recognized in both Jewish and Christian circles as an authoritative text within these religious communities. As such a text, understanding of the Greek presentation of the scriptural text constitutes a worthwhile and commendable goal for commentary. We are fortunate that we already have such a commentary series being published.

RECOMMENDED RESOURCES 1. Aitken (2015b). One of the few volumes to introduce the individual books of the Septuagint, including a variety of information regarding general characteristics, time and place of composition, language, translation and composition issues, text-critical matters, ideology, and reception history. Greek-text-oriented scholars are underrepresented in the contributors. See also Kreuzer (2019). 2. Evans (2001). A detailed study of the use of the verb in the Greek Pentateuch, arguing that the usage conforms to standard Greek usage, rather than being Semitized Greek.

THE SEPTUAGINT: A GREEK-TEXT-ORIENTED APPROACH 377

3. Hess (2016). An accessible explanation and defense of the Brill Septuagint Commentary Series, a Greek-text-oriented commentary that emphasizes the use of a single manuscript of the Greek text from one of the early codices. 4. Muraoka (2016). The first complete grammar of the syntax of the Greek language used in the Septuagint, well illustrated in its taxonomic description. 5. Porter (2016b). A historically based discussion of scholarship on the Greek of the Septuagint, tracing it through four major periods of research: (1) Semitic, (2) Hellenistic/Koine, (3) Hebraic/Jewish-Greek, and (4) revival of Hellenistic periods. 6. Swete (1900). A still useful introduction to the Septuagint, with valuable information on the Greek not readily found in other introductions. 7. Thackeray (1909). An uncompleted but still very valuable introduction to aspects of the Greek of the Septuagint, with early yet still important and informative discussion of translation styles of the various books. 8. van der Louw (2007). Recent discussion of the Septuagint in light of developments within the field of Translation Studies, taking both a historical and a stratified view of the translation process.

378

PART SIX

Survey of Literature

380

CHAPTER 25

The Septuagint and Contemporary Study JENNIFER BROWN JONES

INTRODUCTION The growing interest in Septuagintal research has provided scholars, students, and laypeople interested in the discipline with numerous resources, ranging from introductory volumes and linguistic resources to new translations, commentaries, societies, and conferences. This chapter will offer an overview of these resources to assist those who are new to the field as well as experienced practitioners. In an effort to provide a quick reference tool, the resources will be listed by author(s) or editor(s), a short title, and the publication date, unlike the author-date-style references elsewhere in this volume. A brief annotation will be included for selected resources. Full bibliographic information can be found in the References section at the end of this volume. The chapter will start with bibliographies, surveys, and introductions to the field, followed by original language resources, modern language translations, and secondary literature, finally concluding with an introduction to currently available software resources.1

BIBLIOGRAPHIES The bibliographies listed below provide users with lists of resources related to various aspects of Septuagintal research. The time periods covered vary due to both the dates of preparation and the particular interest or focus. Annotated bibliographies are noted as such and can be particularly helpful for identifying the most useful resources for a given area of research. 1. S. P. Brock, C. T. Fritsch, and S. Jellicoe, eds., A Classified Bibliography of the Septuagint (1973). The included resources generally date between 1860 and 1969 and are classified by type, addressing critical works on the original translation, the revisions, and the daughter/secondary versions. 2. C. Dogniez, ed., A Bibliography of the Septuagint: 1970–1993 (1995). Dogniez’s bibliography updates that of Brock, Fritsch, and Jellicoe through 1993.

The list below is deeply indebted to the list of resources in Jobes and Silva (2015: 365–8), to the bibliographies and introductions to the field noted within the current chapter, and to William A. Ross’s blog, Septuaginta&c., which may be found at https://williamaross.com/ (accessed February 21, 2020).

1

382

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

3. A. Léonas’s “The Septuagint in Premodern Study: A Bibliography” (2008) focuses on material that generally predates the bibliography of Brock, Fritsch, and Jellicoe, dating back as early as the fourteenth century, but focusing on the seventeenth to nineteenth centuries.2 4. J. Lust’s “Selective Bibliography of the Septuagint” (1999) does not include annotations but does include helpful comments by Lust dealing with various resources and issues in the field at the end.3 5. E. Tov, A Classified Bibliography of Lexical and Grammatical Studies (1980). Tov’s work focuses on Septuagintal language more specifically, addressing both the original translation and the revisions. 6. United Bible Societies, “Septuagint Bibliography” (n.d.) deals primarily with resources dated between 1990 and 2002; this bibliography includes annotations on select entries.4 7. J. W. Wevers, “Septuaginta-Forschungen” (1954) and “Septuaginta Forschungen seit 1954” (1968). Wevers provides two German language bibliographies with extended comments on a variety of resources, including editions, texts, and secondary literature, often listed by book. The first includes resources generally dated between 1938 and 1954, and the second between 1954 and 1966.

LITERATURE SURVEYS 1. E. Tov’s “Jewish Greek Scriptures” (1986) includes resources dating back to 1934 and is subdivided into two primary sections focusing on the Septuagint and then its revisions. Of note is Tov’s discussion of the contribution of the Qumran material. For further discussion about the Qumran material, see G. Kotzé’s chapter in the present volume. 2. L. Greenspoon’s “‘It’s All Greek to Me’: Septuagint Studies Since 1968” (1997) follows up on Tov (1986) and offers an overview of the trends and emphases within the field between 1968 and 1997. In particular, he considers the nationally oriented nature of Septuagintal study programs, describing the particular concentrations that have come to be distinctively associated with particular countries. (On the history Septuagint scholarship in various countries, see the articles in vols. 51 and 52 of the JSCS.) He also discusses the textual development found in later revisions, recensions, and daughter/secondary versions. 3. K. De Troyer’s “The Seventy-Two and Their Many Grandchildren” (2012) picks up the literature review where Greenspoon (1997) leaves off. Her review focuses on two categories: “tools and larger projects” and “thematic studies.” 4. Also worthwhile is Meiser’s “Present Perspectives of Septuagint Research” (2018), which is not a survey of literature per se, but does provide a useful and brief overview of current resources and debates.

Available online: http://ccat.sas.upenn.edu/ioscs/journal/volumes/bioscs41.pdf (accessed February 21, 2020). Available online: http://www.bible-researcher.com/lxx-bibliography.html (accessed February 21, 2020). 4 Available online: http://www.ubs-translations.org/bibliographies/#c519 (accessed February 21, 2020). 2 3

THE SEPTUAGINT AND CONTEMPORARY STUDY 383

INTRODUCTIONS TO THE DISCIPLINE Introductions to the Septuagint generally include discussions covering a wide variety of research interests and critical issues in the field, including, but not limited to, Septuagintal origins, textual transmission, textual development in the revisions and recensions, reconstruction of the original text, translation technique, linguistic style, text-critical use of the manuscripts, and reception history. 1. H. B. Swete’s An Introduction to the Old Testament in Greek (1900, second edition 1914) is the classic introduction and was the first of its kind. It gives a thorough, if dated, introduction to the Septuagint, covering such topics as Septuagintal origins, revisions and recensions, daughter/secondary versions,5 manuscripts and printed editions, the character and order of the books included in the various manuscripts relative to the Hebrew text, textual divisions, the character of the language, as well as Septuagintal reception and influence. Although the scholarly discussion around these issues has advanced considerably since its publication, Swete’s work remains valuable, particularly in its discussion of the various manuscripts, their character, and the textual divisions. This introduction does presuppose a working knowledge of Greek, Hebrew, and Latin at various points. 2. R. R. Ottley’s Handbook to the Septuagint (1920) includes a dated introduction to the Septuagint, originally geared toward a popular audience, but often more detailed or advanced than would be expected of such a resource. 3. S. Jellicoe’s The Septuagint and Modern Study (1968) updates Swete’s classic work described above. It offers more recent information on the various revisions, recensions, and daughter/secondary versions, as well as on the various uncials, cursive manuscripts, and papyri and their contributions to text-critical work. Jellicoe occasionally presupposes knowledge of Greek and Hebrew, but often includes English translations. 4. M. Harl’s, G. Dorival’s, and O. Munnich’s La Bible grecque des Septante (1994) offers a comprehensive French introduction. 5. M. Cimosa’s Guida allo studio della Bibbia greca (LXX) (1995) is an Italian volume that includes a selection of Greek texts with translations and commentary. 6. N. Fernández Marcos’s The Septuagint in Context (2000b) is a translation of his Spanish volume, Introducción a las versiones griegas de la Biblia (1998), which sought to update the work of both Swete and Jellicoe. This accessible volume is noteworthy for its discussion of the Septuagint’s Hellenistic origins and influence on Christianity, as well as the various ways in which the Septuagintal text developed in both Judaism and Christianity. 7. F. Siegert, Zwischen hebräischer Bibel und Altem Testament (2001) includes a German language introduction covering a broad range of issues, including origins, transmission, and translation issues. 8. J. Dines’s The Septuagint (2004) offers a brief English-language introduction with information that is useful to both novices and more advanced researchers.

For further discussion of the recensions and secondary versions, see the contributions of Mäkipelto and Cox, respectively, in this volume.

5

384

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

9. M. Tilly’s Einführung in die Septuaginta (2005) includes a brief German language introduction that generally describes the standard scholarly stances on basic Septuagintal issues, although it can at times be idiosyncratic in its perspectives. 10. The second edition of Fernández Marcos’s Septuaginta: La Biblia griega de los judíos y cristianos (2014) offers a Spanish language introduction to the field for non-specialists, including material similar to that found in his Introducción a las versiones griegas de la Biblia (1998), noted above. 11. The second edition of K. Jobes’s and M. Silva’s Invitation to the Septuagint (2015) offers an accessible entry point, addressing the Septuagint’s history and development, its use in biblical research, and the current state of the field. The newer version not only includes an updated discussion of the scholarship discussed in the first edition (2000), but also clarifies certain points and includes a new appendix listing the sigla used in the Göttingen Septuagint’s critical apparatus. 12. E. Tov’s The Text-Critical Use of the Septuagint in Biblical Research (2015a) is not technically an introduction to the field; however, it offers an essential introduction to various aspects of the text-critical enterprise as it relates to the Greek version, revising and expanding the discussion found in the first (1981) and second (1997) editions. 13. For S. Kreuzer’s Introduction to the Septuagint (2019), see the Handbook section below.

TEXTUAL EDITIONS Many of the early editions are available online at archive.org, as noted in the footnotes to the individual resources. 1. R. Holmes’s and J. Parsons’s Vetus Testamentum Graecum cum variis lectionibus (1798–1827) provides a five-volume critical edition of the Sixtine edition of the Septuagint with variants from approximately 300 additional manuscripts. Codex Vaticanus (B) served as the base for the Sixtine edition of 1587. For introductory remarks about Holmes-Parsons and the Sixtine edition, see Jobes and Silva (2015: 65–6).6 2. Swete’s The Old Testament in Greek according to the Septuagint (1887–94) is a three-volume “manual” edition that includes a diplomatic edition of Codex Vaticanus that was supplemented by Alexandrinus and Sinaiticus at the lacunae.7 Listing variants from the major uncials, Swete’s edition was a notable advance over the somewhat inconsistently collated Holmes-Parsons edition and was the most frequently used edition of the early twentieth century. 3. A. Brooke’s, N. McLean’s, and H. Thackeray’s The Old Testament in Greek according to the Text of Codex Vaticanus (1906–40), commonly known as the

Available online: https://archive.org/details/VetusTestamentumGraecumCumVariisLecVol101Genesis (accessed February 21, 2020). 7 Available online: Volume 1, https://archive.org/details/oldtestamentingr01swetuoft/page/n5/mode/2up; Volume 2 https://archive.org/details/oldtestamentingr00swet/page/n5/mode/2up; Volume 3, https://archive.org/details/ oldtestgreek00unknuoft/page/n5/mode/2up (accessed February 21, 2020). 6

THE SEPTUAGINT AND CONTEMPORARY STUDY 385

Larger Cambridge edition, includes a diplomatic edition of Codex Vaticanus. While the project was never completed, the available books are listed in Table 25.1 with those found in the Göttingen edition noted below.8 4. Two well-known textual editions were produced through the SeptuagintaUnternehmen, a center for research in Göttingen, Germany from 1908 to 2015. First, A. Rahlfs’s Septuaginta (1935), which was later revised by Hanhart (2006), includes an eclectic text that seeks to reconstruct an archetypal text from which all subsequent manuscripts derive (Tov 2015a: 12). Second, the Septuaginta: Vetus Testamentum Graecum, which is usually referred to as the Göttingen Septuagint, follows the same approach and to date includes twenty-four fascicles of individual Septuagint books. Since the Göttingen Septuagint is no longer based in the Septuaginta Unternehmen, the new Forschungskommission zur Edition und Erforschung der Septuaginta (Research Commission for the Edition and Study of the Septuagint) is scheduled to complete the remaining fascicles, with ten volumes currently in process.9 The available and forthcoming fascicles, along with those available in the Larger Cambridge edition,10 are provided in Table 25.1. Table 25.1 Comparison of Göttingen and Larger Cambridge Critical Editions English Book Title

Göttingen Septuagint

Larger Cambridge Edition

Genesis

Genesis (Wevers 1974a)

Genesis (1906)

Exodus

Exodus (Wevers 1991)

Exodus (1909)

Leviticus

Leviticus (Wevers 1986)

Leviticus (1909)

Numbers

Numeri (Wevers 1982a)

Numbers (1911)

Deuteronomy

Deuteronomium (Wevers 2006)

Deuteronomy (1911) Joshua (1917)

Joshua

11

Judges

Iudices (forthcoming)

Judges (1917)

Ruth

Ruth (Quast 2009)

Ruth (1917)

1–2 Samuel

Regnorum liber I; Regnorum liber II (forthcoming; two volumes)

1–2 Samuel (1927)

1–2 Kings

Regnorum liber III–IV (forthcoming)

1–2 Kings (1930)

1–2 Chronicles

Paralipomenon liber I (forthcoming); Paralipomenon liber II (Hanhart 2014)

1–2 Chronicles (1932)

Ezra–Nehemiah 1 Esdras

Ezra–Nehemiah (1935) Esdrae liber I (Hanhart 1991)

1 Esdras (1935)

Available online: https://archive.org/details/OldTestamentGreeklxxTextCodexVaticanus (accessed February 21, 2020). 9 Note that the third edition of Sapientia Jesu Filii Sirach (Wisdom of Ben Sira) listed above was completed under the auspices of this new group. Further information about the project is available online: https://adw-goe.de/en/ research/working-groups/edition-und-erforschung-der-septuaginta/ (accessed February 21, 2020). 10 Note that the order of the books is listed according to the Göttingen edition volume numbers. 11 See the edition of Margolis (1931–8, 1992). 8

386

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

English Book Title

Göttingen Septuagint

2 Esdras

Esdrae liber II (Hanhart 2017a)

Esther

Esther (Hanhart 1983a)

Esther (1940)

Judith

Iudith (Hanhart 1979)

Judith (1940)

Tobit

Tobit (Hanhart 1983b)

Tobit (1940)

1 Maccabees

Maccabaeorum liber I (Kappler 1990)

2 Maccabees

Maccabaeorum liber II (Hanhart 2017b)

3 Maccabees

Maccabaeorum liber III (Hanhart 1980)

4 Maccabees

Maccabaeorum liber IV (forthcoming)

Psalms (and Odes)

Psalmi cum Odis (Rahlfs 1979)

Proverbs

Prouerbia (forthcoming)

Ecclesiastes

Ecclesiastes (Gentry 2019)

Song of Songs (Solomon)

Canticum (forthcoming)

Job

Iob (Ziegler 1982)

Wisdom of Solomon

Sapientia Salomonis (Ziegler 2017)

Sirach

Sapientia Jesu Filii Sirach (Ziegler 2016)

Psalms of Solomon

Psalmi Salomonis (Albrecht 2018)

Twelve (Minor) Prophets (Hosea– Malachi)

Duodecim Prophetae (Ziegler 2014)

Isaiah

Isaias (Ziegler 1983)

Jeremiah, Baruch, Lamentations, Letter of Jeremiah

Ieremias, Baruch, Threni, Epistula Ieremiae (Ziegler 2013)

Ezekiel

Ezechiel (Ziegler 2015)

Daniel, Susanna, Bel and the Dragon

Susanna, Daniel, Bel et Draco (Ziegler 1999)



Larger Cambridge Edition

Several supplemental resources for the Göttingen Septuagint should also be noted. First an official list of the sigla used in the textual apparatus has been published by the Septuaginta-Unternehmen (Rahlfs 2012).12 Second, C. Schäfer has produced two handbooks on using the Göttingen Septuagint, Die Edition des Pentateuch (2012a) and Die Edition des Buches Ruth (2012b), which specifically address use

Available online: https://rep.adw-goe.de/bitstream/handle/11858/00-001S-0000-0022-A30C-8/Rahlfs-Sigeln_ Stand_Dezember_2012.pdf?sequence=1 (accessed February 21, 2020).

12

THE SEPTUAGINT AND CONTEMPORARY STUDY 387

of the Pentateuchal volumes and Ruth, respectively. Third, Appendix E of Jobes and Silva (2015: 381–4) contains an English explanation of the sigla used in the textual apparatus of the Göttingen editions. Finally, Abram Kielsmeier-Jones (2012) has produced a helpful guide to reading the edition and its apparatus.13 5. N. Fernández Marcos and B. Sáiz have edited El texto antioqueno de la Biblia griega (1989–96), a critical edition of the Antiochene text, also known as the Antiochian or Lucianic text, which is one of the Septuagintal recensions and is most notably attested in parts of the historical books and Psalms. Scholars suggest that some readings may also exist in Ruth, Job, the prophets, as well as in some of the apocryphal literature, although these readings are at times debated. The three-volumes of Fernández Marcos and Sáiz cover 1–2 Samuel, 1–2 Kings, and 1–2 Chronicles. 6. F. Field’s Origenis Hexaplorum quae supersunt (1875) includes a critical edition of the fragments of Origen’s Hexapla.14 Information on these fragments can also be found in the Göttingen Septuagint’s second apparatus as well as that of Larger Cambridge edition.15 The Hexapla Institute is in the process of preparing an updated version of Field’s work. For more information about the Hexapla, see Gentry’s chapter in the present volume. Some further resources make use of the textual editions discussed above, offering readers annotated Greek texts that include notes assisting the reader with such issues as vocabulary and syntax. Several recent or forthcoming contributions should be noted: 1. K. Jobes’s Discovering the Septuagint (2016) includes ten texts from RahlfsHanhart (2006), representing a variety of genres, and is particularly designed for graduate and seminary students who have already taken Greek coursework. 2. G. Lanier’s and W. Ross’s two-volume Septuaginta: A Reader’s Edition (2018) includes the text of the entire Septuagint from Rahlfs-Hanhart (2006), including the so-called apocryphal books and double texts. It provides both glosses and parsing for less frequent words at the bottom of each page of text along with equivalents for more common words in a glossary. These volumes are also designed for those who are familiar with New Testament Greek. 3. S. Ehorn’s 2 Maccabees 1–7 (2020) provides a new translation and close syntactical analysis in the first volume of the new Baylor Handbook on the Septuagint series (BHLXX). Forthcoming volumes will cover Exodus, Isaiah, Amos, Habakkuk, Zephaniah, Haggai, Ecclesiastes, and 2 Maccabees 8–15.

LEXICONS 1. E. Bons’ Historical and Theological Lexicon of the Septuagint will be a multivolume project with approximately 600 articles for important words and word Available online: http://abramkj.wordpress.com/2012/11/04/how-to-read-and-understand-the-gottingen-septuaginta-short-primer-part-1 (accessed February 21, 2020). 14 Available online: https://archive.org/details/origenhexapla01unknuoft (accessed February 21, 2020). 15 Apparatuses in critical editions provide the attested textual variants from the readings found in the edition’s main text. Jobes and Silva (2015: 148–55) provide helpful examples for reading a critical apparatus. 13

388

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

groups, addressing such issues as their Greek background, their distribution and relationship to their translation equivalents, and subsequent development in later religious literature. The first volume was published in 2020, and the project is likely to go on for some time to come. 2. G. A. Chamberlain’s The Greek of the Septuagint (2011) functions as a supplement to A Greek-English Lexicon of the New Testament and Other Early Christian Literature (BDAG). BDAG focuses primarily on the vocabulary of the New Testament, while Chamberlain’s Chamberlain’s work includes the words and translation equivalents relevant to the Septuagint that are not already found in BDAG. 3. J. Lust’s, E. Eynikel’s, and K. Hauspie’s Greek-English Lexicon of the Septuagint (3rd ed.; 2015), often abbreviated LEH, includes translation equivalents for the Greek words found in Rahlfs’s edition, focusing on the Septuagint as a translation. By adopting this approach, LEH at times draws on the presumed underlying Hebrew and Aramaic Vorlagen, offering renderings based on what the translator appears to have intended. 4. T. Muraoka’s A Greek-English Lexicon of the Septuagint (2010b), often abbreviated GELS, focuses on the Septuagint as a Greek document in its own right and does not draw on the various Vorlagen to derive the definitions; rather, Muraoka takes an audience-centered approach, offering renderings reflecting what an audience in the final centuries BCE would have understood. Additionally, the volume includes explanations of the meanings, rather than the glosses or translation equivalents that are offered in Lust, Eynikel, and Hauspie (noted above). 5. F. Rehkopf’s Septuaginta-Vokabular (1989) is a concise German lexicon of the vocabulary found in Rahlfs, drawing on A Greek-English Lexicon (Liddell et al. 1996) and Griechisch-Deutsches Wörterbuch (Bauer 1988) for its equivalents. 6. B. Taylor’s Analytical Lexicon to the Septuagint (2009) offers parsing for each word form found in Rahlfs, as well as glosses for the lexical forms, which were drawn from the 1992 edition of the lexicon by Lust, Eynikel, and Hauspie.

GRAMMARS 1. L. Cignelli’s and R. Pierri’s Sintassi di Greco biblico (2003) is an Italian resource that currently has two of six expected volumes available, including information on both New Testament and Septuagintal Greek. 2. F. C. Conybeare’s and St. G. Stock’s Grammar of Septuagint Greek ([1905] 1995) includes three sections. The first two address accidence (inflectional morphology) and syntax. The third section includes a number of annotated readings drawn from the Septuagintal text.16 3. R. Helbing has published a German language grammar, Grammatik der Septuaginta (1907), that focuses primarily on the orthography and morphology found in the texts of the major codices and some early papyrological fragments. He also has a Available online: http://www.ccel.org/ccel/conybeare/greekgrammar/Page_Index.html (accessible February 21, 2020).

16

THE SEPTUAGINT AND CONTEMPORARY STUDY 389

resource on verbal syntax, Die Kasussyntax der Verba bei den Septuaginta (1928), that includes the various verbal constructions, meanings, and Hebrew equivalents for various groups of verbs as a part of his discussion of Hebraisms.17 Both of Helbing’s works are still referenced by scholars. 4. T. Muraoka’s A Syntax of Septuagint Greek (2016) includes the most extensive treatment of Septuagintal grammar available. As with his lexicon described above, Muraoka adopts an audience-centered approach that focuses on the audience of the Septuagint as a Greek text without recourse to the Hebrew, although he recognizes Hebrew language interference as appropriate. While he generally deals with the Septuagint as a whole (synchronically), he is sensitive to the diachronic development of the corpus over time. 5. H. Thackeray’s A Grammar of the Old Testament in Greek (1909) focuses on orthography, phonetics, and accidence (inflectional morphology). Thackeray’s classic not only addresses issues of spelling and grammatical form based on his analysis of the material found in the uncial manuscripts and available papyri, it also discusses Hebraisms and the style of the various books relative to one another and to extra-biblical evidence from the period.

CONCORDANCES 1. E. Hatch’s and H. A. Redpath’s Concordance to the Septuagint and other Greek Versions of the Old Testament (1998), often abbreviated HR, is the second edition of the original three volume version published between 1897 and 1906. The Concordance is based on the Alexandrinus, Vaticanus and Sinaiticus codices, and the Sixtine edition, but lacks equivalents for such common Greek words as καί, οὐκ, and δέ. Several limitations of HR should be noted. First, it includes a limited textual base that does not account for many of the variants. Second, the editors had an inconsistent approach to identifying equivalents that often avoided identifying possible or even probable Hebrew equivalents in favor of the actual word found in the Masoretic Text. Ultimately, however, it remains a valuable tool for Septuagintal research.18 A reverse index, the Hebrew/Aramaic Index to the Septuagint, has been prepared by Muraoka (1998). This reverse index provides each Semitic word found in 1 Esdras, Sirach, and the Hebrew Bible along with a list of their associated Greek equivalents as listed in the Hatch and Redpath Concordance; it also lists the page and column number for each Greek term. This index allows researchers to quickly find the Greek equivalents found in the texts used by Hatch and Redpath for any Hebrew or Aramaic lexeme. 2. N. Fernández Marcos’s, M. V. Spottorno Díaz-Caro’s, and J. M. Cañas Reíllo’s (eds.), Índice Griego-Hebreo del Texto Antioqueno en los libros históricos (2005) supplements Hatch and Redpath’s Concordance with material from the Antiochene

Available online: https://archive.org/details/grammatikdersept00helbuoft (accessible February 21, 2020). See the discussion in Tov (2015a: 100–9) about using a concordance in textual criticism, as well as his list of supplementary resources that can offset the issues in HR. The contributions of J. M. Cañas Reíllo and J. Screnock in the present volume also address text-critical issues.

17 18

390

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

text. It includes both Greek-Hebrew and Hebrew-Greek indices as well as an introduction in both Spanish and English.

MODERN LANGUAGE TRANSLATIONS For a more extended treatment of modern language translations of the Septuagint, see Ross’s chapter in the present volume. 1. English: Four full translations are available. First, C. Thomson’s translation was originally published in 1808 and was revised in 1960 by C. A. Muses. Based on Codex Vaticanus, the translation includes Psalm 151, but excludes the deuterocanonical books. Second, L. C. L. Brenton’s translation (1844) also derived from Vaticanus and included the same ordering as Thomson’s, although Brenton did include the apocryphal material. The translation also includes occasional notes on the texts as well as variants noted in Codex Alexandrinus. Third, A New English Translation of the Septuagint (NETS; Pietersma and Wright 2007) was corrected, emended, and reprinted in 2009.19 Following the interlinear paradigm, the translation uses the critical texts available in the Göttingen Septuagint and Rahlfs-Hanhart Septuaginta.20 It does not include an entirely new translation, but rather is based on the 1989 NRSV. In an effort to provide an English text that would be useful for comparing the underlying Hebrew and Greek texts, the NETS translators used the NRSV as a base text, changing its rendering at points where it did not represent the Greek. Fourth, in 2019 K. M. Penner and R. Brannan et  al., published The Lexham English Septuagint (LES) that is based on an earlier digital resource available through the Logos Bible Software platform, which itself had been computer generated from a digital Greek-English interlinear. Note also that a fifth, though partial, translation of the Septuagint is available in each volume of the Brill Septuagint Commentary Series (SEPT), pertaining to individual books in the corpus. 2. French: La Bible d’Alexandrie (BdA) offers a French language translation of and commentary on the Göttingen Septuagint. The translation principles of La Bible d’Alexandrie indicate that the Greek text is translated without regard for its Hebrew source (Vorlage), focusing on the Greek text as being intelligible in its own right. The available commentary on the Greek text addresses such issues as translation, translation variants, and reception. While the project is currently incomplete, Table 25.2 shows the volumes that have been published to date: 3. German: Septuaginta Deutsch (LXX.D) includes a multifaceted project offering a German language translation, commentary, and handbook. The main volume, Das griechische Alte Testament in deutscher Übersetzung (Kraus and Karrer 2009b), includes a translation of the Septuagint that is based on the Göttingen and Rahlfs-Hanhart critical texts. See the discussion in the “Commentaries” and “Handbooks” sections below for information on the related works.

Further corrections and emendations were made in 2014, available online: http://ccat.sas.upenn.edu/nets/ edition/ (accessed February 21, 2020). A further, revised edition of NETS is anticipated in the future. 20 For a discussion of the interlinear paradigm see Hiebert’s chapter in the present volume. 19

THE SEPTUAGINT AND CONTEMPORARY STUDY 391

Table 25.2 Published Volumes in BdA Genesis

La Genèse (Harl 1986)

Exodus

L’Exode (Le Boulluec and Sandevoir 1989)

Leviticus

Le Lévitique (Harlé and Pralon 1988)

Numbers

Les Nombres (Dorival 1994)

Deuteronomy

Le Deutéronome (Dogniez and Harl 1992)

Joshua

Jésus (Josué) (Moatti-Fine 1996)

Judges

Les Juges (Harlé 1999)

Ruth

Ruth (Assan-Dhôte and Moatti-Fine 2009)

1 Samuel

Premier livre des Règnes (Lestienne and Grillet 1997)

2 Esdras

Deuxième livre d’Esdras (Janz 2010)

Esther

Esther (Cavalier 2012)

3 Maccabees

Troisième livre des Maccabées (Mélèze-Modrzejewski 2008)

Proverbs

Les Proverbes (d’Hamonville 2000)

Ecclesiastes

L’Ecclésiaste (Vinel 2002)

Song of Songs

Le Cantique des Cantiques (Auwers 2019)

Hosea

Les Douze Prophètes. Osée (Bons et al. 2002)

Joel, Obadiah, Jonah, Nahum Habakkuk, Zephaniah

Les Douze Prophètes. Joël, Abidou, Jonas, Naoum, Ambakoum, Sophonie (Harl et al. 1999)

Haggai, Zechariah

Les Douze Prophètes. Aggée, Zacharie (Harl et al. 2007)

Malachi

Les Douze Prophètes. Malachie (Vianès 2011)

Baruch, Lamentations, Epistle of Jeremiah

Baruch, Lamentations, Lettre de Jérémie (Assan-Dhôte and Moatti-Fine 2005)

4. Italian: The series La Bibbia dei Settanta has one introductory volume, Introduzione alle versioni greche della Bibbia (Fernández Marcos 2000a), and three translation  volumes: Pentateuco, Libri storici, and Libri poetici (Lucca et al. 2012–16). These volumes include the text of Rahlfs’ editio minor, or “manual” edition (1935), the Italian translation, as well as comments on the translation. Two further Italian translation projects exist. The Pentateuch is available in La Bibbia dei LXX. 1. Il Pentateuco (Mortari 1999), while the entire Septuagint is available in the two-volume version of La Bibbia secondo la versione dei Settanta (Brunello 1960). 5. Spanish: La Biblia griega. Septuaginta, often abbreviated LBGS, was translated by N. Fernández Marcos, M. V. Spottorno Díaz-Caro, and J. M. Cañas Reíllo and includes four volumes: El Pentateuco (2008; second edition, 2016), Libros históricos (2011), Libros poéticos y sapienciales (2013), and Libros proféticos (2015). The translations are based on critical texts, including the Göttingen editions, Rahlfs-Hanhart, and the critical edition of the Antiochene text prepared by Fernández Marcos and Sáiz, each of which is discussed above. Every book

392

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

includes a brief introduction that covers such issues as the date and location of the Greek translation, its literary character and structure, the primary textual witnesses, the book’s reception, and bibliography. In addition to the translation volumes, the introduction Septuaginta: La Biblia griega de los judíos y cristianos (Fernández Marcos 2014), discussed above, is also available.

COMMENTARIES A variety of approaches have been adopted in Septuagintal commentaries, alternately focusing on the text as produced or the text as received. For more thorough discussions relating to these approaches, see the chapters by Hiebert and Porter in the present volume. 1. The Brill Septuagint Commentary Series volumes (SEPT) include a translation of and commentary on a particular Greek manuscript such as Vaticanus, Sinaiticus, or Alexandrinus, rather than on an eclectic text. Focusing on the text as received, these commentaries offer insight into the reception of the Greek text within particular communities with minimal reference to the Hebrew Vorlagen. To date the volumes in Table 25.3 are available. 2. For a description of La Bible d’Alexandrie and a list of available volumes, each of which provides running commentary on the text, see the Modern Language Translations section above. 3. Based on the best possible reconstructed text, the SBL Commentary on the Septuagint (SBLCS) series will focus on the Septuagint as produced. In addition to a verse-by-verse commentary, each volume is expected to address such issues as origin, literary structure, themes, and translation technique. As of October 2020 no volumes have yet been published, but a sample volume edited by D. Büchner, The SBL Commentary on the Septuagint: An Introduction (2017), is available that includes a description of the principles behind the series as well as commentaries on sample passages from seven different books of the Septuagint (Genesis, Exodus, Leviticus, Numbers, Esther, Job, and Psalms). 4. Septuaginta Deutsch includes a multi-stage project offering a German language translation, commentary, and handbook. In addition to the translation volume Table 25.3 Published Volumes in SEPT Genesis (Brayford 2007)

Proverbs (Wolters 2020)

Exodus (Gurtner 2013)

Hosea (Glenny 2013b)

Leviticus (Awabdy 2019)

Amos (Glenny 2013a)

Joshua (Auld 2005)

Micah (Glenny 2015)

1 Esdras (Bird 2012)

Isaiah (Penner 2020)

Tobit (Littman 2008)

Jeremiah (Walser 2012)

3 Maccabees (Croy 2006)

Baruch and the Epistle of Jeremiah (Adams 2014)

4 Maccabees (deSilva 2006)

Ezekiel (Olley 2009)

THE SEPTUAGINT AND CONTEMPORARY STUDY 393

mentioned above (Kraus and Karrer 2009b) and the handbook discussed in the following section, it includes a two-volume commentary: Septuaginta Deutsch: Erläuterungen und Kommentare zum griechischen Alten Testament (Karrer and Kraus 2011a, 2011b).

HANDBOOKS 1. J. K. Aitken’s The T&T Clark Companion to the Septuagint (2015b) is an edited volume with introductions to each book of the Septuagint, with which the present volume is complementary. 2. Eight volumes are anticipated in the Handbuch zur Septuaginta series (LXX.H), which will summarize the results of the Septuaginta Deutsch project and offer a comprehensive overview of Septuagint research. To date, two volumes have been published: Einleitung in die Septuaginta (Kreuzer 2016) and Die Sprache der Septuaginta (Bons and Joosten 2016). Kreuzer’s volume was translated into English in 2019. The German and English editions are identical, including essays covering introductory issues such as Septuagintal origins, textual transmission, and textual witnesses, as well as introductions to each book. These introductions include material similar to that found in Aitken’s The T&T Clark Companion to the Septuagint (2015b), although Kreuzer’s volume includes more extensive discussions of reception history and the Minor Prophets. 3. At the time of publication, A. Salvesen and Timothy Michael Law are in the final stages of editing the forthcoming volume The Oxford Handbook of the Septuagint, which will be part of the Oxford Handbook series and include over fifty articles.

JOURNALS, MONOGRAPHS, AND SERIES 1. The Journal of Septuagint and Cognate Studies (JSCS) is the annual publication of the IOSCS, which was formerly published as the Bulletin of the IOSCS (BIOSCS 1968–2010).21 S. Kreuzer is the present editor. C. Dogniez, A. Salvesen, and G. Wooden comprise the remaining members of the editorial board. 2. The Septuagint and Cognate Studies (SCS) series, published by the Society of Biblical Literature, presently includes over sixty volumes focused on various aspects of Septuagintal, apocryphal, and pseudepigraphical research. Topics include, but are not limited to, text-critical issues, translation, and literary development within both Judaism and Christianity. Key volumes of interest include R. Kraft’s Septuagintal Lexicography (1972b), J. A. L. Lee’s Lexical Study (1983), T. Muraoka’s Melbourne Symposium on Septuagint Lexicography (1990), W. Kraus’s and R. G. Wooden’s Septuagint Research (2006), J. Joosten’s

Volumes 1–43 of JSCS are available online: http://ccat.sas.upenn.edu/ioscs/journal/volumes/ (accessed February 26, 2020). Further information on the IOSCS is available in the Research Centers and Societies section below.

21

394

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

and E. Bons’ Septuagint Vocabulary (2011), S. Kreuzer’s The Bible in Greek (2015b), and the collections of papers delivered at the annual Congresses of the IOSCS.22 The following resources represent noteworthy monographs addressing a variety of topics within Septuagintal studies. 1. P. Walters’s (formerly Katz’s) The Text of the Septuagint (1973), an advanced text requiring competence in Greek, deals with textual corruptions in the Septuagint, considering both those rooted in Greek spelling, grammar, and pronunciation, and those rooted in the Septuagint’s relationship with the Hebrew. 2. M. Hengel’s Septuagint as Christian Scripture (2002) discusses Septuagintal translation and development, Christian reception, and issues of canon. J. Meade offers further insight into the biblical canon in his contribution to the present volume. 3. R. T. McLay’s Use of the Septuagint in New Testament Research (2003) deals with the use of the Old Testament in the New, particularly focusing on the use of the Greek text. His treatment includes a helpful discussion about translation technique. For a discussion of the New Testament’s use of the Old, see the chapter by S. Moyise in the present volume; for translation technique, see M. Dhont’s chapter in the present volume. 4. T. M. Law’s When God Spoke Greek (2013), which is geared to an educated lay audience, considers the origins and development of the Septuagint, its place in the development of the Christian Bible, and its eventual eclipse by the Hebrew text. 5. J. K. Aitken’s No Stone Unturned (2014b) highlights the contribution of ancient Greek inscriptions to our modern knowledge of both Septuagintal vocabulary and the Septuagint’s sociohistorical origins. Also, note Aitken’s chapter in the present volume. 6. E. Bons’, R. Brucker’s, and J. Joosten’s Reception of Septuagint Words (2014) includes essays prepared for two colloquia related to the forthcoming Historical and Theological Lexicon of the Septuagint, noted above, addressing methodological issues and offering a sample article. 7. J. K. Aitken’s and T. V. Evans’s Biblical Greek in Context (2015) includes a collection of essays focusing on biblical Greek within the history of the Greek language and particularly as a representative of Koine Greek as it is attested in the papyrological and inscriptional evidence.23

RESEARCH CENTERS AND SOCIETIES 1. The Association for the Study of the Septuagint in South Africa (LXXSA) seeks to advance Septuagintal research through conferences and publications. Johann Cook is the current president of the organization.

A list of all series titles is available online: see https://www.sbl-site.org/publications/Books_SepandCog.aspx (accessed February 26, 2020). 23 In addition to the lexicographical works noted in this section, see Pouchelle’s chapter in the present volume. 22

THE SEPTUAGINT AND CONTEMPORARY STUDY 395

2. The Centre for Septuagint Studies and Textual Criticism (CSSTC) at the Katholieke Universiteit Leuven in Leuven, Belgium, focuses on text-critical and translation technique research. Founded by J. Lust, the current director is B. Lemmelijn. 3. The Forschungskommission zur Edition und Erforschung der Septuaginta in Göttingen, Germany succeeded the Septuaginta-Unternehmen in January 2016. See above in the “Editions” section for more information. 4. The Groupe de Recherche “Septante et Critique Textuelle” (SSTC) at the Université catholique de Louvain in Louvain-la-Neuve, Belgium, partners with the CSSTC at Leuven, described above, and other research programs in various aspects of text-critical research, addressing both the Old and New Testaments 5. The International Organization for Septuagint and Cognate Studies (IOSCS) promotes research in the Septuagintal, apocryphal, and pseudepigraphical fields. In addition to the various publication projects that have been already mentioned, the IOSCS hosts sessions at the Annual and International meetings of the Society of Biblical Literature. 6. The John William Wevers Institute for Septuagint Studies at Trinity Western University in Langley, British Columbia, focuses on promoting Septuagintal research, particularly hermeneutical issues, by publishing research resources and hosting educational initiatives and scholarly conferences. The current fellows are R. Hiebert, L. Perkins, and D. Büchner.

SOFTWARE PROGRAMS The software programs noted below offer a wide variety of resources for biblical research more generally, but also often include tools that facilitate Septuagintal research in particular. However, the buyer must always beware: If you are specifically seeking a particular resource, make sure that it is available with the program you purchase. Further, key Septuagintal resources are often not included with the base packages, thus requiring additional expenditure. 1. Accordance (https://www.accordancebible.com/) includes a variety of original language and secondary resources for biblical research, including a graphical searching tool and a morphologically tagged and searchable edition of the Göttingen Septuagint. Accordance is available for Mac and PC. It offers a variety of discounts for students, professors, and those in ministry. Accordance is available in Chinese, English, French, German, Portuguese, and Spanish; Korean is currently in preparation. 2. Logos Bible Software (https://www.logos.com/) networks its libraries, which include Hebrew, Greek, and Aramaic language resources, enabling a variety of detailed searches. Many of the resources noted above are available from Logos, including a morphologically tagged and searchable Göttingen Septuagint. Logos is available for Mac and PC and offers an academic discount. It is available in Chinese (Simplified and Traditional), English, German, Korean, Portuguese, and Spanish. 3. While most scholars use one of the programs listed above, database material is also available through the CATSS project (Computer Assisted Tools for Septuagint Studies), which was pioneered by R. Kraft and E. Tov. It includes three tools: the

396

T&T CLARK HANDBOOK OF SEPTUAGINT RESEARCH

“Parallel Aligned Hebrew and Greek Jewish Scriptures”; the “Morphologically Analyzed Septuagint”; and the “Old Greek Variants Files.”24 A variety of resources about the project are available, but two particular ones should be mentioned: “The ‘Accordance’ Search Program for the MT, LXX, and the CATSS Database” (Tov 1997) and “The CATSS Project—A Progress Report” (Tov 1991).25

Available online: http://ccat.sas.upenn.edu/rak/catss.html (accessed February 28, 2020). While one of these articles explicitly mentions Accordance’s program, the other programs mentioned also offer significant resources.

24 25

GLOSSARY

This glossary provides brief explanations of select terms that are either technical in nature or are used in a specific way in Septuagint scholarship or related disciplines. It is certainly not exhaustive but should help the reader avoid having to consult other reference works to understand the discussions in this Handbook. Any terms included in the glossary that are used within a given explanation are put in small capital letters (e.g., ph o n e m e ) to direct the reader to other relevant entries provided. Alexandrian Library Part of the Alexandrian Museon, the library was founded under Pt o l e m y i or Pt o l e m y ii , and it housed a large collection of scrolls. It became a center of intellectual activity, particularly focused on establishing the text of the Homeric epics and their interpretation. Allophone One of possibly multiple ph o n e s that can be used to r e al iz e a ph o n e m e . For example in English, /s/ is sometimes pronounced [s] (boost) and sometimes [z] (position). Allusion A series of words taken from another source without explicitly indicating that it is a q uo t at io n , but usually understood as an intentional attempt by an author to point his/her readers to another text known by both parties. It is most easily spotted when the two texts share some distinctive vocabulary, but sometimes common ideas or themes are sufficient. Anthroponym The name of a person. Antiochene recension See Luc ian ic r e c e n sio n . Apocrypha A term derived from a Greek word meaning “hidden.” Used in the Protestant tradition to refer to the books of the Old Testament that are absent from the Hebrew Bible yet included in the traditional Catholic and Orthodox canons and known in those traditions as d e ut e r o c an o n ic al . Most (but not all) books categorized among the Apocrypha were originally composed in Greek and are typically included in modern c r it ic al e d it io n s of the Se pt uag in t (compare to Pse ud e pig r aph a ). Aquila One of t h e Th r e e early Jewish revisers of the Se pt uag in t . He produced a new Greek translation, or r e c e n sio n , of the Hebrew Scriptures at the beginning of the second century CE. His t r an sl at io n t e c h n iq ue prioritizes a high degree of lexical consistency and word-for-word representation of his Hebrew so ur c e t e x t and was perhaps inspired by the earlier Kai g e style. Archetype In t e x t ual c r it ic ism , an archetype is an idealized form of the text that is considered to be the oldest reconstruction possible on the basis of comparison of manuscripts and other textual evidence. Attenuation A supposed law in Hebrew where /a/ → /i/ in closed unstressed syllables. Baraita The citation of an ancient rabbinic tradition in the Jerusalem or Babylonian Talm ud that is not recorded in the Mish n ah . Catchword Words or phrases found in two or more texts that trigger a connection between  those texts. Catchwords found in the Hebrew Vo rl ag e of the Se pt uag in t

398 GLOSSARY

translator sometimes prompted a mental connection with other biblical texts where the same words or phrases occur. Catena manuscript A manuscript that contains biblical text accompanied by detailed, verse-by-verse commentary (usually excerpts from earlier authoritative commentaries). Citation See q uo t at io n . Clement of Alexandria A Christian church father who lived at the end of the second century CE. Codex An ancient manuscript format much like the modern book, constructed using sheets of paper, vellum, or papyrus, that are folded into q uir e s and bound together. The earliest and most important textual witnesses to the Se pt uag in t as a whole are attested in major codices, such as Alexandrinus or Vaticanus. Collation The process or final result of assembling variant readings that are attested in a set of copies of a given text in order to highlight their differences and similarities for the purpose of t e x t ual c r it ic ism . Composite citation A combination of two or more sources that are introduced by a single q uo t at io n formula. There is debate as to whether readers are expected to know that what follows the quotation formula comes from different sources or treat it as a single quotation. Conflated quotation When one or more words of a q uo t at io n from one source are substituted for another word or words from a different source. Unless readers have a precise knowledge of the actual quotation, they are unlikely to notice this phenomenon. Contamination When the transmission of one text is influenced by the transmission of another text. In particular with reference to a quotation from the Se pt uag in t in the New Testament, when a scribe altered a word or phrase in either the Se pt uag in t text or the New Testament text to bring them into closer alignment. Critical apparatus Textual data that accompanies the main text of a c r it ic al e d it io n , for purposes of t e x t ual c r it ic ism . The apparatus supplies variant readings compiled by c o ll at io n to facilitate understanding the history of textual transmission. Critical edition An edition of a text that features either an e c l e c t ic or d ipl o m at ic t e x t as the main text and a c r it ic al appar at us of variant readings from existing manuscripts or other textual witnesses. Cursive See m in usc ul e . Demetrius of Phalerum In the Letter of Aristeas, he is the head librarian in the A l e x an d r ian L ibr ar y under Pt o l e m y ii . Most scholars see this as an anachronism, since Demetrius most likely was banished from Alexandria shortly after the death of Pt o l e m y i . Dependence The relationship of a text to the in t e r t e x t that has affected it. In the Se pt uag in t , dependence is observed when a particular translation choice owes its existence to an intertext. Deuterocanonical books The term used in Catholic and Orthodox traditions for books that are absent from the canonical Hebrew Bible but included in the Catholic and Orthodox Old Testament canon. See also A po c r yph a and Pse ud e pig r aph a . Diachronic A conceptual perspective that considers the processes by which entities, features, or structures change over time (compare to syn c h r o n ic ). Dimotiki The spoken Greek language of the modern period. A st an d ar d (ized) l an g uag e variety of Demotic Greek is the official language of Greece since 1976. Diplomatic text A text of a c r it ic al e d it io n that reproduces a single textual witness, mostly or entirely unaltered by a modern editor. It is usually (but not always) accompanied by a c r it ic al appar at us that provides variant readings.

GLOSSARY 399

Dittography A mistake made by a copyist while writing a text, caused by writing the same letter, syllable, or phrase more than once. Eastern Orthodox Churches Christian Churches who trace their origin to the Byzantine commonwealth. The list of Eastern Orthodox Churches that mutually recognize each other and therefore were officially invited to the Pan-Orthodox Council of 2016 included churches of Constantinople, Alexandria, Antioch, Jerusalem, Russia, Greece, Cyprus, Romania, Serbia, Bulgaria, Georgia, Poland, Albania, the Czech Republic, and Slovakia. The canonical status of some other Eastern Orthodox Churches is disputed. Echo An echo is similar to an all usio n , but it may be intentional or unintentional. It is not as pronounced as an allusion would be and this makes it difficult to establish intentionality by the author. Eclectic text The text of a c r it ic al e d it io n that is a reconstruction of the oldest possible form of the text (see ar c h e t ype ), produced by a modern editor. Eleazar The name of the Jewish high priest according to the Letter of Aristeas. Although some have tried to identify him as a historical high priest, it is more likely that he is a creation of the author of the Letter. Epiphanius A church father and the bishop of Salamis in Cyprus (d. 403 CE). Etymology The study of the history of words. Eusebius of Caesarea A Christian church father who flourished in the late third and early fourth centuries. He preserves a number of fragments of Jewish writers, which are otherwise not extant, that he has taken from an earlier writer, Alexander Polyhistor (first century BCE). Fricativization The process by which a consonant becomes a fricative, which is a consonant sound involving partial blockage of the airstream, as in English [f ] or [v]. Gemarah The authoritative collection of rabbinical commentary on the m ish n ah , contained in the second portion of the T alm ud . Genizah A storeroom or depository, usually in a synagogue, where worn-out Hebrew books and other documents are placed. This is in accordance with the rabbinic teaching that once a holy book can no longer be used, it should be stored (m. Šabb. 16.1). Of special note is the genizah of the Ben Ezra Synagogue of Fustat (Old Cairo, Egypt), where materials accumulated over a period of almost a thousand years, the earliest dating from the eleventh century CE, including not only scriptural texts but also examples of nearly every kind of writing produced by the Jewish communities of the Near East. Graph A distinct written sign, in formal notation enclosed by the sigla « ». Grapheme The minimal perceptual set of g r aph s that, if interchanged in a given environment, would change the meaning of a m o r ph e m e or word. E.g., the English graphs «s» and «z», when interchanged between the written entities sit and zit, distinguish two different words and therefore express two different graphemes: ‹s› and ‹z›, respectively. In formal notation, a grapheme is enclosed by the sigla ‹ ›. Haplography A mistake by a copyist while writing a text caused by omitting a repeated letter, syllable, or phrase. Hebraism/Semitism A word, expression, or syntactical construction in the t ar g e t t e x t that evinces interference or in fl ue n c e from the Semitic so ur c e t e x t of the Se pt uag in t . Some scholars define this term more narrowly, adding the criterion that the feature under analysis should only be considered Hebraism (or “Semitism,” given the presence of Aramaic in the source text) if it also represents unconventional use in the t ar g e t l an g uag e .

400 GLOSSARY

Hexapla A multi-volume, six-columned textual synopsis of the Hebrew Bible/Old Testament produced by Origen (c. 185–254 CE). The columns included (1) Origen’s Hebrew t e xt us re c e pt us, (2) a transcription of the Hebrew into Greek, (3) A q uil a , (4) Symm ac h us, (5) Origen’s Se pt uag in t textus receptus, and (6) T h e o d o t io n . Hexaplaric recension The result of Origen’s text-critical efforts on the Se pt uag in t , which was derived from the fifth column of the H e x apl a . Hexaplaric readings generally bring the Septuagint closer to the m aso r e t ic t e x t and are found especially in the O manuscript group, as well as in other witnesses. Hierophants In the ancient Greek world, these were people who were thought to have insight into sacred mysteries, usually obtained through inspiration of a god. History of reception The history of the way that texts are interpreted and used by various communities following the initial production of those texts. Homeoarcton A mistake made by a copyist while writing a text, caused by visually skipping over text due to the similarity between the beginning of two words, thus omitting everything between them (compare to H o m e o t e l e ut o n ). Homeoteleuton A mistake made by a copyist while writing a text, caused by visually skipping over text due to the similarity between the ending of two words, thus omitting everything between them (compare to H o m e o ar c t o n ). Hyperbaton The intentional transposition or separation of words (compared to the expected word order), usually for the purposes of style or emphasis in rhetoric. Hypocorism An alternative, less formal, and almost always shorter form of an an t h r o po n ym . For example, English Bill used as a substitution for William. Hypotaxis The subordination of clauses, a common feature of Greek syntax (compare to par at ax is). Influence The effect that A has on B. In the study of in t e r t e x t s or interference in the Se pt uag in t , certain texts affect (i.e., influence) the way the translator interprets or translates his V o rl ag e . Interlinear paradigm A theoretical model that understands the Se pt uag in t —at its point of production—as dependent on the Hebrew text that it translated and not as an independent replacement of it. That is, the linguistic makeup of the Septuagint was intended as a gateway to the Hebrew text. This model served as the basis for A New English Translation of the Septuagint project (see Pietersma and Wright 2009). Intertext A text that exists in connection to one or more other texts. In the case of the Se pt uag in t translation, an intertext would be a text that has exerted some type of discernible in fl ue n c e on the translator’s t ar g e t t e x t . Isomorphism In translation theory, a relationship between a so ur c e t e x t and the t ar g e t l an g uag e in which each grammatical element in the former is represented formally by an element in the latter. Josephus A Jewish historian of the late first century CE. Involved in the First Jewish Revolt, Josephus was captured and taken to Rome, where he wrote several important works, including his Antiquities of the Jews, in which he paraphrases the Letter of Aristeas. Kaige A label or classification that refers to early Jewish revisional activity that began at least in the first century BCE. This activity was more of a movement or tendency than a singular effort and seems to have been characterized by common translation equivalents, such as ‫וגם‬/‫“( גם‬also, even”) and καί γε (“indeed”). Scholars generally believe that the aim of the Kaige r e v isio n was to bring existing Greek translations closer to the pr o t o -M aso r e t ic t e x t . Dominique Barthélemy (1963) first identified the

GLOSSARY 401

Kaige revisional tendency in the Greek Minor Prophets scroll (8ḤevXII gr) that was discovered at the site of Naḥal Ḥever. The members of the Kaige group are debated, but T h e o d o t io n and A q uil a may have worked within this tradition. Karaite A Jewish social and religious movement that developed independently of the rabbinic academies. Originating in the eastern region of the ‘Abbasid caliphate during  the late eighth century CE, the movement coalesced definitively around the turn of the tenth century, and enjoyed a “Golden Age” of literary production through to the end of the eleventh century, by which point its geographical center had shifted to Byzantium. Karaite communities continue to this day, albeit in radically reduced numbers. They emphasize the primacy of written scripture and do not recognize the authority of the traditions incorporated in the t alm ud and later Rabbinic tradition. Katharevousa A (predominantly written) archaizing form of the Greek language conceived in the late eighteenth century as a compromise between classical Greek and the spoken Greek of the time. It was the official language of Greece until 1976. Koine The historical phase of ancient Greek between the classical and Byzantine periods, including the Hellenistic and Roman periods (c. 330 BCE–330 CE). Sometimes known simply as post-classical Greek, the term “Koine” does not designate any particular variety or register, although st an d ar d and n o n -st an d ar d varieties certainly existed, as in every language. Rather, the term refers to the language system as a whole. Lagids The nickname given to the Ptolemaic dynasty in Egypt, after the father of Pt o l e m y   i , Lagus. Loanword A word that is copied from one language to another. Lucianic recension A Greek version historically attributed to Lucian of Antioch in the late third century CE, but now widely recognized as originating much earlier. This tradition sought to update the Greek style of the Se pt uag in t and contains many readings from the H e x apl a . There is considerable disagreement in scholarship as to the extent of the r e c e n sio n (or r e v isio n ?), as well as the text-critical value of Lucianic manuscripts. Some prefer to call this tradition the Antiochene or Antiochean recension in order to distance it from Lucian the historical figure. LXX See Se pt uag in t . Majuscule A term used in both Greek and Latin pal e o g r aph y to refer to large (“capital” or “uppercase”) letters, in contrast to small (“lowercase”) letters, which are referred to as m in usc ul e . There are two kinds of majuscule: (1) capital script, formed by strokes meeting at angles (as in inscriptions on stone or metal) and thereby avoiding curves; and (2) un c ial script, a modification of capital script. Masoretic Text (MT) The standardized form of the Hebrew text of the Bible that was transmitted by medieval scribes called Masoretes. Also see pr o t o -M aso r e t ic t e x t . Megillah The Hebrew word for “scroll” (‫ ;מגילה‬pl. “Megillot”), which designates any of the five books of the Ketubim (‫כתובים‬, “writings”) that are recited in the synagogue liturgy in the course of certain Jewish festivals. The Megillot include the Song of Songs (or Canticles), Ruth, Lamentations, Ecclesiastes, and Esther. Menaion A collection of liturgical texts read or sung in the Orthodox Christian tradition during worship and linked with fixed dates of the calendar year. Minuscule A style of handwriting characterized by smaller, more rounded, and often interconnected letter forms, common in Greek manuscripts from the ninth century on (compare to m ajusc ul e ). Mishnah The authoritative collection of oral exegetical traditions concerning Jewish law, contained in the first portion of the T alm ud .

402 GLOSSARY

Morpheme The smallest meaningful unit in a language, which may or may not be a freestanding word. Morphology The study of m o r ph e m e s. Non-Standard Language A variety of language that may depart from the st an d ar d l an g uag e in various ways, depending upon the speaker or situation. Old Greek (OG) The original translation of any given book of the Hebrew Bible into Greek, in contrast to the later revisional or recensional texts. No actual OG manuscript is known to survive, but it is theoretically possible to r e c o n st r uc t the OG text by means of t e x t ual c r it ic ism . The Göttingen c r it ic al e d it io n of the Se pt uag in t represents the leading attempt at such reconstruction. Also see ar c h e t ype . Onomastics The study of personal names. Oriental Orthodox Churches Christian Churches that split off from the Byzantine commonwealth as the result of the great Christological controversies of fifth century CE. These Churches include the Coptic Church, the Ethiopian Church, the Armenian Church, the Syriac Orthodox Church, the Assyrian Church of the East, and their daughter Churches. Orthography The study of spelling practices. Paleography The study of the written form of g r aph s and g r aph e m e s. Paleo-Hebrew The script used by Jewish scribes prior to the Persian period, at which time they adopted the Aramaic square script. Also known as Old Hebrew, paleo-Hebrew resembles Phoenician script, to which it is closely related, and is quite similar to what is found in the Torah scrolls of the Samaritans. It continued to be used on a limited basis until the first century CE and is found in some of the Torah fragments recovered from q um r an (although the majority of the Dead Sea Scrolls use the square script). In some ancient Jewish manuscripts, including Greek texts, only the tetragrammaton is written in paleo-Hebrew, which may reflect the archaizing tendency that is also evident in ancient Jewish coinage. Palimpsest A parchment or other writing material on which the original text has been effaced (either by scraping or washing) and then overwritten, a practice that was particularly common in the early Middle Ages. In some cases, the lower/earlier script of the palimpsest remains partially legible, and there is a possibility of recovering the text. Scholars have employed various techniques to this end, including, most recently, digital photography, which has led to new textual discoveries. Paranomasia The rhetorical technique of close repetition of words deriving from the same lexical root, often as a play on words. Parataxis The linking of independent clauses by means of a coordinating conjunction, as is typical of Hebrew syntax (compare to h ypo t ax is). Pentateuch The first five books of the canonical Hebrew Bible/Old Testament: Genesis, Exodus, Leviticus, Numbers, and Deuteronomy. Pentecostarion A collection of liturgical texts read or sung in the Orthodox tradition during the fifty-day period following Easter Sunday. Peshitta The translation of the Bible into Syriac, for the Old Testament this was mostly from a Hebrew so ur c e t e x t but also with readings that agree with the Se pt uag in t against the Hebrew. Philo of Alexandria A prolific Jewish writer from Alexandria in the first century CE. Philo used the Se pt uag in t almost exclusively, arguing that it had sacred status and that allegorical interpretation of the text revealed its spiritual meanings. Phone A distinct speech sound, in formal notation enclosed by the sigla [ ].

GLOSSARY 403

Phoneme The minimal perceptual set of ph o n e s that, if interchanged in a given environment, would change the meaning of a m o r ph e m e or word. For example, the English phones [s] and [z], when interchanged between sit and zit, distinguish two different words and therefore express two different phonemes: /s/ and /z/, respectively. In formal notation, a phoneme is enclosed by the sigla / /. Prophetologion A collection of liturgical Old Testament readings of the Orthodox Church. Proto-Masoretic text A consonantal text of the Hebrew Bible from the Second Temple period that was preserved as the standard text by the Masoretes in the Middle Ages (see M aso r e t ic t e x t ). Many scrolls from the Judean Desert, including from Q um r an , represent this tradition, although other textual traditions also existed. Along these lines, the Hebrew so ur c e t e x t of the Se pt uag in t is known to have differed at points from the proto-Masoretic text and agreed with Hebrew texts at Qumran. Pseudepigrapha A term that refers to a body of Jewish texts produced between c. 200 BCE and 200 CE that extends beyond even the A po c r yph a /d e ut e r o c an o n ic al bo o k s of the Catholic and Orthodox Old Testament canons. In the Orthodox world these books are usually called A po c r yph a . Ptolemy I One of Alexander the Great’s generals, who ruled over Egypt from 305/4 BCE to 283 BCE and periodically competed with the Se l e uc id s for control of territory, including ancient Palestine. Ptolemy II Philadelphus The ruler of Ptolemaic Egypt from 283 to 246 BCE, after Pt o l e m y i . Quire A section of pages within a c o d e x . The scribe would take a number of individual sheets of parchment or paper (“leaves,” Latin folia), and make them suitable for writing by forming them into something like a booklet. There were two methods. Either one took a stack of leaves (typically four) and folded it in half to form a bifolio, or else one folded the stack in half again twice, resulting in an octavo. Most codices contain several quires bound together. Qumran Qumran is a site in the Judean Desert near the Dead Sea where thousands of fragments of Hebrew, Aramaic, and Greek manuscripts were discovered in caves between 1947 and 1956. The manuscripts include writings of the Se pt uag in t and Hebrew Bible corpora, copies of the D e ut e r o c an o n ic al bo o k s and Pse ud e pig r aph a , as well as a variety of other types of literature, such as legal and wisdom texts, writings for worship, eschatological works, and commentaries. Quote/quotation A deliberate repetition of words from another source marked either by a particular introductory formula (e.g., “as it is written” or “as Isaiah said”) or by a definite break in grammar, such as a shift from second to third person. See also c o mpo sit e c it at io n and c o n fl at e d q uo t at io n . Realization The actual, physical production of a ph o n e m e . Recension A totally or almost totally new edition of an extant text that involves linguistic, stylistic, and literary modifications. In the case of the Se pt uag in t , recension involves comparison with a Hebrew so ur c e t e x t to adapt the extant text for a different linguistic and/or social context. Reception history See h ist o r y o f r e c e pt io n . Retroversion The reconstruction or “reverse translation” of the wording of the so ur c e t e x t of a translation on the basis of lexical equivalences in the Se pt uag in t and M aso r e t ic t e x t corpora, scholarly intuition, and the support from external sources, such as the Samaritan Pentateuch, Q um r an scrolls, and quotations in extra-biblical writings. The

404 GLOSSARY

reconstructed readings should be plausible from linguistic and textual points of view. In other words, it must be clear how the reconstructed reading developed from those in the available Hebrew-Aramaic manuscripts, or vice versa. Revision Linguistic modification of an extant text. In the case of the Se pt uag in t , revision may or may not involve comparison with the Hebrew text (compare to R e c e n sio n ). Romaniote A distinct branch of Judaism that traces its roots back to the Greek-speaking Jewish communities of the Greco-Roman period. Its language, customs, and liturgy vary significantly from those of the Sephardic and Ashkenazic Jewish traditions. Liturgical differences include their order of service, style of chanting, celebration of the holidays, and use of liturgical poems composed in a mixture of Greek and Hebrew. Romaniote Judaism flourished during the Byzantine period. However, with the arrival of large numbers of Sephardic Jews in the Balkans after their expulsion from Spain in 1492, much of the Romaniote population was eventually assimilated to the language (Ladino) and customs of the more dominant Spanish tradition. Scholium (pl. scholia, i.e., “small annotations”) A commentary or annotation added in the margin of a text to illustrate or explain it in different ways (grammatical, historical, theological, etc.). Mainly used in reference to Greek and Latin texts. Scriptio continua A way of writing without spaces or other markings between words or sentences. This type of writing is most common in classical Greek and Latin manuscripts. Second Temple Judaism (Hellenistic Judaism) The period of Judaism beginning from the construction of the Second Temple after the return from exile (late sixth century BCE) to its destruction in 70 CE by the Romans. The time frame spans the Persian period, the Hellenistic period, and the early Roman period, and it encompasses Jews living throughout the ancient Mediterranean, Europe, Africa, and the Near East. Seleucids The name of the Hellenistic dynasty stemming from Seleucus I, one of the generals of Alexander the Great. The Seleucid Empire was often in competition with the Ptolemies in Egypt. Semitism See H e br aism . Sentence name An an t h r o po n ym involving predication and, unless shortened, usually comprised from multiple words. For example, Hebrew ‫“( אֲמַ ְריָהּו‬Amariah”), literally “YHWH speaks” from ‫ יהוה‬+ ‫אמר‬. Septuagint (LXX) A term used in modern scholarship to refer to the corpus of Greek translations of the Hebrew Bible, along with certain other Jewish scriptural writings that were originally composed in Greek. This corpus is often delineated in extent (somewhat arbitrarily) in terms of the contents of the c r it ic al e d it io n of Rahlfs (1935 and/or 2006), although sometimes in scholarly writing this delimitation is only implied or is simply assumed. Some scholars prefer to use the term “Septuagint” strictly in relation to the O l d G r e e k translation of the Pe n t at e uc h , which was translated in Alexandria in the third century BCE. Source language The language(s) of the so ur c e t e x t (compare to t ar g e t l an g uag e ). Source text The text that is being translated from a so ur c e l an g uag e into a t ar g e t l an g uag e , the product of which is known as the t ar g e t t e x t . The Hebrew Bible is the source text of the Se pt uag in t , just as the Septuagint is the source text of the V e t us L at i n a. See also V o r l ag e . Standard Language A variety of language that is considered (or made) normative within a given society, usually with high prestige and used across various dialectal boundaries.

GLOSSARY 405

Stemma codicum The graphic representation used in t e x t ual c r it ic ism to show the relationships among (groups of ) manuscripts, using a model similar to a genealogical tree. It also contains data related to the chronological development of the texts, with the textual ar c h e t ype at its peak. Symmachus One of t h e T h r e e early Jewish revisers of the Se pt uag in t . Symmachus likely revised the Greek translation of the Hebrew Scriptures in Palestine at the end of the second century CE. His r e v isio n can be described as aiming at elegant and idiomatic Greek language, while at the same time trying to accurately represent the meaning of the Hebrew so ur c e t e x t . Synchronic A conceptual perspective that considers the whole system of entities, features, and structures at any given point in time (compare to d iac h r o n ic ). Syro-Hexapla The Syriac translation of the H e x apl ar ic R e c e n sio n . The ancient witnesses to the Syro-Hexapla preserve many of Origen’s sigla and thus provide valuable evidence for t e x t ual c r it ic ism . Talmud The authoritative collection of Jewish civil and ceremonial law, containing the M ish n ah and G e m ar a , and preserved in the Babylonian and Palestinian/Jerusalem versions. Tannaitic The rabbinic authorities cited in the M ish n ah are referred to as Tannaim. The term “Tannaitic” is used in in rabbinic historiography to designate the period between the destruction of the Second Temple in 70 CE and the death of Patriarch Judah I around 225 CE, at which time the m ish n ah is believed to have been published. Hence it is also referred to as the Mishnaic period. Target language The language of the t ar g e t t e x t (compare to so ur c e l an g uag e ), which in the case of the Se pt uag in t is K o in e Greek. Target text The text that is the result of the translation process from a so ur c e t e x t in another language. The Se pt uag in t is the target text of the Hebrew Bible and Greek is the t ar g e t l an g uag e (compare to so ur c e l an g uag e ). Targums A broad term for the Aramaic translations of the Hebrew Bible, originating as ex tempore translation of the Hebrew and only being written down some centuries later. Textual criticism The practice of comparing manuscripts witnessing to the same text/ writing, identifying variant readings across the evidence, and constructing hypotheses on the textual history of the text/writing. Textual criticism of the Se pt uag in t aims to explain the diversity of readings in the Greek manuscripts from a d iac h r o n ic perspective and thereby to reconstruct the O l d G r e e k . Also see ar c h e t ype . Textus receptus A successively copied or printed text that receive general acceptance in a community. Theodotion One of t h e T h r e e early Jewish revisers of the Se pt uag in t . Theodotion was a Jewish scholar who lived in the middle of the second century CE. The Theodotionic r e v isio n displays a tendency to modify the Greek text toward the Hebrew pr o t o m aso r e t ic text and is therefore also often discussed together with the K ai g e revision. The Three A term originating among early Christians to designate the separate translations or r e c e n sio n s of the Se pt uag in t produced under the names aq uil a , symm ac h us, and T h e o d o t io n , and included in Origen’s H e x apl a . Toponym The name of a place. Translation technique A term used to refer to the characteristics and style of a given translator as evident in a t ar g e t t e x t , often with attention to lexical choice and word order in comparison with the so ur c e t e x t .

406 GLOSSARY

Transliteration The act of transcribing the written form of a so ur c e l an g uag e entity into the t ar g e t l an g uag e . This term encompasses any word that is not part of the target language but is nevertheless written down in the target language. For example, the English word “Cherubim” is a transliteration of the Hebrew word ‫כרובים‬, which may also be transliterated according to scholarly conventions, as in kǝrûbîm (SBL). Trifaria varietas An expression used by Jerome in reference to the three main branches of the Se pt uag in t text that were generally known in his time. Triodion A collection of liturgical texts read or sung in the Orthodox tradition during the Great Lent. Uncial A form of m ajusc ul e Greek script in which curves are freely introduced to facilitate the use of stylus and ink on absorbent materials. Most Greek papyri and vellum manuscripts between the fourth and the ninth centuries CE feature uncial script, but then shift to m in usc ul e script. Vetus Latina (OL) The ancient Latin translations of the Se pt uag in t produced around the second century CE, prior to Jerome’s work, to produce the V ul g at e . Vorlage A German noun meaning “exemplar” or “model.” This term is used in Se pt uag in t scholarship to refer to the specific (if hypothetical) Hebrew or Aramaic manuscript that a given ancient translator was working from to produce the t ar g e t t e x t . Vulgate The Latin translation of the Bible, including most of the A po c r yph a , produced by Jerome (c. 347–420 BCE). For the Old Testament, Jerome translated the Hebrew Bible instead of the Se pt uag in t , a controversial move in his day.

CONTRIBUTORS

James K. Aitken (PhD, University of Cambridge) is Reader in Hebrew and Early Jewish Studies at the Faculty of Divinity, University of Cambridge. His research focuses on Second Temple Judaism, including the use of Hebrew and Greek among Jews. He is currently investigating the origins of the Septuagint, and especially its place within Egyptian society of the time. Publications include The Semantics of Blessing and Cursing in Ancient Hebrew (Peeters, 2007), No Stone Unturned: Greek Inscriptions and Septuagint Vocabulary (Eisenbrauns, 2014), and the T&T Clark Companion to the Septuagint (T&T Clark, 2015). Eberhard Bons (DPhil, University of Mainz; DTh, Sankt Georgen Graduate School of Philosophy and Theology; Habilitation in Biblical Studies, University of Strasbourg) is Professor of Old Testament at the University of Strasbourg, France. He is the editor of the Historical and Theological Lexicon of the Septuagint (Mohr Siebeck). His research focuses on Psalms, Prophets, wisdom literature, Jewish literature of the Hellenistic and Roman era, and the Septuagint. Cameron Boyd-Taylor (PhD, University of Toronto) is Senior Research Associate of the John William Wevers Institute for Septuagint Studies at Trinity Western University (Langley, British Columbia, Canada) and author of Reading Between the Lines (Peeters, 2011). José Manuel Cañas Reíllo is a researcher at the Instituto de Lenguas y Culturas del Mediterráneo y Oriente Próximo (Consejo Superior de Investigaciones Científicas, Spain). His field of research is the Biblical Philology (Vetus Latina and LXX), and especially textual criticism. Claude E. Cox (PhD, University of Toronto) is Adjunct Professor of Old Testament and Hebrew at McMaster Divinity College and chaplain in a long-term care facility in Barrie, Ontario. He is a specialist in the Armenian subversion and author of Hexaplaric Materials Preserved in the Armenian Version (Scholars Press, 1986) and Aquila, Symmachus and Theodotion in Armenia (Scholars Press, 1996). Marieke Dhont (PhD, KU Leuven and Université catholique de Louvain) is a British Academy Postdoctoral Fellow at the University of Cambridge (2018–21) and the author of Style and Context of Old Greek Job (Brill, 2018). She specializes in Hellenistic Judaism, the Septuagint and Jewish-Greek literature, and Greek semantics. Christopher J. Fresch (PhD, University of Cambridge) is Lecturer in Biblical Languages/ Old Testament at the Bible College of South Australia and co-editor of The Greek Verb Revisited (Lexham Press, 2016). His main areas of research are linguistics and Septuagint. Edmun L. Gallagher (PhD, Hebrew Union-College-Jewish Institute of Religion) is Associate Professor of Christian Scripture at Heritage Christian University (Florence,

408 CONTRIBUTORS

Alabama, USA). His main field of research is the history of biblical interpretation. He is author of Hebrew Scripture in Patristic Biblical Theory (Brill, 2012) and coauthor with John D. Meade of The Biblical Canon Lists from Early Christianity (Oxford, 2018). Peter J. Gentry (PhD, University of Toronto) is Donald L. Williams Professor of Old Testament Interpretation and Director of the Hexapla Institute at The Southern Baptist Theological Seminary. He has served on the faculty of Toronto Baptist Seminary and Bible College and also taught at the University of Toronto, Heritage Theological Seminary, and Tyndale Seminary. He has prepared a critical edition of Ecclesiastes for the Göttingen Septuagint. W. Edward Glenny (ThD, Dallas Theological Seminary; PhD, University of Minnesota) is Professor of New Testament and Greek at University of Northwestern–St. Paul (MN), where he has taught for the past twenty-two years. He has published on Septuagint translation technique (Finding Meaning in the Text: Translation Technique and Theology in the Septuagint of Amos) and on the Greek Twelve Prophets (Hosea, Amos, and Micah in the Brill Septuagint Commentary Series). Robert J. V. Hiebert (PhD, University of Toronto) is Professor of Old Testament in the Graduate School of Theological Studies at Trinity Western University (Langley, British Columbia, Canada). He also serves as the Director of the John William Wevers Institute for Septuagint Studies, the President of the International Organization for Septuagint and Cognate Studies, and the Joint-Editor-in-Chief of the Society of Biblical Literature Commentary on the Septuagint (SBLCS). He is the author and editor of books and articles on a range of subjects in the areas of Septuagint and Biblical Studies. He is currently preparing the critical edition of 4 Maccabees for the Göttingen Septuagint series and a commentary on Genesis for the SBLCS series. Jennifer Brown Jones (PhD, McMaster Divinity College) is an Instructor at Liberty University. Her research interests include the Psalter and the Book of the Twelve in both their Greek and Hebrew traditions. Gideon R. Kotzé (PhD, University of Groningen) is Professor in the Research Focus Area, Ancient Texts at the Faculty of Theology of North-West University, South Africa. His main fields of study are textual criticism of early Jewish literature and the Dead Sea Scrolls. Some of his recent major publications include The Qumran Manuscripts of Lamentations (Brill 2013), Septuagint, Sages, and Scripture (Brill 2016), Septuagint, Sages, and Scripture (co-edited with R. X. Gauthier and G. J. Steyn; Brill, 2016), and XVI Congress of the International Organization for Septuagint and Cognate Studies: Stellenbosch, 2016 (coedited with L. C. Jonker and C. M. Maier; Brill, 2017). Ville Mäkipelto (PhD, University of Helsinki) is a postdoctoral researcher in the Faculty of Theology at the University of Helsinki in the Centre of Excellence “Changes in Sacred Texts and Traditions” (www.cstt.fi). His research deals with the Historical Books of the Old Testament, and the usage of the Septuagint and other early textual traditions in reconstructing the textual history of the Hebrew Bible. Mäkipelto is the author of Uncovering Ancient Editing (De Gruyter, 2018). Scott Mandelbrote is Fellow, Director of Studies in History, and Perne Librarian at Peterhouse, Cambridge. He gave the Grinfield Lectures on the Septuagint in the University of Oxford in 2003 and 2004 and has since published widely on the early modern study

CONTRIBUTORS 409

of the Septuagint. His other publications include The Bible, the Ark, the Tower, the Temple (with Jim Bennett: Oxford, 1998); Footprints of the Lion: Isaac Newton at Work (Cambridge, 2001); and, as editor, Nature and Scripture in the Abrahamic Religions (with Jitse van der Meer: Leiden, 2008) and The Reception of Isaac Newton in Europe (with Helmut Pulte: London, 2019). John D. Meade (PhD, The Southern Baptist Theological Seminary) is Associate Professor of Old Testament and co-director of the Text & Canon Institute at Phoenix Seminary. He is coauthor with Edmun L. Gallagher of The Biblical Canon Lists from Early Christianity (Oxford, 2018) and A Critical Edition of the Hexaplaric Fragments of Job 22–42 (Peeters, 2019). Steve Moyise (PhD, University of Birmingham) is Visiting Professor of Newman University, Birmingham, and author of Evoking Scripture: Seeing the Old Testament in the New (T&T Clark, 2008) and co-editor with Maarten Menken of the series The New Testament and the Scriptures of Israel (T&T Clark). Pete Myers is lecturer in Hebrew and Old Testament at the Ethiopian Graduate School of Theology, Addis Ababa. His PhD on the text and grammar of transcriptions in 2 Esdras was completed at the University of Cambridge. His recent publications include peerreviewed articles on the text critical contextualization of transcriptions in corrections on Sinaiticus and the implications of the Greek alphabet for reconstructing the phonology of Canaanite sibilants. Stanley E. Porter (PhD, University of Sheffield) is President, Dean, and Professor of New Testament at McMaster Divinity College (Hamilton, Ontario, Canada). He is the author of nearly thirty volumes, as well as editor of nearly ninety other volumes. He has authored hundreds of journal articles, chapters, and related scholarly items on a wide range of subjects. His primary areas of research are Greek language and linguistics, Pauline studies, Johannine studies, Historical Jesus research, and hermeneutics and the history of interpretation. His interest in the Septuagint grows out of his interest in Greek linguistics and the history of interpretation. Patrick Pouchelle (PhD, University of Strasbourg) is Associate Professor of Old Testament at Centre Sevres (Paris, France), author of Dieu éducateur: Une nouvelle approche d’un concept de la théologie biblique entre Bible Hébraïque, Septante et littérature grecque classique (Mohr Siebeck, 2015), and co-editor with Eberhard Bons of The Psalms of Solomon: Language, History, Theology (SBL Press, 2018). His research focuses on the Septuagint, especially lexicography, and the Pseudepigrapha of the Old Testament, especially Psalms of Solomon. William A. Ross (PhD, University of Cambridge) is Assistant Professor of Old Testament at Reformed Theological Seminary, author of Septuagint Lexicography and Language Change in Greek Judges (SBL Press, forthcoming), and co-editor of Septuaginta: A Reader’s Edition (Hendrickson 2018). His research focuses on the Septuagint, postclassical (Koine) Greek, and the history of biblical philology. John Screnock (PhD, University of Toronto) is Research Fellow in Hebrew Bible at the University of Oxford, and author of Traductor Scriptor (Brill, 2017). He is interested in the development of texts, languages, and translations in Jewish antiquity.

410 CONTRIBUTORS

Mikhail G. Seleznev (PhD, Moscow State University) is Associate Professor at the HSE University (Moscow, Russia). His research is supported by the HSE Academic Fund Programme. Myrto Theocharous (PhD, University of Cambridge) is Lecturer of Hebrew and Old Testament at the Greek Bible College in Athens, Greece. She is the author of Lexical Dependence and Intertextual Allusion in the Septuagint of the Twelve Prophets (T&T Clark, 2012), and various other articles in both Greek and English. Benjamin G. Wright is University Distinguished Professor in the Department of Religion Studies at Lehigh University (Bethlehem, PA, USA). His primary areas of research are Jewish Wisdom Literature, especially the Book of Ben Sira, and ancient translation, especially the Septuagint. His most recent book is The Letter of Aristeas: “Aristeas to Philocrates” or “On the Translation of the Law of the Jews” (Berlin: Walter de Gruyter, 2015).

REFERENCES

Adams, S. A. (2014), Baruch and the Epistle of Jeremiah, SEPT, Leiden: Brill. Adams, S. A. and S. M. Ehorn, eds. (2016), Composite Citations in Antiquity, Volume 1: Jewish, Graeco-Roman, and Early Christian Uses, London: Blooomsbury T&T Clark. Adams, S. A. and S. M. Ehorn, eds. (2018), Composite Citations in Antiquity, Volume 2: New Testament Uses, London: Bloomsbury T&T Clark. Adrados, F. R. (1980–2007), Diccionario griego-español, Madrid: Instituto Antonio de Nebrija. Aejmelaeus, A. (1982), Parataxis in the Septuagint: A Study of the Renderings of the Hebrew Coordinate Clauses in the Greek Pentateuch, AASF Dissertationes Humanarum Litterarum 11, Helsinki: Suomalainen Tiedeakatemia. Aejmelaeus, A. (1993), On the Trail of the Septuagint Translators: Collected Essays, Kampen: Kok Pharos. Revised and expanded in Aejmelaeus (2007a). Aejmelaeus, A. (2001), “What We Talk about When We Talk about Translation Technique,” in B. A. Taylor (ed.), X Congress of the IOSCS, Oslo, 1998, 531–52, SBLSCS 51, Atlanta: SBL Press. Aejmelaeus, A. (2007a), On the Trail of the Septuagint Translators: Collected Essays, CBET 50, Leuven: Peeters. Revised and expanded edition of Aejmelaeus (1993). Aejmelaeus, A. (2007b), “Translation Technique and the Intention of the Translator,” in A. Aejmelaeus (ed.), On the Trail of the Septuagint Translators: Collected Essays, 59–69, CBET 50, Leuven: Peeters. Aejmelaeus, A. (2007c), “What Can We Know about the Hebrew Vorlage?” in A. Aejmelaeus (ed.), On the Trail of the Septuagint Translators: Collected Essays, 71–106, CBET 50, Leuven: Peeters. Aejmelaeus, A. (2013), “The Septuagint and Oral Translation,” in M. K. H. Peters (ed.), XIV Congress of the International Organization for Septuagint and Cognate Studies, Helsinki, 2010, 5–13, SBLSCS 59, Atlanta: SBL Press. Aejmelaeus, A. (2017a), “‘Does God Regret? A Theological Problem That Concerned the Kaige Reviser,’” in A. Aejmelaeus and T. Kauhanen (eds.), The Legacy of Barthélemy: 50 Years after Les Devanciers d’Aquila, 41–53, DSI 9, Göttingen: Vandenhoeck & Ruprecht. Aejmelaeus, A. (2017b), “Kaige Readings in a Non-Kaige Section in 1 Samuel,” in A. Aejmelaeus and T. Kauhanen (eds.), The Legacy of Barthélemy: 50 Years after Les Devanciers d’Aquila, 169–84, DSI 9, Göttingen: Vandenhoeck & Ruprecht. Aejmelaeus, A. (2017c), “Textual History of the Septuagint and the Principles of Critical Editing,” in A. Piquer Otero and P. A. Torijano (eds.), The Text of the Hebrew Bible and Its Editions, 160–79, THBSup 1, Leiden: Brill. Aejmelaeus, A. and T. Kauhanen, eds. (2017), The Legacy of Barthélemy: 50 Years after Les Devanciers d’Aquila, DSI 9, Göttingen: Vandenhoeck & Ruprecht. Aejmelaeus, A. and U. Quast (2000), Der Septuaginta-Psalter und seine Tochterübersetzungen. Symposium in Göttingen 1997, Abhandlungen der Akademie der Wissenschaftern in Göttingen, Philologische-Historische Klasse Dritte Folge 230, MSU 24, Göttingen: Vandenhoeck & Ruprecht.

412 REFERENCES

Aichele, G. and G. A. Phillips (1995), “Introduction: Exegesis, Eisegesis, Intergesis,” Semeia, 69/70: 7–18. Aitken, J. K. (2005), “Rhetoric and Poetry in Greek Ecclesiastes,” BIOSCS 38: 55–77. Aitken, J. K. (2009), “Jewish Use of Greek Proverbs,” in N. de Lange, J. Krivoruchko, and C. Boyd-Taylor (eds.), Jewish Reception of Greek Bible Versions, 53–77, TSMEMJ 23, Tübingen: Mohr Siebeck. Aitken, J. K. (2013), “Neologisms: A Septuagint Problem,” in J. K. Aitken, J. M. S. Clines and C. M. Maier (eds.), Interested Readers: Essays in the Hebrew Bible in Honor of David J. A. Clines, 315–29, Atlanta: SBL Press. Aitken, J. K. (2014a), “Jewish Worship amid Greeks: The Lexical Context of the Old Greek Psalter,” in R. T. McLay (ed.), The Temple in Text and Tradition, 48–70, London: T&T Clark. Aitken, J. K. (2014b), No Stone Unturned: Greek Inscriptions and Septuagint Vocabulary, CSHB 5, Winona Lake: Eisenbrauns. Aitken, J. K. (2014c), “The Language of the Septuagint and Jewish-Greek Identity,” in J. K. Aitken and J. Carleton Padget (eds.), The Jewish-Greek Tradition in Antiquity and the Byzantine Empire, 120–34, Cambridge: Cambridge University Press. Aitken, J. K. (2015a), “The Origins of KAI GE,” in J. K. Aitken and T. V. Evans (eds.), Biblical Greek in Context: Essays in Honour of John A. L. Lee, 21–40, BiTS 22, Leuven: Peeters. Aitken, J. K., ed. (2015b), T&T Clark Companion to the Septuagint, London: Bloomsbury T&T Clark. Aitken, J. K. (2015c), “Why Is the Giraffe Kosher? Exoticism in Dietary Laws of the Second Temple Period,” BN 164: 21–34. Aitken, J. K. (2015d), “Wisdom of Solomon,” in J. K. Aitken (ed.), The T&T Clark Companion to the Septuagint, 401–9, London: Bloomsbury T&T Clark. Aitken, J. K. (2016), “The Septuagint and Egyptian Translation Methods,” in W. Kraus, M. N. van der Meer, and M. Meiser (eds.), XV Congress of the International Organization for Septuagint and Cognate Studies, Munich, 2013, 269–94, SBLSCS 64, Atlanta: SBL Press. Aitken, J. K. and J. Carleton Paget, eds. (2014), The Jewish-Greek Tradition in Antiquity and the Byzantine Empire, Cambridge: Cambridge University Press. Aitken, J. K. and L. Cuppi (2015), “Proverbs,” in J. K. Aitken (ed.), The T&T Clark Companion to the Septuagint, 341–55, London: Bloomsbury T&T Clark. Aitken, J. K. and T. V. Evans, eds. (2015), Biblical Greek in Context: Essays in Honour of John A. Lee, BiTS 22, Leuven: Peeters. Aland, B., K. Aland, J. Karavidopoulos, C. M. Martini, M. Black, A. Wikgren, and B. Metzger, eds. (2001), The Greek New Testament. Fourth Revised Edition, Stuttgart: Beutsche Bibelgesellschaft/United Bible Societies. Aland, B., K. Aland, J. Karavidopoulos, C. M. Martini, and B. Metzger, eds. (2014), The Greek New Testament. Fifth Revised Edition, Stuttgart: Beutsche Bibelgesellschaft/United Bible Societies. Albl, M. C. (1999), “And Scripture Cannot Be Broken”: The Form and Function of the Early Christian Testimonia Collections, NovTSup 96, Leiden: Brill. Albrecht, F. (2012), “Cantacuzenus, Mathaeus Asanes,” in F. W. Bautz and T. Bautz (eds.), Biographisch-Bibliographisches Kirchenlexikon, Volume 33: 212–16, Nordhausen: Verlag Traugott Bautz. Albrecht, F. (2013), “Malachias Monachus,” in F. W. Bautz and T. Bautz (eds.), BiographischBibliographisches Kirchenlexikon, Volume 34: 882–4, Nordhausen: Verlag. Albrecht, F., ed. (2018), Septuaginta. Vetus Testamentum Graecum. XII.3. Psalmi Salomonis, Göttingen: Vandenhoeck & Ruprecht.

REFERENCES 413

Alekseev, A. A. (1999), Textual Criticism of the Slavonic Bible (in Russian) [Алексеев, А. А. Текстология Славянской Библии], St. Petersburg/Köln: Bulanin/Bölau. Alekseev, A. A. (2014), “Russian-Jewish Literary Ties in the Kievan Period. Results and Prospects of the Research,” (in Russian) [Алексеев, А. А., Русско-еврейские литературные связи Киевской эпохи. Результаты и перспективы исследования, in В. Москович, М. Членов, and А. Торпусман (eds.), Кенааниты: Евреи в средневековом славянском мире, 166–182, Москва-Иерусалим: Мосты культуры. Alexander, P. (1999), “How Did the Rabbis Learn Hebrew?” in W. Horbury (ed.), Hebrew Study from Ezra to Ben-Yehuda, 71–89, Edinburgh: T&T Clark. Alexander, P. (2007), “The Formation of the Biblical Canon in Rabbinic Judaism,” in P. S. Alexander and J.-D. Kaestli (eds.), The Canon of Scripture in Jewish and Christian Tradition, 57–80, Prahins: Éditions du Zebre. Alexander, P. (2014), “The Rabbis, the Greek Bible and Hellenism,” in J. K. Aitken and J. Carleton Padget (eds.), The Jewish-Greek Tradition in Antiquity and the Byzantine Empire, 229–46, Cambridge: Cambridge University Press. Alexanian, J. M. (1992), “Versions, Ancient (Armenian),” in D. N. Freedman (ed.), The Anchor Bible Dictionary, Volume 6: 805–8, New York: Doubleday. Alexanian, J. M. (2012), The Ancient Armenian Text of the Acts of the Apostles, CSCO 643, Scriptores Armeniaci 31, Leuven: Peeters. Allen, W. C. (1907), A Critical and Exegetical Commentary on the Gospel According to S. Matthew, 2nd edn., ICC, Edinburgh: T&T Clark. Allen, W. S. (1987), Vox Graeca: A Guide to the Pronunciation of Classical Greek, 3rd edn., Cambridge: Cambridge University Press. Alter, R. (1981), The Art of Biblical Narrative, London: George Allen and Unwin. Alter, R. (1992), The World of Biblical Literature, London: SPCK. Anania, B. V., ed. (2001), Biblia sau Sfânta Scriptură, Bucuresti: Editura Institutului si de Misiune al Bisericii Ortodoxe Române. Andersen, R. G. (2000), Glossary of Greek Rhetorical Terms, CBET 24, Leuven: Peeters. Arar, D. (2005), “Le Pentateuque de Constantinople (1547): une traduction littérale?” Ph.D. dissertation, Université Paris-Sorbonne, Paris. Archer, G. L. and G. Chirichigno ([1983] 2005), Old Testament Quotations in the New Testament, Eugene: Wipf & Stock. Arnold, B. T. and J. H. Choi (2003), A Guide to Biblical Hebrew Syntax, Cambridge: Cambridge University Press. Arvaniti, A. (1999), “Standard Modern Greek,” JIPA 29: 167–72. Arvaniti, A. (2007), “Greek Phonetics: The State of the Art,” JGL 8: 97–208. Askin, L. (2018), “What Did Ben Sira’s Bible and Desk Look Like?” in G. V. Allen and J. A. Dunne (eds.), Ancient Readers and Their Scriptures: Engaging the Hebrew Bible in Early Judaism and Christianity, 3–26, AJEC 107, Leiden: Brill. Aslanov, C. (1998), “Les gloses judéo-hellénistiques du ‘Commentaire de Reuel sur Ezéchiel’ découvert à la Geniza du Caire,” REJ 157: 7–45. Aslanov, C. (1999a), “La place du Venetus Graecus dans l’histoire des traductions grecques de la Bible,” RPLHA 3/73: 155–74. Aslanov, C. (1999b), “The Judeo-Greek and Ladino columns in the Constantinople edition of the Pentateuch (1547): A Linguistic Commentary on Gen. 1:1–15,” REJ 158: 385–97. Assan-Dhôte, I. and J. Moatti-Fine (2005), La Bible d’Alexandrie. XXV.2. Baruch, Lamentations, Lettre de Jérémie, BdA 25.2, Paris: Cerf. Assan-Dhôte, I. and J. Moatti-Fine (2009), La Bible d’Alexandrie. VIII. Ruth, BdA 8, Paris: Cerf. Augustine (1555), De Civitate Dei libri xxii, J. L. Vives (ed.), Basel: Froben.

414 REFERENCES

Auld, G. (2005), Joshua, SEPT, Leiden: Brill. Ausloos, H. (2017), “Sept défis posés à une thèologie de la Septante,” in L. C. Jonker, G. R. Kotzé, and C. M. Maier (eds.), Congress Volume Stellenbosch 2016, 228–50, VTSup 177, Leiden: Brill. Ausloos, H. and B. Lemmelijn, eds. (2020), Die Theologie der Septuaginta/The Theology of the Septuagint, LXX.H 5, Gütersloh: Gütersloher Verlagshaus. Auwers, J.-M. (2019), La Bible d’Alexandrie. XIX. Le Cantique des Cantiques, BdA 19, Paris: Cerf. Avemarie, F. (2009), “Interpreting Scripture Through Scripture: Exegesis Based on Lexematic Association in the Dead Sea Scrolls and the Pauline Epistles,” in F. G. Martínez (ed.), Echoes from the Caves: Qumran and the New Testament, 83–102, Leiden: Brill. Awabdy, M. (2019), Leviticus, SEPT, Leiden: Brill. Backfish, E. H. P. (2019), Hebrew Wordplay and Septuagint Translation Technique in the Fourth Book of the Psalter, LHBOTS 682, New York: Bloomsbury T&T Clark. Bădiliţă, C., F. Băltăceanu, M. Broşteanu, and D. Sluşanschi, eds. (2004–8), Septuaginta, 8 volumes, Bucharest: Polirom. Baillet, M. (1962a), “Exode,” in M. Baillet, J. T. Milik, and R. de Vaux (eds.), Les “Petites Grottes” de Qumrân, 142–3, DJD 3, Oxford: Clarendon Press. Baillet, M. (1962b), “Lettre de Jérémie,” in M. Baillet, J. T. Milik, and R. de Vaux (eds.), Les “Petites Grottes” de Qumrân, 143, DJD 3, Oxford: Clarendon Press. Baird, W. (1992–2013), History of New Testament Research, 3 volumes, Minneapolis: Fortress. Balz, H. and G. Schneider (1990), Exegetical Dictionary of the New Testament, Edinburgh: T&T Clark. Bar-Asher, M. (2010), “The Verse ‫‘( ְשׁמַ ע יִ ְש ָׂראֵ ל‬Hear, O Israel’) in Greek Transcription on an Ancient Amulet,” JAJ 1: 227–32. Barone, F. P. (2009), “Pour une édition critique de la Synopsis Scripturae Sacrae du PseudoJean Chrysostome,” RPLHA 1/83: 7–19. Barr, J. (1961), The Semantics of Biblical Language, Oxford: Oxford University Press. Barr, J. (1967a), “St. Jerome and the Sounds of Hebrew,” JSS 12: 1–36. Barr, J. (1967b), “Vocalization and the Analysis of Hebrew among Ancient Translations,” in B. Hartmann, E. Jenni, E. Y. Kutscher, and V. Maag (eds.), Hebräische Wortforschung: Festschrift zum 80. Geburtstage von Walter Baumgartner, 1–11, VTSup 16, Leiden: Brill. Barr, J. (1968), Comparative Philology and the Text of the Old Testament, Oxford: Clarendon. Barr, J. (1979), The Typology of Literalism in Ancient Biblical Translations, MSU 15, Göttingen: Vandenhoeck & Ruprecht. Barr, J. (1989), The Variable Spellings of the Hebrew Bible, Schweich Lectures of the British Academy 1986, Oxford: Oxford University Press. Barr, J. (1990), “‘Guessing’ in the Septuagint,” in D. Fraenkel, U. Quast, and J. W. Wevers (eds.), Studien zur Septuaginta—Robert Hanhart zu Ehren: Aus Anlass seines 65. Geburtstages, 19–34, MSU 20, Göttingen: Vandenhoeck & Ruprecht. Barr, J. (2002), Review of Karen H. Jobes and Moisés Silva (2000), Invitation to the Septuagint, Grand Rapids: Baker Academic, RBL, Online: http://www.bookreviews.org. Barr, J. (2003), “Did the Greek Pentateuch Really Serve as a Dictionary for the Translation of the Later Books?” in M. F. J. Baasten and W. Th. van Peursen (eds.), Hamlet on a Hill: Semitic and Greek Studies Presented to Professor T. Muraoka on the Occasion of his SixtyFifth Birthday, 523–43, OLA 118, Leuven: Peeters. Barthélemy, D. (1963), Les Devanciers d’Aquila. Première publication intégrale du texte des fragments du Dodécaprophéton, VTSup 10, Leiden: Brill.

REFERENCES 415

Barthélemy, D. (1978), Études d’histoire du texte de l’Ancien Testament, OBO 21, Göttingen: Vandenhoeck & Ruprecht. Barthélemy, D. (1990), “Les Relations de la Complutensis avec le Papyrus 967,” in D. Fraenkel, U. Quast, and J. W. Wevers (eds.), Studien zur Septuaginta—Robert Hanhart zu Ehren: Aus Anlass seines 65. Geburtstages, 253–61, Göttingen: Vandenhoeck & Ruprecht. Barton, J. (1986), The Oracles of God: Perceptions of Ancient Prophecy in Israel after the Exile, Oxford: Oxford University Press. Batalden, S. K. (2013), Russian Bible Wars: Modern Scriptural Translation and Cultural Authority, Cambridge: Cambridge University Press. Batalden, S. K. (2017), “The Septuagint Challenge to Russian Old Testament Translation: Fault Lines in fin-de-siècle Russian Religious Culture,” in L. J. Frary and Th. G. Stavrou (eds.), Thresholds into the Orthodox Commonwealth: Essays in Honor of Theofanis G. Stavrou, 137–46, Bloomington: Slavica. Bauer, H. and P. Leander (1927), Grammatik des Biblisch-Aramäischen, Halle/Saale: Max Niemeyer Verlag. Bauer, H., P. Leander, and P. Kahle (1922), Historische Grammatik der hebräischen Sprache des alten Testamentes: Erste Band: Einleitung. Schriftlehre. Laut- und Formenlehre, Halle/Saale: Max Niemeyer Verlag. Bauer, W. (1988), Griechisch-Deutsches Wörterbuch zu den Schriften des Neuen Testaments und der frühchristlichen Literatur, 6th edn., Berlin: De Gruyter. Bauer, W., F. W. Danker, W. F. Arndt, and F. W. Gingrich (2000), A Greek-English Lexicon of the New Testament and Other Early Christian Literature, 3rd edn., Chicago: University of Chicago Press. Beale, G. K. (1999), The Book of Revelation, NIGTC, Grand Rapids: Eerdmans. Beale, G. K. and D. A. Carson, eds. (2007), Commentary on the New Testament Use of the Old Testament, Grand Rapids: Baker Academic. Beckwith, R. (1985), The Old Testament Canon for the New Testament Church, Grand Rapids: Eerdmans. Beekes, R. (2010), Etymological Dictionary of Greek, 2 volumes, LEIEDS 10, Leiden: Brill. Bellantuono, A. (2019), “Divine Epithets in Jewish-Hellenistic Literature,” Ph.D. dissertation, University of Strasbourg, Strasbourg. Ben-Ḥayyim, Z. (1988), “‫בשולי ׳׳חוק פיליפי׳׳‬,” Lĕšonénu 53: 113–20. Ben-Ḥayyim, Z. (2000), A Grammar of Samaritan Hebrew Based on the Recitation of the Law in Comparison with the Tiberian and Other Jewish Traditions, Winona Lake: Eisenbrauns. Benoit, P. (1963), “L’inspiration des Septante d’après les Pères,” in L’homme devant Dieu: mélanges offerts au père Henri de Lubac. Exégèse et Patristique, 169–87, Théologie. Études Publiées Sous La Direction De La Faculté De Théologie S. J. de Lyon-Fourvière 56, Paris: Aubier. Bentein, K. (2013), “Register and the Diachrony of Post-Classical and Early Byzantine Greek,” Revue belge de Philologie et d’Histoire 91: 5–44. Bentein, K. (2017), “Towards a Socio-Historical Analysis of Ancient Greek? Some Problems and Prospects,” Byzantios. Studies in Byzantine History and Civilization 12: 35–45. Bertram, G. (1936), “Die religiöse Umdeutung altorientalischer Lebensweisheit in der griechischen Übersetzung des ATs,” ZAW 54: 153–67. Bertram, G. (1939), “Der Sprachschatz der Septuaginta und der des hebräischen Alten Testaments,” ZAW 57: 85–101. Bertram, G. (1957), “Praeparatio Evangelica in der Septuaginta,” VT 7: 225–49.

416 REFERENCES

Beyer, K. (1986), The Aramaic Language: Its Distribution and Subdivisions, J. F. Healey (trans.), Göttingen: Vandenhoeck & Ruprecht. Bickerman, E. (1959), “The Septuagint as a Translation,” PAAJR 28: 1–39. Biel, J. C. (1779–80), Novus Thesaurus philologicus, sive lexicon in LXX et alios interpretes et scriptures apocryphos Veteris Testamenti, 3 volumes, The Hague: Bouvink. Bird, M. F. (2012), 1 Esdras, SEPT, Leiden: Brill. Birdsall, J. N. (1992), “Versions, Ancient (Georgian),” in D. N. Freedman (ed.), The Anchor Bible Dictionary, Volume 6: 810–13, New York: Doubleday. Blau, J. (1982), On Polyphony in Biblical Hebrew, Proceedings of the Israel Academy of Sciences and Humanities 6.2, Jerusalem: Ahva Press, Ltd. Blau, J. (2010), Phonology and Morphology of Biblical Hebrew: An Introduction, LSAWS 2, Winona Lake: Eisenbrauns. Blondheim, D. S. (1924), “Échos du judéo-hellénisme. Étude sur l'influence de la Septante et d'Aquila sur les versions néo-grecques des Juifs,” REJ 78: 1–14. Bogaert, P.-M. (1992), “Versions, Ancient (Latin),” in D. N. Freedman (ed.), The Anchor Bible Dictionary, Volume 6: 799–803, New York: Doubleday. Bogaert, P.-M. (2013), “The Latin Bible,” in J. Carleton Paget and J. Schaper (eds.), The New Cambridge History of the Bible, Volume 1: From the Beginnings to 600, 505–26, Cambridge: Cambridge University Press. Bogaert, P.-M. (2014), “Les frontiers du canon de l’Ancien Testament dans l’Occident latin,” in R. Gounelle and J. Joosten (eds.), La Bible juive dans l’Antiquité, 41–95, Lausanne: Éditions du Zebre. Boned Colera, P. and J. R. Somolinos, eds. (1998), Repertorio bibliográfico de la lexicografía griega, ADGE 3, Madrid: Instituto de filología CSIC. Bons, E. (2011), “Rhetorical Devices in the Septuagint Psalter,” in E. Bons and T. J. Kraus (eds.), Et sapienter et eloquenter: Studies on Rhetorical and Stylistic Features of the Septuagint, 69–79, FRLANT 241, Göttingen: Vandenhoeck & Ruprecht. Bons, E., ed. (2020), The Historical and Theological Lexicon of the Septuagint, Volume 1, Tübingen: Mohr Siebeck. Bons, E. and R. Brucker (2019), “Psalms,” in S. Kreuzer (ed.), Introduction to the Septuagint, D. A. Brenner and P. Altman (trans.), 297–316, Waco: Baylor University Press. Bons, E., R. Brucker, and J. Joosten, eds. (2014), The Reception of Septuagint Words in JewishHellenistic and Christian Literature, WUNT 2/367, Tubingen: Mohr Siebeck. Bons, E., and J. Joosten, eds. (2016), Die Sprache der Septuaginta/The Language of the Septuagint, LXX.H 3, Gütersloh: Gütersloher Verlagshaus. Bons, E., J. Joosten, and R. Hunziker-Rodewald, eds. (2015), Biblical Lexicology, Hebrew and Greek: Semantics, Exegesis, Translation, BZAW 443, Berlin: De Gruyter. Bons, E., J. Joosten, S. Kessler, and P. Le Moigne (2002), La Bible d’Alexandrie: Les Douze Prophetes: Osée, BdA. 23.1, Paris: Cerf. Bons, E. and T. J. Kraus, eds. (2011), Et sapienter et eloquenter: Studies on Rhetorical and Stylistic Features of the Septuagint, FRLANT 241, Gottingen: Vandenhoeck & Ruprecht. Bons, E., P. Pouchelle, and D. Scialabba, eds. (2019), The Vocabulary of the Septuagint and its Hellenistic Background, WUNT 2/496, Tübingen: Mohr Siebeck. Borée, W. (1930), Die alten Ortsnamen Palästinas, Hildesheim: Georg Olms. Botterweck, G. J., H. Ringgren, and H.-J. Fabry (1970–2019), Theological Dictionary of the Old Testament, 16 volumes, Grand Rapids: Eerdmans. Le Boulluec, A. and P. Sandevoir (1989), La Bible d’Alexandrie. II. L’Exode, BdA 2, Paris: Cerf.

REFERENCES 417

Boyd-Taylor, C. (1998), “A Place in the Sun: The Interpretative Significance of LXX-Psalm 18:5c,” BIOSCS 31: 71–105. Boyd-Taylor, C. (2001), “The Evidentiary Value of Septuagint Usage for Greek Lexicography: Alice’s Reply to Humpty Dumpty,” BIOSCS 34: 47–80. Boyd-Taylor, C. (2004), “Lexicography and Interlanguage: Gaining Our Bearings,” BIOSCS 37: 55–72. Boyd-Taylor, C. (2006a), “In a Mirror, Dimly: Reading the Septuagint as a Document of Its Times,” in W. Kraus and R. Glenn Wooden (eds.), Septuagint Research: Issues and Challenges in the Study of the Greek Jewish Scriptures, 15–31, SBLSCS 53, Atlanta: SBL Press. Boyd-Taylor, C. (2006b), “Toward the Analysis of Translational Norms: A Sighting Shot,” BIOSCS 39: 27–46. Boyd-Taylor, C. (2008a), “An Ear for an Eye: Lay Literacy and the Septuagint,” in A. Voitila and J. Jokiranta (eds.), Scripture in Transition: Essays on Septuagint, Hebrew Bible, and Dead Sea Scrolls in Honour of Raija Sollamo, 127–46, JSJSup 126, Leiden: Brill. Boyd-Taylor, C. (2008b), “The Greek Bible among Jews and Christians in the Middle Ages: The Evidence of Codex Ambrosianus,” in M. K. H. Peters (ed.), XIII Congress of the International Organization for Septuagint and Cognate Studies, Ljubljana, 2007, 29–39, SBLSCS 55, Atlanta: SBL Press. Boyd-Taylor, C. (2010a), “Echoes of the Septuagint in Byzantine Judaism,” in M. Karrer and W. Kraus (eds.), Die Septuaginta—Texte, Theologien, Einflüsse 2. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 23.-27. Juli 2008, 272–88, WUNT 252, Tübingen: Mohr Siebeck. Boyd-Taylor, C. (2010b), “The Semantics of Biblical Language Redux,” in R. J. V. Hiebert (ed.), “Translation Is Required”: The Septuagint in Retrospect and Prospect, 41–57, SBLSCS 56, Atlanta: SBL Press. Boyd-Taylor, C. (2011), Reading between the Lines: The Interlinear Paradigm for Septuagint Studies, BiTS 8, Leuven: Peeters. Boyd-Taylor, C., ed. (2013), A Question of Methodology: Albert Pietersma, Collected Essays on the Septuagint, Biblical Tools and Studies, BiTS 14, Leuven: Peeters. Boyd-Taylor, C. (2014), “Afterlives of the Septuagint: A Christian Witness to the Greek Bible in Byzantine Judaism,” in J. K. Aitken and J. C. Paget (eds.), The Jewish-Greek Tradition in Antiquity and the Byzantine Empire, 135–151, Cambridge: Cambridge University Press. Boyd-Taylor, C. (2015), “The τὸ ἰουδαϊκόν Glosses: Lost Jewish Version or Red-herring?” in J. K. Aitken and T. V. Evans (eds.), Biblical Greek in Context: Essays in Honour of John A. L. Lee, 41–57, BiTS 22, Leuven: Peeters. Boyd-Taylor, C. (2018), “Faithful Scribes and Phantom Texts: Jewish Transmission of the Septuagint Prior to the Amoraic Period,” in M. Meiser, M. Geiger, S. Kreuzer, and M. Sigismund (eds.), Die Septuaginta—Geschichte, Wirkung, Relevanz. 6. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 21.-24. Juli 2016, 359–382, WUNT 405, Tübingen: Mohr Siebeck Bowman, S. (1985), The Jews of Byzantium 1204–1453, Tuscaloosa: University of Alabama Press. Bowman, S. (2014), “The Jewish Experience in Byzantium,” in J. K. Aitken and J. Carleton Paget (eds.), The Jewish-Greek Tradition in Antiquity and the Byzantine Empire, 37–53, Cambridge: Cambridge University Press.

418 REFERENCES

Bratsiotis, N. (1966), “Nepheš-Psychè, ein Beitrag zur Erforschung der Sprache und Theologie der Septuaginta,” in E. J. Brill (ed.), Volume du congrès: Genève 1965, 58–89, VTSup 15, Leiden: Brill. Brayford, S. A. (2007), Genesis, SEPT, Leiden: Brill. Brenton, L. C. (1844), The Septuagint Version of the Old Testament, According to the Vatican Text, Translated into English: With the Principal Various Readings of the Alexandrine Copy, and a Table of Comparative Chronology, 2 volumes, London: Samuel Bagster & Sons. Brock, S. P. (1972), “The Phenomenon of the Septuagint,” OtSt 17: 11–36. Brock, S. P. (1992), “Versions, Ancient (Syriac),” in D. N. Freedman (ed.), The Anchor Bible Dictionary, Volume 6: 794–9, New York: Doubleday. Brock, S. P. (1996), The Recensions of the Septuaginta Version of 1 Samuel, Turin: Silvio Zamorani. Brock, S. P. and J. A. L. Lee (1972), “Memorandum on the Proposed LXX Lexicon Project,” in R. Kraft (ed.), Septuagintal Lexicography, 20–4, SBLSCS 1, Missoula: Scholars Press. Brock, S. P., C. T. Fritsch, and S. Jellicoe, eds. (1973), A Classified Bibliography of the Septuagint, ALGHJ 6, Leiden: Brill. Brodersen, A. (2017), The End of the Psalter, BZAW 505, Berlin: De Gruyter. Brønno, E. (1940), “Some Nominal Types in the Septuagint: Contributions to Pre-Masoretic Hebrew Grammar,” CM 3: 180–213. Brønno, E. (1943), Studien über hebraïsche Morphologie und Vokalismus: auf Grundlage der merchatischen Fragmente der zweiten Kolumne der Hexapla des Origenes, AKM 28, Leipzig: Deutsche Morgenländische Gesellschaft. Brooke, A. E. and N. McLean, eds. (1906), The Old Testament in Greek according to the Septuagint, Volume I, Part I. Genesis, Cambridge: Cambridge University Press. Brooke, A. E. and N. McLean (1927), The Old Testament in Greek according to the Septuagint, Volume II, Part I. I and II Samuel, Cambridge: Cambridge University Press. Brooke, A. E. and N. McLean (1906–40), The Old Testament in Greek according to the Text of Codex Vaticanus, Supplemented from Other Uncial Manuscripts, with a Critical Apparatus Containing the Variants of the Chief Ancient Authorities for the Text of the Septuagint, with H. St. J. Thackeray, 3 volumes, Cambridge: Cambridge University Press. Brooke, G. J. (1985), Exegesis at Qumran: 4QFlorilegium in Its Jewish Context, JSOTSup 29, Sheffield: JSOT Press. Brooke, G. J. (1998), “Shared Intertextual Interpretations in the Dead Sea Scrolls and the New Testament,” in M. Stone and E. G. Chazon (eds.), Biblical Perspectives: Early Use and Interpretation of the Bible in Light of the Dead Sea Scrolls. Proceedings of the First International Symposium of the Orion Center for the Study of the Dead Sea Scrolls and Associated Literature, 12–14 May, 1996, 35–57, STDJ 28, Leiden: Brill. Brooke, G. J. (2019), “Scrolls and Early Judaism,” in G. J. Brooke and C. Hempel (eds.), T&T Clark Companion to the Dead Sea Scrolls, 119–27, London: T&T Clark. Brooke, G. J. and B. Lindars, eds. (1992), Septuagint, Scrolls and Cognate Writings, SBLSCS 33, Atlanta: Scholars Press. Brown, C. (1975–1986), New International Dictionary of New Testament Theology, Grand Rapids: Zondervan. Brunello, A. (1960), La Bibbia secondo la versione dei Settanta: Prima ed unica traduzione in lingua moderna con introduzioni, comment e note, 2 volumes, Rome: Istituto Diffusione Edizioni Culturali. Bruni, A. M. (2016), “Old Church Slavonic Translations,” in A. Lange and E. Tov (eds.), Textual History of the Bible, Volume 1A: 393–407, Leiden: Brill.

REFERENCES 419

Büchner, D., ed. (2017), The SBL Commentary on the Septuagint: An Introduction, SBLSCS 67, Atlanta: SBL Press. Bührer, W. (2014), Am Anfang: Untersuchungen zur Textgenese und zur relativ-chronologischen Einordnung von Gen 1–3, FRLANT 256, Göttingen: Vandenhoeck & Ruprecht. Bugatus, C. (1788), Daniel secundum editionem LXX. Interpretum, Milan: Biblioteca Ambrosiana. Burkitt, F. C. (1897), Fragments of the Books of Kings according to the Translation of Aquila, Cambridge: Cambridge University Press. Burris, K. R. (2009), “A Critical Edition of the Hexaplaric Fragments of Numbers 1–18,” Ph.D. dissertation, Southern Baptist Theological Seminary, Louisville. Butera, C. J. and D. M. Moffit (2011), “P.Duk.inv. 727: A Dispute with ‘Proselytes’ in Egypt,” ZPE 177: 201–6. Butera, C. J. and D. M. Moffit (2013), “P.Duk. inv. 727r: New Evidence for the Meaning and Provenance of the Word Προσήλυτος,” JBL 132: 159–78. Buth, R. (1995), “Functional Grammar, Hebrew and Aramaic: An Integrated, Textlinguistic Approach to Syntax,” in W. R. Bodine (ed.), Discourse Analysis of Biblical Literature: What It Is and What It Offers, 77–102, SBLSS, Atlanta: Scholars Press. Byun, S. L. (2017), The Influence of Post-Biblical Hebrew and Aramaic on the Translator of Septuagint Isaiah, LHBOTS 635, London: Bloomsbury T&T Clark. Cacciari, A., and S. Tampellini (1998), “La Bibbia dei Settanta: A New Italian Translation of the Septuagint,” BIOSCS 31: 36–8. Cadwallader, A. H. (1992), “The Correction of the Text of Hebrews Towards the LXX,” NovT 34: 257–92. Cain, A. (2009), The Letters of Jerome: Asceticism, Biblical Exegesis, and the Construction of Christian Authority in Late Antiquity, Oxford: Oxford University Press. Caird, G. B. (1968), “Towards a Lexicon of the Septuagint. I,” JTS 19: 453–75. Caird, G. B. (1969), “Towards a Lexicon of the Septuagint. II,” JTS 20: 21–40. Caloz, M. (1978), Étude sur la LXX origénienne du psautier: les relations entre les leçons des psaumes du manuscrit Coislin 44, les fragments des Hexaples et le texte du psautier Gallican, OBO 19, Göttingen/Fribourg: Vandenhoeck & Ruprecht/Éditions universitaires. Cameron, E., ed. (2016), The New Cambridge History of the Bible, Volume 3: From 1450 to 1750, Cambridge: Cambridge University Press. Campbell, L. (2013), Historical Linguistics: An Introduction, 3rd edn., Edinburgh: Edinburgh University Press. Campbell, L. and M. J. Mixco (2007), A Glossary of Historical Linguistics, Edinburgh: Edinburgh University Press. Canart, P., P.-M. Bogaert, and S. Pisano (1999), Bibliothecae Apostolicae Vaticanae Codex Vaticanus Graecus 1209: Bibliorum Sacrorum Graecorum Codex Vaticanus B, Rome: Instituto Poligrafico e Zecca Della Stato. Cappelletti, S. (2009), “Biblical Quotations in the Greek Jewish Inscriptions,” in N. de Lange, J. Krivoruchko, and C. Boyd-Taylor (eds.), Jewish Reception of Greek Bible Versions, 128–41, TSMEMJ 23, Tübingen: Mohr Siebeck. Carbonaro, P. (2012), La Lettre d’Aristée et le Mythe des Ages du Monde, CahRB 79, Pendé: Garabalda. Carey, H. J. (2009), Jesus’ Cry from the Cross. Towards a First-Century Understanding of the Intertextual Relationship between Psalm 22 and the Narrative of Mark’s Gospel, LNTS 398, London: T&T Clark.

420 REFERENCES

Carleton Paget, J. (2014), “The Origins of the Septuagint,” in J. K. Aitken and J. C. Paget (eds.), The Jewish-Greek Tradition in Antiquity and the Byzantine Empire, 105–19, Cambridge: Cambridge University Press. Carson, D. A. (1979), The King James Debate: A Plea for Realism, Grand Rapids: Baker. Cavalier, C. (2012), La Bible d’Alexandrie. XII. Esther, BdA 12, Paris: Cerf. Ceulemans, R. (2008), “The Greek Christian Afterlife of the Minor Versions: The Possibilities of a Shift in Perspective,” in H. Ausloos, B. Lemmelijn, and M. Vervenne (eds.), Florilegium Lovaniense: Studies in Septuagint and Textual Criticism in Honour of Florentino García Martínez, 101–17, BETL 224, Leuven: Peeters. Ceulemans, R. (2009), “A Critical Edition of the Hexaplaric Fragments of the Book of Canticles, with Emphasis on Their Reception in Greek Christian Exegesis,” Ph.D. dissertation, Katholieke Universiteit Leuven, Leuven. Ceulemans, R. (2012), “Greek Christian access to ‘the Three,’ 250–600 CE,” in T. M. Law and A. Salvesen (eds.), Greek Scripture and the Rabbis, 165–91, CBET 66, Leuven: Peeters. Ceulemans, R. (2016), “Theodoret and the Antiochene Text of the Psalms,” in W. Kraus, M. N. van der Meer, and M. Meiser (eds.), XV Congress of the International Organization for Septuagint and Cognate Studies, Munich, 2013, 149–64, SBLSCS 64, Atlanta: SBL Press. Chamberlain, G. A. (2011), The Greek of the Septuagint: A Supplemental Lexicon, Peabody: Hendrickson Publishers. Chantraine, P. (2009), Dictionnaire étymologique de la langue grecque. Histoire des mots, Nouvelle edn., Paris: Klincksieck. Chapman, S. B. (2000), The Law and the Prophets: A Study in Old Testament Canon Formation, FAT 27, Tübingen: Mohr Siebeck. Chaze, M. (1980), “Remarques et notes sur les versions grecque et ladino du Pentateuque de Constantinople, 1547,” in G. Nahon and C. Touati (eds.), Hommage à Georges Vajda, Études d’histoire et de pensée juives, 323–32, CREJ 1, Leuven: Peeters. Childs, B. S. (1974), The Book of Exodus, OTL, Philadelphia: Westminster Press. Cignelli, L., and R. Pierri (2003–), Sintassi di Greco biblico (LXX e NT), 2 volumes, Analecta 61/77, Jerusalem: Franciscan Press. Cimosa, M. (1995), Guida allo studio della Bibbia greca (LXX): Storia-Lingua-Testi, Rome: Società biblica britannica e forestiera. Clogg, R. (1969), “The Correspondence of Adhamantios Korais with the British and Foreign Bible Society (1808),” GOTR 14: 65–84. Clogg, R. (2004), “Enlightening ‘A Poor, Oppressed, and Darkened Nation’: Some Early Activities of the BFBS in the Levant,” in S. Batalden, K. Cann, and J. Dean (eds.), Sowing the Word: The Cultural Impact of the British and Foreign Bible Society 1804–2004, 234–50, MBW 3, Sheffield: Sheffield Phoenix. Codex Sinaiticus (2010), Codex Sinaiticus: Facsimile Edition with Reference Guide, Peabody: Hendrickson. Coenen, L., K. Hacker, E. Beyreuther, J. Kabiersch, S. Kreuzer, H. Lichtenberger, G. Mayer, and H. Seebass, eds. (2014), Theologisches Begriffslexikon zum Neuen Testament: Ausgabe mit aktualisierten Literaturangaben, Bodenborn: SCM Brockhaus. Collins, J. J. (1985), “Artapanus,” in J. H. Charlesworth (ed.), The Old Testament Pseudepigrapha, Volume 2: 889–903, Garden City: Doubleday. Collins, J. J. (2006), “Messianism and Exegetical Tradition: Evidence of the LXX Pentateuch,” in M. A. Knibb (ed.), The Septuagint and Messianism, 129–37, BETL 195, Leuven: Peeters. Collins, N. (2000), The Library in Alexandria & the Bible in Greek, VTSup 82, Leiden: Brill.

REFERENCES 421

Colson, F. H., ed. and trans. (1935), Philo: On Abraham. On Joseph. On Moses, LCL 289, Cambridge: Harvard University Press. Colwell, E. C. (1969), Studies in Methodology in Textual Criticism of the New Testament, NTTS 9, Leiden: Brill. Conybeare, F. C. (1898), The Dialogues of Athanasius and Zacchaeus and of Timothy and Aquila: Edited with Prolegomena and Facsimiles, Oxford: Oxford University Press. Conybeare, F. C., and St. G. Stock (1905), Grammar of Septuagint Greek, Boston: Ginn. Conybeare, F. C. and St. G. Stock ([1905] 1995), A Grammar of Septuagint Greek: With Selected Readings, Vocabularies, and Updated Indexes, Expanded edn., Grand Rapids: Hendrickson. Cook, J. (2001), “Intertextual Relationships between the Septuagint of Psalms and Proverbs,” in R. J. V. Hiebert, C. E. Cox, and P. J. Gentry (eds.), The Old Greek Psalter: Studies in Honour of Albert Pietersma, 218–28, JSOTSup 332, Sheffield: Sheffield Academic Press. Cook, J. (2005), “Theological/Ideological Tendenz in the Septuagint—LXX Proverbs: A Case Study,” in F. G. Martinez and M. Vervenne (eds.), Interpreting Translation: Studies on the LXX and Ezekiel in Honour of Johann Lust, 65–79, BETL 192, Leuven: Peeters. Cook, J. (2006a), “Intertextual Readings in the Septuagint,” in C. Breytenback, J. C. Thom, and J. Punt (eds.), The New Testament Interpreted: Essays in Honour of Bernard C. Lategan, 119–34, NovTSup 124, Leiden: Brill. Cook, J. (2006b), “The Translation of a Translation: Some Methodological Considerations on the Translation of the Septuagint,” in M. K. H. Peters (ed.), XII Congress of the International Organization for Septuagint and Cognate Studies, Leiden, 2004, 29–40, SBLSCS 54, Leiden: Brill. Cook, J. (2010a), “The Relationship between the Septuagint Versions of Isaiah and Proverbs,” in M. van der Meer, P. van Keulen, W. van Peursen, and B. Ter Haar Romeny (eds.), Isaiah in Context: Studies in Honour of Arie van der Kooij on the Occasion of his Sixty-Fifth Birthday, 199–214, VTSup 138, Leiden: Brill. Cook, J. (2010b), “Towards the Formulation of a Theology of the Septuagint,” in A. Lemaire (ed.), Congress Volume Ljubljana 2007, 621–37, VTSup 133, Leiden: Brill. Cox, C. E. (1985), “The Use of the Armenian Version for the Textual Criticism of the Septuagint,” in N. Fernández Marcos (ed.), La Septuaginta en la investigación contemporánea: (V Congreso de la IOSCS), 25–35, Madrid: CSIC. Cox, C. E. (1993), “The Translation of Aquila, Symmachus, and Theodotion Found in the Margins of Armenian Manuscripts,” in C. Burchard (ed.), Armenia and the Bible, Papers Presented to the International Symposium Held at Heidelberg, 1990, 35–45, UPATS 12, Leuven: Peeters. Cox, C. E. (2001), “Schaper’s Eschatology Meets Kraus’s Theology of the Psalms,” in R. J. V. Hiebert, C. E. Cox, and P. J. Gentry (eds.), The Old Greek Psalter: Studies in Honour of Albert Pietersma, 289–311, JSOTSup 332, Sheffield: Sheffield Academic. Cox, C. E. (2006), Armenian Job: Reconstructed Greek Text, Critical Edition of the Armenian with English Translation, HUAS 8, Leuven: Peeters. Cox, C. E. (2015), “Job,” in J. K. Aitken (ed.), The T&T Clark Companion to the Septuagint, 385–400, London: Bloomsbury T&T Clark. Cox, C. E. (2016), “Armenian Translations: Secondary Translations,” in A. Lange and E. Tov (eds.), Textual History of the Bible, Volume 1A: 370–75, Leiden: Brill. Cribiore, R. (2001), Gymnastics of the Mind: Greek Education in Hellenistic and Roman Egypt, Princeton: Princeton University Press.

422 REFERENCES

Crisp, S. (2016), “The Septuagint as Canon,” BT 67: 137–50. De Crom, D. (2009), “The Book of Canticles in Codex Graecus Venetus 7,” in N. de Lange, J. Krivoruchko, and C. Boyd-Taylor (eds.), Jewish Reception of Greek Bible Versions, 287–301, TSMEMJ 23, Tübingen: Mohr Siebeck. Cross, F. M. (1964), “The History of the Biblical Text in the Light of Discoveries in the Judaean Desert,” HTR 57: 281–99. Cross, F. M. (1995), The Ancient Library of Qumran, 3rd edn., Minneapolis: Fortress Press. Croy, N. C. (2006), 3 Maccabees, SEPT, Leiden: Brill. Cumming, S., T. Ono, and R. Laury (2011), “Discourse, Grammar and Interaction,” in T. A. V. Dijk (ed.), Discourse as Structure and Process, 8–36, 2nd edn., London: Sage Publications. Cuppi, L. (2012), “The Treatment of Personal Names in the Book of Proverbs from the Septuagint to the Masoretic Text,” in T. M. Law and A. Salvesen (eds.), Greek Scripture and the Rabbis, 19–38, CBET 66, Leuven: Peeters. Dafni, E. (2002), “Theologie der Sprache der Septuaginta,” TZ 58: 315–28. Dafni, E. (2010), “Konstantinos Oikonomos ex Oikonomon als Septuaginta-Interpret,” in A. Lemaire (ed.), Congress Volume Ljubljana 2007, 265–92, VTSup 133, Leiden: Brill. Davidson, T. (1874), The Grammar of Dionysios Thrax: Translated from the Greek, St. Louis: Studley. Davila, J. (2005), “(How) Can We Tell if a Greek Apocryphon or Pseudepigraphon Has Been Translated from Hebrew or Aramaic?” JSP 15: 3–61. Dean, J. E., ed. (1935), Epiphanius’ Treatise on Weights and Measures: The Syriac Version, Chicago: University of Chicago Press. Deissmann, A. (1901), Bible Studies: Contributions Chiefly from Papyri and Inscriptions to the History of the Language, the Literature, and the Religion of Hellenistic Judaism and Primitive Christianity, A. Grieve (trans.), Edinburgh: T&T Clark. Deissmann, A. (1903), “Die Hellenisierung des semitischen Montheismus,” NJahrb 11: 162–77. Deissmann, A. (1908), Licht vom Osten. Das Neue Testament und die neuentdeckten Texte der hellenistisch-römischen Welt, Tübingen: Mohr Siebeck. DelCogliano, M. (2008), “Basil of Caesarea on Proverbs 8:22 and the Sources of Pro-Nicene Theology,” JTS 59: 183–90. Delicostopoulos, A. (1998), “Major Greek Translations of the Bible,” in J. Krašovec (ed.), The Interpretation of the Bible: The International Symposium in Slovenia, 297–316, JSOTSup 289, Sheffield Academic. Delitzsch, F. (1886), Fortgesetzte Studien zur Entstehungsgeschichte der Complutensischen Polyglotte, Leipzig: Edelmann. Dempster, S. (2008), “Torah, Torah, Torah: The Emergence of the Tripartite Canon,” in C. A. Evans and E. Tov (eds.), Exploring the Origins of the Bible: Canon Formation in Historical, Literary, and Theological Perspective, 87–127, Grand Rapids: Baker. Derenbourg, J. (1867), Essai sur l’Histoire et la Géographie de la Palestine d’Après les Thalmuds et les Autres Sources Rabbiniques: I., Histoire de la Palestine, Paris: l’Imprimerie Impériale. Desnitsky, A. S. (2006), “The Vision of the Old Testament Canon in the Russian Orthodox Church,” in L. J. de Regt (ed.), Canon and Modern Bible Translation in Interconfessional Perspective, 91–105, Istanbul: UBS. deSilva, D. (2006), 4 Maccabees, SEPT, Leiden: Brill. Devreesse, R. (1937), Bibliothecae Apostolicae Vaticanae Codices Vaticani Graeci, Volume 2: Codices 330–603, Vatican City: Biblioteca Vaticana. Devreesse, R. (1954), Introduction à l’étude des manuscrits grecs, Paris: Klincksieck. Dhont, M. (2018), Style and Context of Old Greek Job, JSJSup 183, Leiden: Brill.

REFERENCES 423

Dhont, M. (2019), “Towards a Comprehensive Explanation for the Stylistic Diversity of the Septuagint Corpus,” VT 69: 388–407. Diakonoff, I. M. (1988), Afrasian Languages, Moscow: Nauka. Dickey, E. (2015), “Columnar Translation: An Ancient Interpretive Tool That the Romans Gave the Greeks,” CQ 65: 807–21. Dik, S. C. (1997), The Theory of Functional Grammar: Part 1: The Structure of the Clause, K. Hengeveld (trans.), 2nd edn., Functional Grammar Series 20, Berlin: De Gruyter. Dimitrova, B. E. (2010), “Translation Process,” in Y. Gambier and L. van Doorslaer (eds.), Handbook of Translation Studies: Volume 1, 406–11, Amsterdam: John Benjamins. Dines, J. M. (1995), “Imaging Creation: The Septuagint Translation of Genesis 1:2,” HeyJ 36: 439–50. Dines, J. M. (2004), The Septuagint, London: T&T Clark. Dines, J. M. (2007), “The Book of the Twelve: Translation, Interpretation, and Current Research,” Paper presented as a Grinfield Lecture on the Septuagint, Oxford, March 1. Dines, J. M. (2016), “Stylistic Features of the Septuagint,” in E. Bons and J. Joosten (eds.), Die Sprache der Septuaginta/The Language of the Septuagint, 375–85, LXX.H 3, Gütersloh: Gütersloher Verlagshaus. Docherty, S. E. (2009), The Use of the Old Testament in Hebrews: A Case Study in Early Jewish Bible Interpretation, WUNT 2/260, Tübingen: Mohr Siebeck. Dodd, C. H. (1935), The Bible and the Greeks, London: Hodder and Stoughton. Dogniez, C. (1995), Bibliography of the Septuagint / Bibliographie de La Septante, 1970–1993, VTSup 60, Leiden: Brill. Dogniez, C. (2005), “L’intertextualité dans la LXX de Zacharie 9–14,” in F. G. Martinez and M. Vervenne (eds.), Interpreting Translation: Studies on the LXX and Ezekiel in Honour of Johann Lust, 81–96, BETL 192, Leuven: Peeters. Dogniez, C. (2010), “L’indépendance du traducteur grec d’Isaïe par rapport au Dodekapropheton,” in M. van der Meer, P. van Keulen, W. van Peursen, and B. Ter Haar Romeny (eds.), Isaiah in Context: Studies in Honour of Arie van der Kooij on the Occasion of His Sixty-Fifth Birthday, 229–46, VTSup 138, Leiden: Brill. Dogniez, C. and M. Harl (1992), La Bible d’Alexandrie. V. Le Deutéronome, BdA 5, Paris: Cerf. Dogniez, C. and M. Harl (2001), La Bible des Septante. Le Pentateuque d’Alexandrie, Paris: Cerf. Donaldson, A. M. (2009), “Explicit References to New Testament Variant Readings among Greek and Latin Church Fathers,” Ph.D. dissertation, 2 volumes, Notre Dame: University of Notre Dame. Donner, H. (2013), Wilhelm Gesenius: Hebräisches und Aramäisches Handwörterbuch über das Alte Testament, 18th edn., Berlin: Springer. Dooley, R. A. and S. Levinsohn (2001), Analyzing Discourse: A Manual of Basic Concepts, Dallas: SIL International. Dorival, G. (1994), La Bible d’Alexandrie. IV. Les Nombres, BdA 4, Paris: Cerf. Dorival, G. (1995), “Les phénomènes d’intertextualité dans le livre grec des Nombres,” in G. Dorival and O. Munnich (eds.), Κατὰ τοὺς ο ́—Selon les Septante. Trente études sur la Bible grecque des Septante. En hommage à Marguerite Harl, 253–85, Paris: Cerf. Dorival, G. (2003), “L’apport des Peres de l’Église a la question de la clôture du canon de l’Ancien Testament,” in J.-M. Auwers and H. J. de Jonge (eds.), The Biblical Canons, 81–110, BETL 163, Leuven: Leuven University Press. Dorival, G. (2008), “‘La Bible d’Alexandrie’, Which Changes?” in H. Ausloos, J. Cook, F. G. Martínez, B. Lemmelijn, and M. Vervenne (eds.), Translating a Translation, 65–78, BETL 213, Leuven: Peeters.

424 REFERENCES

Dorival, G. (2010), “New Light about the Origins of the Septuagint,” in M. Karrer and W. Kraus (eds.), Die Septuaginta—Texte, Theologien, Einflüsse 2. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 23.-27. Juli 2008, 36–47, WUNT 252, Tübingen: Mohr Siebeck. Dorival, G. (2016a), “La lexicographie de la Septante,” in E. Bons and J. Joosten (eds.), Die Sprache der Septuaginta/The Language of the Septuagint, 271–305, LXX.H 3, Gütersloh: Gütersloher Verlagshaus. Dorival, G. (2016b), “Textes des écritures et exégèse patristique: des relations complexes,” in L. Mellerin (ed.), La reception du livre de Qohélet, 1er–XIIIe siècle, 17–34, Paris: Cerf. Douglas, A. (2012), “Limitations to Writing a Theology of the Septuagint,” JSCS 45: 104–17. Driver, S. R. (1890), Notes on the Hebrew Text of the Books of Samuel: With an Introduction on Hebrew Palaeography and the Ancient Versions, Oxford: Clarendon Press. Drusius, J. (1588), Parallela sacra, hoc est, locorum veteris testamenti cum iis, quae in nouo citantur, coniuncta commemoratio Ebraicè & Graecè, Franeker: Ægidius Radæus. Dunkelgrün, T. W. (2010–11), “The Hebrew Library of a Renaissance Humanist. Andreas Masius and the Bibliography to His Iosuae Imperatoris Historia (1574), with a Latin Edition and an Annotated English Translation,” Studia Rosenthaliana 42–3: 197–252. Dunkelgrün, T. W. (2012), “The Multiplicity of Scripture: The Confluence of Textual Traditions in the Making of the Antwerp Polyglot Bible (1568–1573),” Ph.D. dissertation, University of Chicago, Chicago. Dunkelgrün, T. W. (2016), “The Testimonium Flavium Canonicum: Josephus as a Witness to the Biblical Canon, 1566–1823,” International Journal of the Classical Tradition 23: 252–68. Edwards, T. (2012), “Aquila in the Psalter: A Prolegomenon,” in T. M. Law and A. Salvesen (eds.), Greek Scripture and the Rabbis, 87–106, CBET 66, Leuven: Peeters. Ehorn, S. (2020), 2 Maccabees 1–7, Baylor Handbook on the Septuagint, Waco: Baylor University Press. Eichler, E., G. Hilty H. Löffler, H. Steger, and L. Zgusta, eds. (1995), Namenforschung/Name Studies/Les noms propres: Ein internationales Handbuch zur Onomastik/An International Handbook of Onomastics/Manuel international d’onomastique, HSK 11.1, Berlin: De Gruyter. Eldon Jay Epp, “Textual Criticism (NT),” in D. N. Freedman (ed.), The Anchor Bible Dictionary, Volume 6, 412–35, New York: Doubleday. Ernest, J. D. (2004), The Bible in Athanasius of Alexandria, Leiden: Brill. Evans, C. A., and S. E. Porter (2017), “A Recent Discovery in Byzantine-Era Galilee and the Problem of ‘Regularized’ Spelling of Koine and Byzantine Greek,” JGRChJ 13: 172–88. Evans, T. (2001), Verbal Syntax in the Greek Pentateuch: Natural Greek Usage and Hebrew Interference, Oxford: Oxford University Press. Evans, T. (2010), “Standard Koine Greek in Third Century BC Papyri,” in T. Gagos and A. Hyatt (eds.), Proceedings of the 25th International Congress of Papyrology. Ann Arbor, July 29–August 4, 2007, 197–205, Ann Arbor: University of Michigan Library. Even-Zohar, I. (1975), “Decisions in Translating Poetry,” Ha-Sifrut/Literature 21: 32–45. [Hebrew] Even-Zohar, I. (1978), Papers in Historical Poetics, Tel Aviv: Porter Institute. Even-Zohar, I. (1990), Polysystem Studies, Durham: Duke University Press. [= Poetics Today 11.1]. Evseev, I. (1911), “Manuscript Tradition of the Slavonic Bible,” (in Russian): Евсеев И., “Рукописная традиция славянской Библии,” Христианское чтение 4: 435–50.

REFERENCES 425

Ewen, C. J. and H. V. D. Hulst (2001), The Phonological Structure of Words: An Introduction, Cambridge Textbooks in Linguistics, Cambridge: Cambridge University Press. Fabry, H.-J. (2001), “Die griechischen Handschriften vom Toten Meer,” in H.-J. Fabry and U. Offerhaus (eds.), Im Brennpunkt: Die Septuaginta. Studien zur Entstehung und Bedeutung der Griechischen Bibel, 131–53, BWANT 8, Stuttgart: Kohlhammer. Fassberg, S. E. (1990), A Grammar of the Palestinian Targum Fragments from the Cairo Geniza, HSS 38, Atlanta: Scholars Press. Faulkenberry Miller, J. B. (2007), “4QLXXLeva and Proto-Septuagint Studies: Reassessing Qumran Evidence for the Urtext Theory,” in M. T. Davis and B. Strawn (eds.), Qumran Studies: New Approaches, New Questions, 1–27, Grand Rapids: Eerdmans. Fernández Marcos, N., ed. (1985), La Septuaginta en la investigación contemporánea: (V Congreso de la IOSCS), TECC 34, Madrid: Instituto “Arias Montano” CSIC. Fernández Marcos, N. (1987), “The Use of the Septuagint in the Criticism of the Hebrew Bible,” Sef 47: 59–72. Fernández Marcos, N. (1994a), Scribes and Translators: Septuagint and Old Latin in the Books of Kings, VTSup 54, Leiden: Brill. Fernández Marcos, N. (1994b), “The Septuagint Reading of the Book of Job,” in W. A. M. Beuken (ed.), The Book of Job, 251–66, BETL 114, Leuven: Peeters. Fernández Marcos, N. (1998), Introducción a las versiones griegas de la Biblia, 2nd edn., TECC 64, Madrid: CSIC. Fernández Marcos, N. (2000a), La Bibbia dei Settanta. Introduzione alle versioni greche della Bibbia, Supplementi all’ Introduzione allo studio della Bibbia 6, Brescia: Morcelliana. Fernández Marcos, N. (2000b), The Septuagint in Context: Introduction to the Greek Versions of the Bible, Leiden/Boston: Brill. Translation of Fernández Marcos (1998). Fernández Marcos, N. (2001), “Reactions to the Panel on Modern Translations,” in B. A. Taylor (ed.), X Congress of the International Organization for Septuagint and Cognate Studies, Oslo, 1998, 233–40, SBLSCS 51, Atlanta: SBL Press. Fernández Marcos, N. (2005), “Un manuscrito complutense redivivo,” Sefarad 65: 65–83. Fernández Marcos, N. (2008), “A New Spanish Translation of the Septuagint,” in H. Ausloos, J. Cook, F. G. Martínez, B. Lemmelijn, and M. Vervenne (eds.), Translating a Translation, 283–91, BETL 213, Leuven: Peeters. Fernández Marcos, N. (2009a), “Non placet Septuaginta: Revisions and New Greek Versions of the Bible in Byzantium,” in N. de Lange, J. Krivoruchko, and C. Boyd-Taylor (eds.), Jewish Reception of Greek Bible Versions, 39–50, TSMEMJ 23, Tübingen: Mohr Siebeck. Fernández Marcos, N. (2009b), “The Greek Pentateuch and the Scholarly Milieu of Alexandria,” SemClass 2: 81–9. Fernández Marcos, N. (2011), Judges, BHQ, Stuttgart: Deutsche Bibelgesellschaft. Fernández Marcos, N. (2013), “The Antiochene Edition in the Text History of the Greek Bible,” in S. Kreuzer and M. Sigismund (eds.), Der Antiochenische Text der Septuaginta in seiner Bezeugung und seiner Bedeutung, 57–73, DSI 4, Göttingen: Vandenhoeck & Ruprecht. Fernández Marcos, N. (2014), Septuaginta: La Biblia griega de los judíos y cristianos, 2nd edn., Biblioteca de Estudios Bíblicos Minor 12, Salamanca: Ediciones Sígueme. Fernández Marcos, N. (2017), “The First Polyglot Bible,” in A. Piquer Otero and P. Torijano Morales (eds.), The Text of The Hebrew Bible and Its Editions: Studies in Celebration of the Fifth Centennial of the Complutensian Polyglot, 3–18, THBSup 1, Leiden: Brill. Fernández Marcos, N. and J. R. Busto Sáiz, eds. (1989), El texto antioqueno de la Biblia griega. Tomo I. Libros 1–2 Samuel, TECC 50, Madrid: CSIC. Fernández Marcos, N. and J. R. Busto Sáiz, eds. (1992), El texto antioqueno de la Biblia griega. Tomo II. 1-2 Reyes, TECC 53, Madrid: CSIC.

426 REFERENCES

Fernández Marcos, N. and J. R. Busto Sáiz, eds. (1996), El texto antioqueno de la Biblia griega. Tomo III. 1-2 Crónicas, TECC 60, Madrid: CSIC. Fernández Marcos, N. and J. R. Busto Sáiz (1989–96), El Texto Antioqueno de la biblia griega, TECC 50/53/60, Madrid: CSIC. Fernández Marcos, N. and M. V. Spottorno Díaz-Caro, eds. (2008), La Biblia griega. Septuaginta, Volume 1: Pentateuco, BEB 125, Salamanca: Ediciones Sígueme. Fernández Marcos, N., M. V. Spottorno Díaz-Caro, and J. M. Cañas Reíllo, eds. (2011), La Biblia griega. Septuaginta, Volume 2: Libros Históricos, BEB 126, Salamanca: Ediciones Sígueme. Fernández Marcos, N., M. V. Spottorno Díaz-Caro, and J. M. Cañas Reíllo, eds. (2013), La Biblia griega. Septuaginta, Volume 3: Libros poéticos y sapienciales, BEB 127, Salamanca: Ediciones Sígueme. Fernández Marcos, N., M. V. Spottorno Díaz-Caro, and J. M. Cañas Reíllo, eds. (2015), La Biblia griega. Septuaginta, Volume 4: Libros proféticos, BEB 128, Salamanca: Ediciones Sígueme. Fernández Marcos, N., M. V. Spottorno Díaz-Caro, and J. M. Cañas Reíllo, eds. (2016), La Biblia griega. Septuaginta, Volume 1: El pentateuco, 2nd edn., BEB 125, Salamanca: Ediciones Sígueme. Fernández Marcos, N., M. V. Spottorno Díaz Caro, and J. M. Cañas Reíllo, eds. (2005), Índice griego-hebreo del texto antioqueno en los libros históricos, 2 volumes, TECC 75, Madrid: CSIC. Field, F., ed. (1875), Origenis Hexaplorum quae supersunt: Veterum interpretum graecorum in totum Vetus Testamentum fragmenta, 2 volumes, Oxford: Oxford University Press. Filaret (Drozdov) (1858), “On Dogmatic Value and Conservative Usage of the Greek Septuagint and Slavonic translations of the Holy Scripture,” (in Russian): [Митрополит Филарет (Дроздов), “О догматическом достоинстве и охранительном употреблении греческого семидесяти толковников и славенского переводов Священного Писания],” Прибавления к творениям Святых Отцов в русском переводе 32: 452–84. Filson, F. V. (1972), “Translate the Septuagint,” in E. H. Barth and R. E. Cocroft (eds.), Festschrift to Honor F. Wilbur Gingrich: Lexicographer, Scholar, Teacher and Committed Christian Layman, 142–6, Leiden: Brill. Fincati, M. (2016), The Medieval Revision of the Ambrosian Hexateuch: Critical Editing between Septuaginta and Hebraic Veritas in MS Ambrosianus A 147 inf., DSI 5, Göttingen: Vandenhoeck & Ruprecht. Finsterbusch, K., and N. Jacoby (2016), MT-Jeremia und LXX-Jeremia 1–24: Synoptische Übersetzung und Analyse der Kommunikationsstruktur, WMANT 145, Göttingen: Vandenhoeck & Ruprecht. Fischer, G. (2017), “Jeremiah: Septuagint,” in A. Lange and E. Tov (eds.), Textual History of the Bible, Volume 1B: 543–55, Leiden: Brill. Fishbane, M. (1985), Biblical Interpretation in Ancient Israel, Oxford: Clarendon Press. Fitzmyer, J. A. (1995), “Tobit,” in M. Broshi, E. Eshel, J. Fitzmyer, E. Larson, C. Newsom, L. Schiffman, M. Smith, M. Stone, J. Strugnell, and A. Yardeni (eds.), Qumran Cave 4 XIV: Parabiblical Texts, Part 2, 1–76, DJD 19, Oxford: Clarendon Press. Fitzmyer, J. A. (2000), “The Significance of the Qumran Tobit Texts for the Study of Tobit,” in J. A. Fitzmyer (ed.), The Dead Sea Scrolls and Christian Origins, 131–58, Studies in the Dead Sea Scrolls and Related Literature, Grand Rapids: Eerdmans. Flashar, M. (1912), “Exegetische Studien zum Septuagintapsalter,” ZAW 32: 81–116, 161–89, 241–68.

REFERENCES 427

Fox, J. (2003), Semitic Noun Patterns, HSS 52, Winona Lake: Eisenbrauns. Fox, M. V. (2015), Proverbs: An Eclectic Edition with Introduction and Textual Commentary, HBCE 1, Atlanta: SBL Press. François, W. (2009), “Andreas Masius (1514–73): Humanist, Exegete and Syriac Scholar,” Journal of Eastern Christian Studies 61: 199–244. Frankel, Z. (1841), Vorstudien zu der Septuaginta, Leipzig: Vogel. Frankel, Z. (1851), Über der Einfluss der palästinischen Exegese auf die alexandrinische Hermeneutik, Leipzig: J. A. Barth. Fraser, P. M. (1972), Ptolemiac Alexandria, 3 volumes, Oxford: Clarendon Press. Freed, E. D. (1965), Old Testament Quotations in the Gospel of John, NovTSup 11, Leiden: Brill. Fresch, C. J. (2015), “Discourse Markers in the Septuagint and Early Koine Greek with Special Reference to the Twelve,” Ph.D. dissertation, University of Cambridge, Cambridge. Fresch, C. J. (2016), “The Peculiar Occurrences of οὖν in Septuagint Genesis and Exodus,” in W. Kraus, M. N. van der Meer, and M. Meiser (eds.), XV Congress of the International Organization for Septuagint and Cognate Studies, Munich, 2013, 455–72, SBLSCS 64, Atlanta: SBL Press. Freudenthal, J. (1890), “Are there Traces of Greek Philosophy in the Septuagint?” JQR 2: 205–22. Fritsch, C. T. (1943), Anti-Anthropomorphism in the Greek Pentateuch, Princeton: Princeton University Press. Fritsch, C. T. (1970a), “Minutes of IOSCS Meeting Wednesday, November 19, 1969, Royal York Hotel, Toronto, Canada,” BIOSCS 3: 3. Fritsch, C. T. (1970b), “The Future of Septuagint Studies: A Brief Survey,” BIOSCS 3:4–8. Gallagher, E. L. (2012), Hebrew Scripture in Patristic Biblical Theory: Canon, Language, Text, VCSup 114, Leiden: Brill. Gallagher, E. L. (2013a), “The Religious Provenance of the Aquila Manuscripts from the Cairo Genizah,” JJS 64: 283–305. Gallagher, E. L. (2013b), “The Septuagint’s Fidelty to Its Vorlage in Greek Patristic Thought,” in M. K. H. Peters (ed.), XIV Congress of the International Organization for Septuagint and Cognate Studies, Helsinki, 2010, 663–76, SBLSCS 59, Atlanta: SBL Press. Gallagher, E. L. (2014), “Writings Labeled ‘Apocrypha’ in Latin Patristic Sources,” in J. H. Charlesworth and L. M. McDonald (eds.), Sacra Scriptura: How “Non-Canonical” Texts Functioned in Early Judaism and Early Christianity, 1–14, Jewish and Christian Texts, London: Bloomsbury. Gallagher, E. L. (2015), “Why Did Jerome Translate Tobit and Judith?” HTR 108: 356–75. Gallagher, E. L. (2016), “Augustine on the Hebrew Bible,” JTS 67: 97–114. Gallagher, E. L. (2020), “The Hebraica Veritas,” in T. Denecker (ed.), Language and Culture in Early Christianity: A Companion, n.p., Leuven: Peeters. Gallagher, E. L. and J. D. Meade (2017), The Biblical Canon Lists from Early Christianity: Texts and Analysis, Oxford: Oxford University Press. Garbini, G. (1960), Il Semitico di Nord-Ovest [= Quaderni della Sezione Linguistica degli Annali 1], Napoli: Istituto Universitario Orientale di Napoli. García Martínez, F. (2005), “La Bible d’Alexandrie au miroir de Qumrân,” RevQ 22/2: 253–68. Gebhardt, O., ed. (1875), Graecus Venetus: Pentateuchi Proverbiorum Ruth Cantici Ecclesiastae Threnorum Danielis versio Graeca. Leipzig: Brockhaus. Gehman, H. S. (1951), “The Hebraic Character of Septuagint Greek,” VT 1: 81–90.

428 REFERENCES

Gehman, H. S. (1974), “Peregrinations in Septuagint Lexicography,” in H. N. Bream, R. D. Heim, and C. A. Moore (eds.), A Light unto My Path. Old Testament Studies in Honor of Jacob M. Myers, 223–40, Philadelphia: Temple University Press. Geiger, A. (1857), Urschrift und Übersetzungen der Bibel in ihrer Abhängigkeit von der inneren Entwickelung des Judentums, Breslau: Hainauer. Gelston, A. (2002), “Some Hebrew Misreadings in the Septuagint of Amos,” VT 52: 493–500. Gelston, A. (2006), “Some Difficulties Encountered by Ancient Translators,” in Y. A. P. Goldman, A. van der Kooij, and R. D. Weis (eds.), Sôfer Mahîr: Essays in Honour of Adrian Schenker Offered by the Editors of Biblia Hebraica Quinta, 47–58, VTSup 110, Leiden: Brill. Gentry, P. J. (2006), “The Septuagint and the Text of the Old Testament,” BBR 16: 193–218. Gentry, P. J. (2015), “The Greek Genizah Fragment of Ecclesiastes and Its Relation to Earlier Greek Versions,” in J. K. Aitken and T. V. Evans (eds.), Biblical Greek in Context: Essays in Honour of John A. L. Lee, 91–116, BiTS 22, Leuven: Peeters. Gentry, P. J. (2016a), “Did Origen Use the Artistarchian Signs in the Hexapla?” in W. Kraus, M. N. van der Meer, and M. Meiser (eds.), XV Congress of the International Organization for Septuagint and Cognate Studies, Munich, 2013, 133–148, SBLSCS 64, Atlanta: SBL Press. Gentry, P. J. (2016b), “Pre-Hexaplaric Translations, Hexapla, Post-Hexaplaric Translations,” in A. Lange and E. Tov (eds.), Textual History of the Bible, Volume 1A: 211–35, Leiden: Brill. Gentry, P. J., ed. (2019), Septuaginta. Vetus Testamentum Graecum. XI.2 Ecclesiastes, Göttingen: Vandenhoeck & Ruprecht. Gerleman, G. (1946), Studies in the Septuagint I: Book of Job, Lunds universitets årsskrift 43/2, Lund: C. W. K. Gleerup. Gerleman, G. (1956), Studies in the Septuagint III. Proverbs, Lund: C. W. K. Gleerup. Giannakis, G. K., ed. (2014), Encyclopedia of Ancient Greek Language and Linguistics, 3 volumes, Leiden: Brill. Gignac, F. T., S. J. (1976), A Grammar of the Greek Papyri of the Roman and Byzantine Periods, Volume 1: Phonology, Testi e Documenti per lo Studio dell’antichità 55, Milano: Istituto Editoriale Cisalpino—La Goliardica. Gignac, F. T., S. J. (1989), “The Development of Greek Phonology: The Fifteenth Century B.C. to the Twentieth Century After Christ,” in R. F. Sutton, Jr. (ed.), Daidalikon: Studies in Memory of Raymond V. Schoder, S. J., 131–7, Wauconda: Bolchazy-Carducci. Gil, L. (2016), “La columna griega de la Biblia Poliglota Complutense,” in A. Alvar Ezquerra (ed.), La Biblia Poliglota Complutense en su contexto, 261–83, Alcalá: Universidad de Alcalá Servicio de Publicaciones. Glenny, W. E. (2007), “Hebrew Misreadings or Free Translation in the Septuagint of Amos?” VT 57: 524–47. Glenny, W. E. (2009), Finding Meaning in the Text: Translation Technique and Theology in the Septuagint of Amos, VTSup 126, Leiden: Brill. Glenny, W. E. (2012), “The Septuagint and Apostolic Hermeneutics: Amos 9 in Acts 15,” BBR 22: 1–26. Glenny, W. E. (2013a), Amos, SEPT, Leiden: Brill. Glenny, W. E. (2013b), Hosea, SEPT, Leiden: Brill. Glenny, W. E. (2015), Micah, SEPT, Leiden: Brill. Glenny, W. E. (2016), “The Septuagint and Biblical Theology,” Them 41: 263–78. Glenny, W. E. (2018), “Messianism in Septuagint Amos?” in C. J. P. Friesen (ed.), Envisioning God in the Humanities: Essays on Christianity, Judaism, and Ancient Religion in Honor of Melissa Harl Sellew, 175–88, Eugene: Cascade Books.

REFERENCES 429

Glenny, W. E. (forthcoming), “The Intention of the Translator and Theology in the Septuagint,” in G. Kotzé, W. Kraus, and M. N. van der Meer (eds.), XVII Congress of the International Organization for Septuagint and Cognate Studies: Aberdeen, 2019, n.p., SBLSCS, Atlanta: SBL Press. Görg, M. (2001), “Die Septuaginta im Kontext spätägyptischer Kultur-Beispiele lokaler Inspiration bei der Übersetzungsarbeit am Pentateuch,” in H.-J. Fabry and U. Offerhaus (eds.), Im Brennpunkt: Die Septuaginta—Studien zur Entstehung und Bedeutung der Griechischen Bibele, 115–30, BWANT 153, Stuttgart: Kohlhammer. Goshen-Gottstein, M. H. (1963), “Theory and Practice of Textual Criticism: The Text-Critical Use of the Septuagint,” Text 3: 130–58. Grabe, J. E. (1709–20), Septuaginta Interpretum, 4 volumes, Zurich: Heidegger. Grabbe, L. L. (1988), Etymology in Early Jewish Interpretation: The Hebrew Names in Philo, Atlanta: Scholars Press. Grafton, A. (1983–93), Joseph Scaliger. A Study in the History of Classical Scholarship, 2 volumes, Oxford: Clarendon Press. Grafton, A. (2016), “‘Pandects of the Jews’: A French, Swiss and Italian Prelude to John Selden,” in S. Mandelbrote and J. Weinberg (eds.), Jewish Books and Their Readers: Aspects of the Intellectual Life of Christians and Jews in Early Modern Europe, 169–88, BSCHRC 75, Leiden: Brill. Grafton, A. and J. Weinberg (2011), “I have always loved the Holy Tongue.” Isaac Casaubon, the Jews, and a Forgotten Chapter in Renaissance Scholarship, Cambridge, MA: Belknap Press of Harvard University Press. Grafton, A. and M. Williams (2006), Christianity and the Transformation of the Book: Origen, Eusebius, and the Library of Caesarea, Cambridge: Belknap Press. Graves, M. (2007), Jerome’s Hebrew Philology: A Study Based on His Commentary on Jeremiah, VCSup 90, Leiden: Brill. Graves, M. (2014), The Inspiration and Interpretation of Scripture: What the Early Church Can Teach Us, Grand Rapids: Eerdmans. Gray, L. H. (1936), “Observations on the Phonology of the ΒƎΓΑΔΚƎΦΑΘ,” AJSL 52: 171–7. Greenspoon, L. (1997), “‘It’s All Greek to Me’: Septuagint Studies since 1968,” CurBS 5: 147–74. Greenspoon, L. (1998), “The Dead Sea Scrolls and the Greek Bible,” in P. W. Flint and J. C. VanderKam (eds.), The Dead Sea Scrolls after Fifty Years: A Comprehensive Assessment, Volume 1: 101–27, Leiden: Brill. Grinfield, E. W. (1850), An Apology for the Septuagint, London: W. Pickering. Grishchenko, A. (2018), “The Linguistic and Literary Contact between East Slavs and Jews in the Middle Ages: Results and Perspectives of the Study,” (in Russian) [Грищенко А. И., “Языковые и литературные контакты восточных славян и евреев в средние века: итоги и перспективы изучения,”] Studi Slavistici 15: 29–60. Guillaume, P. (2004), “New Light on the Nebiim from Alexandria: A Chronography to Replace the Deuteronomistic History,” JHS 5: 2–51. Gundry, R. H. (1967), The Use of the Old Testament in St. Matthew’s Gospel with Special Reference to the Messianic Hope, NovTSup 18, Leiden: Brill. Gundry, R. H. (1994), Matthew. A Commentary on His Handbook for a Mixed Church under Persecution, 2nd edn., Grand Rapids: Eerdmans. Gurtner, D. M. (2013), Exodus, SEPT, Leiden: Brill. Gzella, H. (2019), “Languages: Hebrew, Aramaic, Greek,” in G. J. Brooke and C. Hempel (eds.), T&T Clark Companion to the Dead Sea Scrolls, 192–203, London: Bloomsbury T&T Clark.

430 REFERENCES

Haar Romeny, R. B. ter (1997), A Syrian in Greek Dress: The Use of Greek, Hebrew, and Syriac Biblical Texts in Eusebius of Emesa’s Commentary on Genesis, Traditio Exegetica Graeca 6, Leuven: Peeters. Hacham, N. (2007), “3 Maccabees and Esther: Parallels, Intertextuality, and Diaspora Identity,” JBL 126: 765–85. Hadas, M., ed. and trans. ([1951] 2007), Aristeas to Philocrates (Letter of Aristeas), Eugene: Wipf & Stock. Hallermayer, M. and T. Elgvin (2006), “Schøyen Ms. 5234: Ein Neues ‘Tobit’-Fragment vom Toten Meer,” RevQ 22: 451–61. Hamilton, A. (1999), The Apocryphal Apocalypse, Oxford: Clarendon Press. d’Hamonville, D. M. (2000), La Bible d’Alexandrie. XVII. Les Proverbes, BdA 18, Paris: Cerf. Hanhart, R. (1967), “Die Bedeutung der Septuaginta-Forschung für die Theologie,” in R. Hanhart (ed.), Drei Studien zum Judentum, 38–64, Theologische Existenz heute 140, Munich: Kaiser. Hanhart, R., ed. (1979), Septuaginta. Vetus Testamentum Graecum. VIII.4. Iudith, Göttingen: Vandenhoeck & Ruprecht. Hanhart, R., ed. (1980), Septuaginta. Vetus Testamentum Graecum. IX.3. Maccabaeorum Liber III, 2nd edn., Göttingen: Vandenhoeck & Ruprecht. Hanhart, R., ed. (1983a), Septuaginta. Vetus Testamentum Graecum. VIII.3. Esther, 2nd edn., Göttingen: Vandenhoeck & Ruprecht. Hanhart, R., ed. (1983b), Septuaginta. Vetus Testamentum Graecum. VIII.5. Tobit, Göttingen: Vandenhoeck & Ruprecht. Hanhart, R., ed. (1991), Septuaginta. Vetus Testamentum Graecum. VIII.1. Esdrae Liber I, 2nd edn., Göttingen: Vandenhoeck & Ruprecht. Hanhart, R. (1992), “The Translation of the Septuagint in Light of Earlier Tradition and Subsequent Influences,” in G. J. Brooke and B. Lindars (eds.), Septuagint, Scrolls and Cognate Writings: Papers Presented to the International Symposium on the Septuagint and Its Relations to the Dead Sea Scrolls and Other Writings (Manchester, 1990), 339–379, SBLSCS 33, Atlanta: Scholars Press. Hanhart, R., ed. (1993), Septuaginta. Vetus Testamentum Graecum. VIII.2. Esdrae Liber II, Göttingen: Vandenhoeck & Ruprecht. Hanhart, R. (2002), “Introduction: Problems in the History of the LXX Text from Its Beginnings to Origen,” in M. Hengel (ed.), The Septuagint as Christian Scripture: Its Prehistory and the Problem of Its Canon, 1–17, Grand Rapids: Baker. Hanhart, R., ed. (2008), Septuaginta. Vetus Testamentum Graecum. IX.2. Maccabaeorum Liber II, 3rd edn., Göttingen: Vandenhoeck & Ruprecht. Hanhart, R., ed. (2014), Septuaginta. Vetus Testamentum Graecum. VII.2. Paralipomenon Liber II, Göttingen: Vandenhoeck & Ruprecht. Hanhart, R., ed. (2017a), Septuaginta. Vetus Testamentum Graecum. VIII.2. Esdrae Liber II, 2nd edn., Göttingen: Vandenhoeck & Ruprecht. Hanhart, R., ed. (2017b), Septuaginta. Vetus Testamentum Graecum. IX.2. Maccabaeorum Liber II, 4th edn., Göttingen: Vandenhoeck & Ruprecht. Hardy, N. (2015), “The Septuagint and the Transformation of Biblical Scholarship in England, from the King James Bible (1611) to the London Polyglot (1657),” in K. Killeen, H. Smith, and R. Willie (eds.), The Oxford Handbook of the Bible in Early Modern England, c. 1530–1700, 117–30, Oxford: Oxford University Press. Hardy, N. (2017), Criticism and Confession. The Bible in the Seventeenth-Century Republic of Letters, Oxford: Oxford University Press.

REFERENCES 431

Hardy, N. (2018), “Revising the King James Apocrypha: John Bois, Isaac Casaubon, and the Case of 1 Esdras,” in M. Feingold (ed.), Labourers in the Vineyard of the Lord. Scholarship and the Making of the King James Version of the Bible, 266–327, SLCTI 22, Leiden: Brill. Harl, M. (1986), La Bible d’Alexandrie. I. La Genèse, BdA 1, Paris: Cerf. Harl, M. (1988), “La Septante chez les Pères grecs et dans la vie des chrétiens,” in M. Harl, G. Dorival, and O. Munnich (eds.), La Bible grecque des Septante: du judaïsme hellénistique au christianisme ancien, 289–320, Initiations au christianisme ancien, Paris: Cerf. Harl, M. (1992), “Traduire la Septante en français : pourquoi et comment?” in M. Harl (ed.), La langue de Japhet. Quinze études sur la Septante et le grec des chrétiens, 33–42, Paris: Cerf. Harl, M. (1998), “Translating the LXX: Experience of ‘La Bible d’Alexandrie,’” BIOSCS 31: 31–5. Harl, M. (1999), “L’usage des commentaires patristiques pour l’étude de la Septante,” RevScRel 73: 184–201. Harl, M. (2001), “La Bible d’Alexandrie I: The Translation Principles,” in B. A. Taylor (ed.), X Congress of the International Organization for Septuagint and Cognate Studies, Oslo, 1998, 181–98, SBLSCS 51, Atlanta: SBL Press. Harl, M. (2004), La Bible en Sorbonne, ou la revanche d’Érasme, L’histoire à vif, Paris: Cerf. Harl, M. (2010), La Genèse, 3rd edn., BdA 1, Paris: Cerf. Harl, M., C. Dogniez, L. Brottier, M. Casevitz, and P. Sandevoir (1999), La Bible d’Alexandrie. XXIII.4–9. Les Douze Prophètes. Joël, Abidiou, Jonas, Naoum, Ambakoum, Sophonie, BdA 23, 4–9, Paris: Cerf. Harl, M., C. Dogniez, and M. Casevitz (2007), La Bible d’Alexandrie. XXIII.10–11. Les Douze Prophètes. Aggée, Zacharie, BdA 23, 10–11, Paris: Cerf. Harl, M., G. Dorival, and O. Munnich (1988), La Bible grecque des Septante: du judaïsme hellenistique au christianisme ancien, Initiations au christianisme ancien, Paris: Cerf. Harl, M., G. Dorival, and O. Munnich (1994), La Bible grecque des Septante: du judaïsme hellenistique au christianisme ancien, 2nd edn., Initiations au christianisme ancien, Paris: Cerf. Harl, M. and N. de Lange, eds. (1983), Origène. Philocalie, 1–20 Sur Les Écritures; et la Lettre à Africanus sur I’Histoire de Suzanne, Sources Chrétiennes 302, Paris: Cerf. Harlé, P. (1999), La Bible d’Alexandrie. VII. Les Juges, BdA 7, Paris: Cerf. Harlé, P. and D. Pralon (1988), La Bible d’Alexandrie. III. Le Lévitique, BdA 3, Paris: Cerf. Harris, R. J. (1916, 1920), Testimonies, 2 volumes, Cambridge: Cambridge University Press. Harris, W. V. (1989) Ancient Literacy, Cambridge, MA: Harvard University Press. Harvey, W. W. (1857), Sancti Irenaei: Libros quinque adversus Haereses, Cambridge: Typis Academicis. Haspelmath, M. (2009), “Lexical Borrowing: Concepts and Issues,” in M. Haspelmath and U. Tadmor (eds.), Loanwords in the World’s Languages: A Comparative Handbook, 35–54, Berlin: De Gruyter Mouton. Haspelmath, M. and U. Tadmore, eds. (2009), Loanwords in the World’s Languages: A Comparative Handbook, Berlin: De Gruyter. Hatch, E. (1889), Essays in Biblical Greek, Oxford: Oxford University Press. Hatch, E. and H. A. Redpath (1897–1906), A Concordance to the Septuagint and the Other Greek Versions of the Old testament (Including the Apocryphal Books), 2 Volumes and a Supplement, Oxford: Clarendon Press.

432 REFERENCES

Hatch, E., and H. A. Redpath (1998), A Concordance to the Septuagint and the Other Greek Versions of the Old Testament (Including the Apocryphal Books), 2nd edn., Grand Rapids: Baker Books. Hatina, T., ed. (2006), Biblical Interpretation in Early Christian Gospels, Volume 1: The Gospel of Mark, LNTS 304, New York: T&T Clark. Hatzmichali, M. (2013), “Encyclopaedism in the Alexandrian Library,” in J. König and G. Woolf (eds.), Encyclopaedism from Antiquity to the Renaissance, 64–83, Cambridge: Cambridge University Press. Hays, R. B. (1989), Echoes of Scripture in the Letters of Paul, New Haven: Yale University Press. Hays, R. B. (2005), The Conversion of the Imagination: Paul as Interpreter of Israel’s Scripture, Grand Rapids: Eerdmans. Hays, R. B. (2014), Reading Backwards: Figural Christology and the Fourfold Gospel Witness, Waco: Baylor University Press. Heater, H. (1982), A Septuagint Translation Technique in the Book of Job, CBQMS 11, Washington: Catholic Biblical Association of America. Heijmans, S. (2013), “Vocalization, Palestinian,” in G. Khan (ed.), Encyclopedia of Hebrew Language and Linguistics, Volume 3: 964a–67a, Leiden: Brill. Heine, R. E. (2010), Origen: Scholarship in the Service of the Church, Christian Theology in Context, Oxford: Oxford University Press. Helbing, R. (1907), Grammatik der Septuaginta: Laut- und Wortlehre, Göttingen: Vandenhoeck & Ruprecht. Helbing, R. (1928), Die Kasussyntax der Verba bei den Septuaginta. Ein Beitrag zur Hebraismenfrage und zur Syntax der Koine, Göttingen: Vandenhoeck & Ruprecht. Hengel, M., with R. Deines (2002), The Septuagint as Christian Scripture: Its Prehistory and the Problem of its Canon, M. E. Biddle (trans.), Edinburgh: T&T Clark. Den Hertog, C. G. (2002), “The Contribution of the Daughter Translations to the Lexicography of the Septuagint: With Special Emphasis on the Sahidic Translation of Deuteronomy 1–10,” JNSL 28: 57–67. Heschel, S. (2008), The Aryan Jesus: Christian Theologians and the Bible in Nazi Germany, Princeton: Princeton University Press. Hess, R. S. (2016), “Setting Scholarship Back a Hundred Years? Method in the Septuagint Commentary Series,” in L. K. Fuller Dow, C. A. Evans, and A. W. Pitts (eds.), The Language and Literature of the New Testament: Essays in Honor of Stanley E. Porter’s 60th Birthday, 63–8, BibInt 150, Leiden: Brill. Hesseling, D. C. (1897), Les cinq livres de la loi (le Pentateuque): traduction en néo-grec publiée en caractères hébraïques à Constantinople en 1547, transcrite et accompagnée d’une introduction, d’un glossaire et d’un facsimile, Leiden-Leipzig: van Doesburgh-Harrasowitz. Hesseling, D. C. (1901), “Le livre de Jonas,” BZ 10: 208–17. Hiebert, R. J. V. (1994), “Deuteronomy 22: 28–29 and Its Premishnaic Interpretations,” CBQ 56: 203–20. Hiebert, R. J. V. (2000), “Translation Technique in the Septuagint of Genesis and Its Implications for the NETS Version,” BIOSCS 33: 76–93. Hiebert, R. J. V. (2001), “Syriac Biblical Textual History and the Greek Psalter,” in R. J. V. Hiebert, C. E. Cox, and P. J. Gentry (eds.), The Old Greek Psalter, 178–204, JSOTSup 332, Sheffield: Sheffield Academic Press. Hiebert, R. J. V. (2004), “Lexicography and the Translation of a Translation: The NETS Version and the Septuagint of Genesis,” BIOSCS 37: 73–86.

REFERENCES 433

Hiebert, R. J. V., ed. (2010), “Translation Is Required”: The Septuagint in Retrospect and Prospect, SBLSCS 56, Atlanta: SBL Press. Hilarion (Alfeev) (2017), “Translations of the Bible: History and Present,” (in Russian) [Митрополит Волоколамский Иларион. Переводы Библии: история и современность, in Митрополит Волоколамский Иларион], Cовременная библеистика и предание церкви, 27–43, Москва: Познание. Himbaza, I. (2016), “What Are the Consequences if 4QLXXLeva Contains Earliest Formulation of the Septuagint?” in S. Kreuzer, M. Meiser, and M. Sigismund (eds.), Die Septuaginta— Orte und Intentionen: 5. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 24.-27. Juli 2014, 294–308, WUNT 361, Tubingen: Mohr Siebeck. Hody, H. (1705), De Bibliorum Textibus Originibus, Versionibus Graecis, et Latine Vulgata, Oxford: Sheldonian. Höpfl, H. (1908), Kardinal Wilhelm Sirlets Annotationen zum Neuen Testament, Freiburg: Herder. Höpfl, H. (1913), Beiträge zur Geschichte der Sixto-Klementinischen Vulgata, Freiburg: Herder. Hoey, M. (2001), Textual Interaction: An Introduction to Written Discourse Analysis, London: Routledge. Holladay, C. R. (1983a), Fragments from Hellenistic Jewish Authors, Volume I: Historians, SBLTT 20/SBLPS 10, Chico: Scholars Press. Holladay, C. R. (1983b), Fragments from Hellenistic Jewish Authors, Volume II: Poets, SBLTT 30/SBLPS 12, Chico: Scholars Press. Holladay, C. R. (1995), Fragments from Hellenistic Jewish Authors, Volume III: Aristobulus, SBLTT 39/SBLPS 13, Atlanta: Scholars Press. Holmes, R. (1789), The First Annual Account of the Collation of the MSS. of the SeptuagintVersion, Oxford: Clarendon Press. Holmes, R. and J. Parsons, eds. (1798–1827), Vetus Testamentum Graecum cum variis lectionibus, 5 volumes, Oxford: Clarendon. Holton, D., G. Horrocks, M. Janssen, T. Lendari, I. Manolessou, and N. Toufexis (2019), The Cambridge Grammar of Medieval and Early Modern Greek, 4 volumes, Cambridge: Cambridge University Press. Honigman, S. (2003), The Septuagint and Homeric Scholarship in Alexandria: A Study in the Narrative of the Letter of Aristeas, London: Routledge. Honigman, S. (2017), “The Library and the Septuagint: Between Representations and Reality,” in C. Rico and A. Dan (eds.), The Library of Alexandria: A Cultural Crossroads of the Ancient World: Proceedings of the Second Polis Institute Interdisciplinary Conference, 45–77, Jerusalem: Polis Institute Press. Horrocks, G. (2010), Greek: A History of the Language and Its Speakers, 2nd edn., Chichester: Wiley-Blackwell. Horsley, G. H. R. (1989), “The Fiction of ‘Jewish Greek’,” in G. H. R. Horsley (ed.), New Documents Illustrating Early Christianity, 5–40, North Ryde: Ancient History Documentary Research Centre, Macquarie University. Houghton, H. A. G. (2016), The Latin New Testament: A Guide to Its Early History, Texts, and Manuscripts, Oxford: Oxford University Press. Houtman, A., E. van Staalduine-sulman, and H.-M. Kirn, eds. (2014), A Jewish Targum in a Christian World, JCPS 27, Leiden: Brill. Huehnergard, J. (2013), “Philippi’s Law,” in G. Khan (ed.), Encyclopedia of Hebrew Language and Linguistics, Volume 3: 70b–1b, Leiden: Brill.

434 REFERENCES

Hughes, R. and M. McCarthy (1998), “From Sentence to Discourse: Discourse Grammar and English Language Teaching,” TQ 32: 263–87. International Phonetic Association (1999), Handbook of the International Phonetic Association: A Guide to the Use of the International Phonetic Alphabet, Cambridge: Cambridge University Press. Jacobson, H. (1983), The Exagoge of Ezekiel, Cambridge: Cambridge University Press. Jacoby, D. (2009), “Jewish Communities of the Byzantine World,” in N. de Lange, J. Krivoruchko, and C. Boyd-Taylor (eds.), Jewish Reception of Greek Bible Versions, 157–81, TSMEMJ 23, Tübingen: Mohr Siebeck. Jacques, X. (1972), List of Septuagint Words Sharing Common Elements: Supplement to Concordance or Dictionary, Subsidia Biblica 1, Rome: Biblical Institute Press. Janse, M. (2002), “Aspects of Bilingualism in the History of the Greek Language,” in J. N. Adams, M. Janse, and S. Swain (eds.), Bilingualism in Ancient Society: Language Contact and the Written Text, 332–90, Oxford: Oxford University Press. Janz, T. (2010), La Bible d’Alexandrie. XI.2. Deuxième Livre d’Esdras, BdA 11.2, Paris: Cerf. Jassen, A. P. (2016), “The Prophets in the Dead Sea Scrolls,” in C. J. Sharp (ed.), The Oxford Handbook of The Prophets, 353–72, Oxford: Oxford University Press. Jay, P. (1985), L’Exégèse de Saint Jérôme d’après son “Commentaire sur Isaïe”, Paris: Études Augustiniennes. Jellicoe, S. (1968), The Septuagint and Modern Study, Oxford: Clarendon Press. Jellicoe, S. (1969), “Septuagint Studies in the Current Century,” JBL 88: 191–9. Jenkins, R. G. (1991), “Colophons of the Syrohexapla and the Textgeschichte of the Recensions of Origen,” in C. E. Cox (ed.), VII Congress of the International Organization for Septuagint and Cognate Studies, Leuven 1989, 261–77, SBLSCS 31, Atlanta: Scholars Press. Jenkins, R. G. (1998a), “Hexaplaric Marginalia and the Hexapla-Tetrapla Question,” in A. Salvesen (ed.), Origen’s Hexapla and Fragments: Papers Presented at the Rich Seminar on the Hexapla, Oxford Centre for Hebrew and Jewish Studies, 25th July–3rd August 1994, 73–87, TSAJ 58, Tübingen: Mohr Siebeck. Jenkins, R. G. (1998b), “The First Column of the Hexapla: The Evidence of the Milan Codex (Rahlfs 1098) and the Cairo Genizah Fragment (Rahlfs 2005),” in A. Salvesen (ed.), Origen’s Hexapla and Fragments: Papers Presented at the Rich Seminar on the Hexapla, Oxford Centre for Hebrew and Jewish Studies, 25th July–3rd August 1994, 88–102, TSAJ 58, Tübingen: Mohr Siebeck. Jenni, E. and C. Westermann (1997), Theological Lexicon of the Old Testament, Peabody, MA: Hendrickson Publishers. Jobes, K. H., ed. (2016), Discovering the Septuagint: A Guided Reader, Grand Rapids: Kegel. Jobes, K. H., and M. Silva (2000), Invitation to the Septuagint, Grand Rapids: Baker Academic. Jobes, K. and M. Silva (2015), Invitation to the Septuagint, 2nd edn., Grand Rapids: Baker Academic. Johnson, A. P. (2006), Ethnicity and Argument in Eusebius’, Praeparatio Evangelica, Oxford: Oxford University Press. Johnson, B. J. M. (2016), “Reading the Septuagint: The Hermeneutical Problem of a Translated Text,” in D. Batovici and K. De Troyter, Authoritative Texts and Reception History: Aspects and Approaches, 20–40, BibInt 151, Leiden: Brill. Johnson, L. T. (2002), Septuagintal Midrash in the Speeches of Acts, Milwaukee: Marquette University Press. Jones, B. A. (1995), The Formation of the Twelve: A Study in Text and Canon, SBLDS 149, Atlanta: Scholars Press.

REFERENCES 435

Jones, R. N. (1992), “Embalming,” in D. N. Freedman (ed.), The Anchor Bible Dictionary, Volume 2: 490–6, New York: Doubleday. Jongkind, D. (2013), Scribal Habits of Codex Sinaiticus, TS 5, Piscataway: Gorgias Press. Joosten, J. (2000), “Une théologie de la Septante? Réflexions méthodologiques sur l’interprétation de la version grecque,” RTP 132: 31–46. Joosten, J. (2003), “On Aramaising Renderings in the Septuagint,” in M. F. J. Baasten and W. Th. van Peursen (eds.), Hamlet on a Hill: Semitic and Greek Studies Presented to Professor T. Muraoka on the Occasion of His Sixty-Fifth Birthday, 587–600, OLA 118, Leuven: Peeters. Joosten, J. (2005), “Source-Language Oriented Remarks on the Lexicography of the Greek Versions of the Bible,” ETL 81: 152–64. Joosten, J. (2006), “Le milieu producteur du Pentateuque Grec,” REJ 165: 349–61. Joosten, J. (2007b), “The Original Language and Historical Milieu of the Book of Judith,” in M. Ben-Asher and E. Tov (eds.), Meghillot 5–6: A Festschrift for Devorah Dimant, *159–*176, Jerusalem: Bialik Institute. Joosten, J. (2008a), “Reflections on the ‘Interlinear Paradigm,’ in Septuagintal Studies.” in A. Voitila and J. Jokiranta (eds.), Scripture in Transition: Essays on Septuagint, Hebrew Bible, and Dead Sea Scrolls in Honour of Raija Sollamo, 163–78, JSJSup 126, Leiden: Brill. Joosten, J. (2008b), Review of Susan Brayford (2007), Genesis, SEPT, Leiden: Brill, RBL, Online: http://www.bookreviews.org. Joosten, J. (2015), “Reflections on the Original Language of the Psalms of Solomon,” in E. Bons and P. Pouchelle (eds.), The Psalms of Solomon: Language, History, Theology, 31–47, SBLEJL 40, Atlanta: SBL Press. Joosten, J. (2016a), “Septuagint Greek and the Jewish Sociolect in Egypt,” in E. Bons and J. Joosten (eds.), Die Sprache der Septuaginta/The Language of the Septuagint, 246–56, LXX.H 3, Gütersloh: Gütersloher Verlagshaus. Joosten, J. (2016b), “The Origin of the Septuagint Canon,” in S. Kreuzer, M. Meiser, and M. Sigismund (eds.), Die Septuaginta—Orte und Intentionen: 5. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 24.-27. Juli 2014, 688–99, WUNT 361, Tübingen: Mohr Siebeck. Joosten, J., and E. Bons, eds. (2011), Septuagint Vocabulary: Pre-History, Usage, Reception, SBLSCS 58, Atlanta: SBL Press. Joosten, J. and P. J. Tomson, eds. (2007), Voces Biblicae: Septuagint Greek and its Significance for the New Testament, CBET 49, Leuven: Peeters. Joüon, P. and T. Muraoka (2006), A Grammar of Biblical Hebrew, 2nd edn., SubBi 27, Rome: Gregorian & Biblical Press. Jüngling, H.-W., H. von Lips, and R. Scoralick (2011), “Paroimiai/Proverbia/Sprichwörter/ Sprüche Salomos” in M. Karrer and W. Kraus (eds.), Septuaginta Deutsch. Erläuterungen und Kommentare zum griechischen Alten Testament, 1950–2000, Volume 2: Psalmen bis Daniel, Stuttgart: Deutsche Bibelgesellschaft. Kahle, P. E. (1959), The Cairo Geniza, 2nd edn., Oxford: Blackwell. Kamesar, A. (1990), “The Virgin of Isaiah 7:14: The Philological Argument from the Second to the Fifth Century,” JTS 41: 51–75. Kamesar, A. (1993), Jerome, Greek Scholarship, and the Hebrew Bible: A Study of the Quaestiones Hebraicae in Genesim, Oxford: Oxford University Press. Kamesar, A. (2009), “Biblical Interpretation in Philo,” in A. Kamesar (ed.), The Cambridge Comanion to Philo, 65–91, Cambridge: Cambridge University Press. Kamesar, A. (2013), “Jerome,” in J. Carleton Paget and J. Schaper (eds.), The New Cambridge History of the Bible, Volume 1: From the Beginnings to 600, 653–75, Cambridge: Cambridge University Press.

436 REFERENCES

Kaminka, A. (1928), “Studien zur Septuaginta an der Hand der zwölf kleinen Prophetenbücher,” MGWJ 72: 49–60, 242–73. Kannengiesser, C. (2004), Handbook of Patristic Exegesis: The Bible in Ancient Christianity, 2 volumes, Leiden: Brill. Kantor, B. P. (2017), The Second Column (Secunda) of Origen’s Hexapla in Light of Greek Pronunciation, Ph.D. dissertation, University of Texas, Austin. Kappler, W., ed. (1990), Septuaginta. Vetus Testamentum Graecum. IX.1. Maccabaeorum Liber I, 3rd edn., Göttingen: Vandenhoeck & Ruprecht. Karo, G. and Lietzmann, J. (1902), “Catenarum Graecarum Catalogus,” in Akademie der Wissenschaften in Göttingen, Nachrichten von der Königl. Gesellschaft der Wissenschaften zu Göttingen, Philologisch-Historische Klasse, 31–19, Göttingen: Lüder Horstmann. Karrer, M. (2008), “Septuaginta Deutsch (LXX.D). Characteristics of the German Translation Project,” in H. Ausloos, J. Cook, F. García Martínez, B. Lemmelijn, and M. Vervenne (eds.), Translating a Translation, 105–18, BETL 213, Leuven: Peeters. Karrer, M. (2010), “LXX Psalm 39: 7-10in Hebrews 10: 5-7,” in D. J. Human and G. J. Steyn (eds.), Psalms and Hebrews: Studies in Reception, LHBOTS 527, 126–46, New York: T&T Clark. Karrer, M. and W. Kraus, eds. (2011a), Septuaginta Deutsch: Erläuterungen und Kommentare zum griechischen Alten Testament, 2 volumes, Volume 1: Genesis bis Makkabäer, Stuttgart: Deutsche Bibelgesellschaft. Karrer, M. and W. Kraus, eds. (2011b), Septuaginta Deutsch: Erläuterungen und Kommentare zum griechischen Alten Testament, 2 volumes, Volume 2: Psalmen bis Daniel, Stuttgart: Deutsche Bibelgesellschaft. Kato, T. (2013), “Jerome’s Understanding of Old Testament Quotations in the New Testament,” VC 67: 289–315. Katz, P. (1956), “Septuagintal Studies in the Mid-Century: Their Links with the Past and Their Present Tendencies,” in W. D. Davies and D. Daube (eds.), The Background of the New Testament and its Eschatology, 176–208, Cambridge, Cambridge University Press. Kauhanen, T. (2012), The Proto-Lucianic Problem in 1 Samuel, DSI 3, Göttingen: Vandenhoeck & Ruprecht. Kenyon, F. G. (1957), The Codex Alexandrinus (Royal MS. 1 D. V-VIII) in Reduced Photographic Facsimile: Old Testament Part IV 1 Esdras—Ecclesiasticus, London: The Trustees of the British Museum. Khan, G. (2012), A Short Introduction to the Tiberian Masoretic Bible and Its Reading Tradition, Piscataway, NJ: Gorgias Press. Khan, G., ed. (2013a), Encyclopedia of Hebrew Language and Linguistics, 4 volumes, Leiden: Brill. Khan, G. (2013b), “Biblical Hebrew: Pronunciation Traditions,” in G. Khan (ed.), Encyclopedia of Hebrew Language and Linguistics, Volume 1: 341a–52b, Leiden: Brill. Khan, G. (2013c), “Vocalization, Babylonian,” in G. Khan (ed.), Encyclopedia of Hebrew Language and Linguistics, Volume 3: 953a–63b, Leiden: Brill. Kharanauli, A. (2018), “Origen and Lucian in the Light of Ancient Editorial Techniques,” in A. Aejmelaeus, D. Longacre, and N. Mirotadze (eds.), From Scribal Error to Rewriting: How Ancient Texts Could and Could Not Be Changed, 15–52, DSI 12, Göttingen: Vandenhoeck & Ruprecht. Kielsmeier-Jones, A. (2012), “How to Read and Understand the Göttingen Septuagint: A Short Primer, Part 1,” Words on the Word, November 4, 2012, Online: https://abramkj. com/2012/11/04/how-to-read-and-understand-the-gottingen-septuagint-a-short-primerpart-1/ (accessed February 21, 2020).

REFERENCES 437

Kiessling, E. and F. Preisigke (1925–), Wörterbuch der griechischen Papyrusurkunden mit Einschluss der griechischen Inschriften, Aufschriften, Ostraka, Mumienschilder, usw., aus Ägypten, Berlin/Wiesbaden: Harrassowitz. King, N. (2013), The Bible: A Study Bible Freshly Translated by Nicholas King, Suffolk: Kevin Mayhew. Kippenberg, H. G. (2002), Discovering Religious History in the Modern Age, B. Harshav (trans.), Princeton: Princeton University Press. Kircher, C. (1607), Concordantiae Veteris Testamenti Graecae, Ebreis vocibus respondents, 2 volumes, Francfurt am Main: C. Marnium & heredes I. Aubrii. Kircher, C. (1622), Concordantiarum biblicarum maxime veteris Testamenti graecarum, hebraeis vocibus respondentium, Wittenburg: Sumptibus Caspari Heiden bibl. Typis Johannis Gormanni. Kitchen, K. A. (2003), On the Reliability of the Old Testament, Grand Rapids: Eerdmans. Kittel, G. and G. Friedrich (1964), Theological Dictionary of the New Testament, G. W. Bromiley (trans.), Grand Rapids: Eerdmans. Klein, R. W. (1974), Textual Criticism of the Old Testament: From the Septuagint to Qumran, Philadelphia: Fortress Press. Knobloch, F. W. (1995), Hebrew Sounds in Greek Script: Transcriptions and Related Phenomena in the Septuagint, with Special Focus on Genesis, Ph.D. dissertation, University of Pennsylvania, Philadelphia. Koch, K. (2005), Daniel. 1. Teilband Daniel 1–4, BKAT 22.1, Neukirchen-Vluyn: Neukirchener Verlag. Koch, K. and M. Rösel (2000), Polyglottensynopse zum Buch Daniel, Neukirchen-Vluyn: Neukirchener Verlag. Koenig, J. (1982), L’Herméneutique analogique du Judaïsme antique d’après les témoins textuels d’Isaïe, VTSup 33, Leiden: Brill. Koller, A. (2013), “Attenuation,” in G. Khan (ed.), Encyclopedia of Hebrew Language and Linguistics, volume 1: 231b–2b, Leiden: Brill. Konstantinou, M. (2012), “Bible Translation and National Identity: The Greek Case,” ΣΥΝΘΕΣΙΣ 1: 34–53. Koriwn (1941), Vark‘ Mashtots’i, M. Abeghyan (ed. and trans.), Yerevan: Haypethrat. Korteweg, P. (2006). De nieuwtestamentische commentaren van Johannes Drusius (1550– 1616), Leiden: Melissant. Kraft, R. A. (1972a), “Approaches to Translation Greek Lexicography,” in R. Kraft (ed.), Septuagintal Lexicography, 31–5, SBLSCS 1, Missoula: Scholars Press. Kraft, R. A., ed. (1972b), Septuagintal Lexicography, SBLSCS 1, Missoula: Scholars Press. Kraft, R. A. (2003), “The ‘Textual Mechanics’ of Early Jewish LXX/OG Papyri and Fragments,” in S. McKendrick and O. A. O’Sullivan (eds.), The Bible as Book: The Transmission of the Greek Text, 51–72, London: The British Library. Kraft, R. A. (2013), “What Did Constantine Want from Eusebius? The 50 Booklets of Sacred Scriptures Badly/Sadly Misunderstood,” Paper presented at the annual meeting of the Society of Biblical Literature, Baltimore. Kraus, M. (2003), “Hebraisms in the Old Latin Version of the Bible,” VT 53: 487–513. Kraus, M. A. (2017), Jewish, Christian, and Classical Exegetical Traditions in Jerome’s Translation of the Book of Exodus: Translation Technique and the Vulgate, VCSup 141, Leiden: Brill. Kraus, W. (2006), “Contemporary Translations of the Septuagint: Problems and Perspectives,” in W. Kraus and R. Glenn Wooden (eds.), Septuagint Research: Issues and Challenges in the Study of the Greek Jewish Scriptures, 63–83, SBLSCS 53, Atlanta: SBL Press.

438 REFERENCES

Kraus, W. (2010), “Septuaginta Deutsch (LXX.D): The Value of a German Translation of the Septuagint,” in R. J. V. Hiebert (ed.), “Translation Is Required”: The Septuagint in Retrospect and Prospect, 243–8, SBLSCS 56, Atlanta: SBL Press. Kraus, W. and M. Karrer (2005), “Septuaginta Deutsch Annual Report to IOSCS, October 2005,” Online: http://ccat.sas.upenn.edu/ioscs/reports/deutsch2005.html (accessed January 14, 2020). Kraus, W. and M. Karrer (2009a), “Einführung in den Gebrauch des Übersetzungsbandes,” in W. Kraus and M. Karrer (eds.), Septuaginta Deutsch: Das griechische Alte Testament in deutscher Übersetzung, xvii–xxiii, Stuttgart: Deutsche Bibelgesellschaft. Kraus, W. and M. Karrer, eds. (2009b), Septuaginta Deutsch: Das griechische Alte Testament in deutscher Übersetzung, Stuttgart: Deutsche Bibelgesellschaft. Kraus, W. and R. G. Wooden, eds. (2006), Septuagint Research: Issues and Challenges in the Study of the Greek Jewish Scriptures, SBLSCS 53, Atlanta: SBL Press. Krašovec, J. (2010), The Transformation of Biblical Proper Names, LHBOTS 418, London: T&T Clark. Kreuzer, S. (2001), “A German Translation of the Septuagint,” BIOSCS 34: 40–5. Kreuzer, S. (2004), “Lexicography and Translation: Experiences, Examples, and Expectations in the Context of the Septuaginta-Deutsch Project,” BIOSCS 37: 107–17. Kreuzer, S. (2010), “A Reply to M. Law and T. Kauhanen, ‘Methodological Remarks …’,” BIOSCS 43: 89–95. Kreuzer, S. (2013), “Der antiochenische Text der Septuaginta. Forschungsgeschichte und eine neue Perspektive,” in S. Kreuzer and M. Sigismund (eds.), Der Antiochenische Text der Septuaginta in seiner Bezeugung und seiner Bedeutung, 23–56, DSI 4, Göttingen: Vandenhoeck & Ruprecht. Kreuzer, S. (2015a), “Origin and Development of the Septuagint,” in S. Kreuzer (ed.), The Bible in Greek: Translation, Transmission, and Theology of the Septuagint, 3–46, SBLSCS 63, Atlanta: SBL Press. Kreuzer, S. (2015b), The Bible in Greek: Translation, Transmission, and Theology of the Septuagint, SBLSCS 63, Atlanta: SBL Press. Kreuzer, S., ed. (2016), Einleitung in die Septuaginta, LXX.H 1, Gütersloh: Gütersloher Verlaghaus. English translation Kreuzer (2019). Kreuzer, S. (2017), “Septuagint and Septuagint Research in Germany,” JSCS 50: 71–104. Kreuzer, S. (2018), “Zur Relevanz editorischer Prinzipien,” in M. Meiser, M. Geiger, S. Kreuzer, and M. Sigismund (eds.), Die Septuaginta—Geschichte, Wirkung, Relevanz. 6. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 21.-24. Juli 2016, 130–45, WUNT 405, Tübingen: Mohr Siebeck. Kreuzer, S., ed. (2019), Introduction to the Septuagint, D. A. Brenner and P. Altman (trans.), Waco: Baylor University Press. Translation of Kreuzer (2016). Kristeva, J. (1980), Desire in Language: A Semiotic Approach to Literature and Art, L. S. Roudiez (ed.), T. Gora, A. Jardine, and L. S. Roudiez (trans.), New York: Columbia University Press. Krivoruchko, J. G. (2008), “The Constantinople Pentateuch within the Context of Septuagint Studies,” in M. K. H. Peters (ed.), XIII Congress of the International Organization for Septuagint and Cognate Studies, Ljubljana, 2007, 255–76, SBLSCS 55, Atlanta: SBL Press. Kühner, R. and B. Gerth (1904), Ausführliche Grammatik der griechischen Sprache, Hannover/ Leipzig: Hahn. Kujanpää, K. (2018), The Rhetorical Functions of Scriptural Quotations in Romans: Paul’s Argumentation by Quotations, NovTSup 172, Leiden: Brill.

REFERENCES 439

Kutscher, E. Y. (1965), “Contemporary Studies in North-Western Semitic,” JSS 10: 21–51. Kutscher, E. Y., A. Ben-Ḥayyim, A. Dotan, and G. Sarfatti, eds. (1977), Hebrew and Aramaic Studies, Jerusalem: Magnes. Ladefoged, P., and K. Johnson (2011), A Course in Phonetics, 6th edn., Wadsworth: Cengage Learning. de Lagarde, P. (1883), Librorum Veteris Testamenti Canonicorum Pars Prior Graece, Göttingen: Arnoldi Hoyer. Lambrecht, K. (1998), Information Structure and Sentence Form: Topic, Focus, and the Mental Representations of Discourse Referents, CSL 71, Cambridge: Cambridge University Press. Lampe, G. W. H. (1961), A Patristic Greek Lexicon, Oxford: Clarendon Press. Langacker, R. W. (2008), Cognitive Grammar: A Basic Introduction, Oxford: Oxford University Press. Lange, A. (2016a), “Ancient Hebrew-Aramaic Texts,” in A. Lange and E. Tov (eds.), Textual History of the Bible, Volume 1A: 82–166, Leiden: Brill. Lange, A. (2016b), “Canonical History of the Hebrew Bible,” in A. Lange and E. Tov (eds.), Textual History of the Bible, Volume 1A: 35–81, Leiden: Brill. Lange, A. (2017a), “Samuel–Kings: Ancient and Late Ancient Manuscript Evidence,” in A. Lange and E. Tov (eds.), Textual History of the Bible, Volume 1B: 319–32, Leiden: Brill. Lange, A. (2017b), “Jeremiah: Ancient Hebrew-Aramaic Text,” in A. Lange and E. Tov (eds.), Textual History of the Bible, Volume 1B: 514–42, Leiden: Brill. Lange, A. and E. Tov, eds. (2016), Textual History of the Bible, Volume 1: The Hebrew Bible, A: Overview Articles, Leiden: Brill. Lange, A. and E. Tov, eds. (2017a), Textual History of the Bible, Volume 1: The Hebrew Bible, B: Pentateuch, Former and Latter Prophets, Leiden: Brill. Lange, A. and E. Tov, eds. (2017b), Textual History of the Bible, Volume 1: The Hebrew Bible, B: Writings, Leiden: Brill. de Lange, N. (1980), “Some New Fragments of Aquila on Malachi and Job,” VT 30: 291–4. de Lange, N. (1993), “The Jews of Byzantium and the Greek Bible: Outline of the Problems and Suggestions for Future Research,” in G. Sed-Rajna (ed.), Rashi 1040–1998: Hommage à Ephraïm E. Urbach: Congrès Européen des études juives, 203–10, Paris: Cerf. de Lange, N. (1995), “Hebrew/Greek Manuscripts: Some Notes,” JJS 46: 262–70. de Lange, N. (1996), Greek Jewish Texts from the Cairo Genizah, TSAJ 51, Tübingen: Mohr Siebeck. de Lange, N. (2003), “The Greek Glosses of the Fitzwilliam Museum Bible (MS 364*),” in S. Berger, M. Brocke, and I. Zwiep (eds.), Zutot 2002, 138–47, London/Dordrecht: Kluwer. de Lange, N. (2005), “Jews in the Age of Justinian,” in M. Maas (ed.), The Cambridge Companion to the Age of Justinian, 401–26, Cambridge: Cambridge University Press. de Lange, N. (2007), “An Early Hebrew–Greek Bible Glossary from the Cairo Genizah and its Significance for the Study of Jewish Bible Translations into Greek,” in M. F. J. Baasten and R. Munk (eds.), Studies in Hebrew Literature and Jewish Culture: Presented to Albert van der Heide on the Occasion of His Sixty-Fifth Birthday, 31–9, ASJT 12, Dordrecht: Springer. de Lange, N. (2008), “Jewish Transmission of Greek Bible Versions,” in M. K. H. Peters (ed.), XIII Congress of the International Organization for Septuagint and Cognate Studies, Ljubljana, 2007, 109–17, SBLSCS 55, Atlanta: SBL Press. de Lange, N., J. Krivoruchko, and C. Boyd-Taylor, eds. (2009), Jewish Reception of Greek Bible Versions, TSMEMJ 23, Tübingen: Mohr Siebeck.

440 REFERENCES

de Lange, N. (2010), “The Greek Bible Translations of the Byzantine Jews,” in P. Magdalino and R. Nelson (eds.), The Old Testament in Byzantium, 39–54, Washington: Dumbarton Oaks. de Lange, N. (2012), “The Greek Bible in the Medieval Synagogue,” in R. Bonfil, O. Irshai, G. G. Stroumsa, and R. Talgam (eds.), Jews in Byzantium: Dialectics of Minority and Majority Cultures, 371–84, Leiden: Brill. de Lange, N. (2015), Japheth in the Tents of Shem: Greek Bible Translations in Byzantine Judaism, TSMEMJ 30, Tübingen: Mohr Siebeck. Lanier, G. R. and W. A. Ross, eds. (2018), Septuaginta: A Reader’s Edition, 2 volumes, Peabody: Hendrickson. Lanier, G. R. and W. A. Ross (2019), A Book-by-Book Guide to Septuagint Vocabulary. Peabody: Hendrickson. Lass, R. (1984), Phonology: An Introduction to Basic Concepts, Cambridge Textbooks in Linguistics, Cambridge: Cambridge University Press. Laver, J. (1994), Principles of Phonetics, Cambridge Textbooks in Linguistics, Cambridge: Cambridge University Press. Law, T. M. (2008), “Origen’s Parallel Bible: Textual Criticism, Apologetics, or Exegesis?” JTS 59: 1–21. Law, T. M. (2009), “The Glosses on Solomon’s Building Campaign,” in N. de Lange, J. Krivoruchko, and C. Boyd-Taylor (eds.), Jewish Reception of Greek Bible Versions, 263–83, TSMEMJ 23, Tübingen: Mohr Siebeck. Law, T. M. (2013), When God Spoke Greek: The Septuagint and the Making of the Christian Bible, Oxford: Oxford University Press. Law, T. M. and T. Kauhanen (2010), “Methodological Remarks on the Textual History of Reigns: A Response to Siegfried Kreuzer,” BIOSCS 43: 73–88. Leaney, A. R. C. (1976), “Greek Manuscripts from the Judaean Desert,” in J. K. Elliot (ed.), Studies in New Testament Language and Text, NovTSup 44, 283–300, Leiden: Brill. Lebram, J. C. H. (1975), “Ein Streit um die Hebraïsche Bibel und die Septuaginta,” in Th. H. Lunsingh Scheurleer and G. H. M. Posthumus Meyjes (eds.), Leiden University in the Seventeenth Century: An Exchange of Learning, 20–63, Leiden: Brill. Lee, J. A. L. (1969), “A Note on Septuagint Material in the Supplement to Liddell and Scott,” Glotta 47: 234–42. Lee, J. A. L. (1983), A Lexical Study of the Septuagint Version of the Pentateuch, SBLSCS 14, Chico: Scholars Press. Lee, J. A. L. (2016), “The Vocabulary of the Septuagint and Documentary Evidence,” in E. Bons and J. Joosten (eds.), Die Sprache der Septuaginta/The Language of the Septuagint, 98–108, LXX.H 3, Gütersloh: Gütersloher Verlagshaus. Lee, J. A. L. (2018), The Greek of the Pentateuch: Grinfield Lectures On The Septuagint 2011–2012, Oxford: Oxford University Press. Lee, J. A. L. (2020), “Back to the Question of Greek Idiom,” in T. Kauhanen and H. Vanonen (eds.), The Legacy of Soisalon-Soininen: Towards a Syntax of Septuagint Greek, n.p., DSI 13, Göttingen: Vandenhoeck & Ruprecht. Leemans, J. (1997), “Athanasius and the Book of Wisdom,” ETL 73: 349–68. Lemmelijn, B. (2001), “Two Methodological Trails in Recent Studies on the Translation Technique of the Septuagint,” in R. Sollamo and S. Sipilä (eds.), Helsinki Perspectives on the Translation Technique of the Septuagint, 43–63, PFES 82, Göttingen: Vandenhoeck & Ruprecht. Lemmelijn, B. (2009), A Plague of Texts? A Text-Critical Study of the So-Called ‘Plague Narrative’ in Exodus 7: 14–11:10, OtSt 56, Leiden: Brill.

REFERENCES 441

Léonas, A. (2008), “The Septuagint in Premodern Study: A Bibliography,” BIOSCS 41: 93–113. Léonas, A. (2016), “The Language of the Septuagint between Greek and Hebrew,” in E. Bons and J. Joosten (eds.), Die Sprache der Septuaginta/The Language of the Septuagint, 357–74, LXX.H 3, Gütersloh: Gütersloher Verlagshaus. Lestienne, M. and B. Grillet (1997), La Bible d’Alexandrie. IX.1. Premier Livre des Règnes, BdA 9.1, Paris: Cerf. Levinsohn, S. H. (2000), Discourse Features of New Testament Greek: A Coursebook on the Information Structure of New Testament Greek, 2nd edn., Dallas: SIL International. Levitin, D. (forthcoming), “Historicising the New Testament Canon, c. 1700,” in D. Levitin and E. Maclean (eds.), Beyond Ancients and Moderns. Comparative Approaches to the Reception of the Classical Tradition in Early Modern Europe, c. 1600–1750, Oxford: Oxford University Press. Lewis, J. P. (1964), “What Do We Mean By Jabneh?” JBR 32: 125–32. Liddell, H. G., R. Scott, and H. S. Jones (1925–40), A Greek-English Lexicon: A New Edition, Oxford: Clarendon Press. Liddell, H. G., R. Scott, and H. S. Jones (1968), Greek-English Lexicon: A Supplement, Oxford: Clarendon Press. Liddell, H. G., R. Scott, and H. S. Jones (1996), Greek-English Lexicon: Revised Supplement, Oxford: Clarendon Press. Lim, T. H. (2000), “The Qumran Scrolls, Multilingualism, and Biblical Interpretation,” in J. J. Collins and R. A. Kugler (eds.), Religion in the Dead Sea Scrolls, 57–73, Grand Rapids: Eerdmans. Lim, T. H. (2013), The Formation of the Jewish Canon, New Haven: Yale University Press. Linde, C. (2011), How To Correct the Sacra Scriptura? Textual Criticism and the Latin Bible between the Twelfth and Fifteenth Century, Oxford: Society for the Study of Medieval Languages and Literature. Lisowsky, G. (1940), Die Transskription der hebraeischen Eigennamen des Pentateuch in der LXX, Ph.D. dissertation, Universität Basel, Basel. Littman, R. J. (2008), Tobit, SEPT, Leiden Brill. Litwak, K. D. (2005), Echoes of Scripture in Luke-Acts: Telling the History of God’s People Intertextually, London: T&T Clark. Livanios, D. (2014), “‘In the Beginning Was the Word’: Orthodoxy and Bible Translation into Modern Greek (16th–19th Centuries),” Mediterranean Chronicle 4: 101–20. Loiseau, A.-F. (2016), L’influence de l’araméen sur les traducteurs de la LXX principalement, sur les traducteurs grecs postérieurs, ainsi que sur les scribes de la Vorlage de la LXX, SBLSCS 65, Atlanta: SBL Press. Longacre, R. E. and S. J. J. Hwang (2012), Holistic Discourse Analysis, 2nd edn., Dallas: SIL International. Longenecker, R. N. (1975), Biblical Exegesis in the Apostolic Period, Carlisle: Paternoster. Lougee, R. W. (1962), Paul de Lagarde, 1827–1891. A Study of Radical Conservatism in Germany, Cambridge, MA: Harvard University Press. Lourenço, F., ed. and trans. (2017), Bíblia, Volume 3: Antigo Testamento, Os Livros Proféticos, Lisbon: Quetzal Editores. Lourenço, F., ed. and trans. (2018), Bíblia, Volume 4.1: Antigo Testamento. Os Livros Sapiencias, Lisbon: Quetzal Editores. Lourenço, F., ed. and trans. (2019), Bíblia, Volume 4.2: Antigo Testamento. Os Livros Sapiencias, Lisbon: Quetzal Editores.

442 REFERENCES

Louw, J. P. and E. A. Nida (1989), Greek-English Lexicon of the New Testament Based on Semantic Domains, New York: United Bible Societies. Lucca, P., C. Martone, and L. Mazzinghi, eds. (2012–16), La Bibbia dei Settanta, 3 volumes, Brescia: Morcelliana. Lust, J. (1993), “Translation Greek and the Lexicography of the Septuagint,” JSOT 59: 109–20. Lust, J. (1999), “Selective Bibliography of the Septuagint,” Online: http://www.bibleresearcher.com/lxx-bibliography.html (accessed February 28, 2020). Lust, J. (2003), “Introduction,” in J. Lust, E. Eynikel, and K. Hauspie (eds.), Greek-English Lexicon of the Septuagint, revised edn., xi–xxiv, Stuttgart: Deutsche Bibelgesellschaft. Lust, J., E. Eynikel, and K. Hauspie (1992, 1996), Greek-English Lexicon of the Septuagint, 2 volumes, Stuttgart: Deutsche Bibelgesellschaft. Lust, J., E. Eynikel, and K. Hauspie (2003), Greek-English Lexicon of the Septuagint, revised edn., Stuttgart: Deutsche Bibelgesellschaft. Lust, J., E. Eynikel, and K. Hauspie (2015), Greek-English Lexicon of the Septuagint, 3rd edn., Stuttgart: Deutsche Bibelgesellschaft. Lys, D. (1966), “The Israelite Soul According to the LXX,” VT 16: 181–228. MacDonald, P. J. (1992), “Discourse Analysis and Biblical Interpretation,” in W. R. Bodine (ed.), Linguistics and Biblical Hebrew, 153–75. Winona Lake: Eisenbrauns. Mäkipelto, V. (2018), Uncovering Ancient Editing: Documented Evidence of Changes in Joshua 24 and Related Texts, BZAW 513, Berlin: De Gruyter. Magdalino, P. and N. Robert, eds. (2010), The Old Testament in Byzantium, Washington: Dumbarton Oaks. Mai, A. (1825), Scriptorum veterum nova collection è Vaticanis codicibus edita, Volume 1, Part 2, 1 Auflage, Rome: In Collegio Urbano apud Burlaeum. Maksim Grek (1862), Works (in Russian) [Сочинения Преподобного Максима Грека, Казань] Mandelbrote, S. (2006), “English Scholarship and the Greek Text of the Old Testament, 16201720: The Impact of Codex Alexandrinus,” in A. Hessayon and N. Keene (eds.), Scripture and Scholarship in Early Modern England, 74–93, Aldershot: Ashgate. Mandelbrote, S. (2012), “Isaac Vossius and the Septuagint,” in E. Jorink and E. van Miert (eds.), Isaac Vossius (1618–1689) between Science and Scholarship, 85–117, BSIH 214, Leiden: Brill. Mandelbrote, S. (2013), “Making the Bible English,” History Workshop Journal 75: 265–73. Mandelbrote, S. (2016a), “The Letter of Aristeas: Three Phases in the Readership of a Jewish Text,” in S. Mandelbrote and J. Weinberg (eds.), Jewish Books and their Readers: Aspects of the Intellectual Life of Christians and Jews in Early Modern Europe, 15–44, BSCHRC 75, Leiden: Brill. Mandelbrote, S. (2016b), “When Manuscripts Meet: Editing the Bible in Greek during and after the Council of Trent,” in A. Blair and A.-S. Goeing (eds.), For the Sake of Learning: Essays in Honor of Anthony Grafton, 2 volumes, 1: 251–67, SLCTI 18, Leiden: Brill. Mandelbrote, S. (2016c), “The Old Testament and Its Ancient Versions in Manuscript and Print in the West, from c. 1480 to c. 1780,” in E. Cameron (ed.), The New Cambridge History of the Bible, Volume 3: From 1450 to 1750, 82–109, Cambridge: Cambridge University Press. Mandelbrote, S. (2017), “Witches and Forgers: Anthonie van Dale on Biblical History and the Authority of the Septuagint,” in D. van Miert, H. Nellen, P. Steenbakkers, and J. Touber (eds.), Scriptural Authority and Biblical Criticism in the Dutch Golden Age, 270–306, Oxford: Oxford University Press.

REFERENCES 443

Mandelbrote, S. (2019), “Philology and Scepticism: Early Modern Scholars at Work on the Text of the Bible,” in G. M. Cao, A. Grafton, and J. Kraye (eds.), The Marriage of Philology and Scepticism: Uncertainty and Conjecture in Early Modern Scholarship and Thought, 123–42, London: Warburg Institute. Manson, T. W. (1938), “Sadducee and Pharisee—The Origin and Significance of the Names,” BJRL 22: 144–59. Marchand, S. L. (2009), German Orientalism in the Age of Empire, Cambridge: Cambridge University Press. Margolis, M. L. (1910), “Complete Induction for the Identification of the Vocabulary in the Greek Versions of the Old Testament with Its Semitic Equivalents: Its Necessity and the Means of Obtaining It,” JAOS 30: 301–12. Margolis, M. L. (1931–8), The Book of Joshua in Greek According to the Critically Restored Text with an Apparatus Containing the Variants of the Principal Recensions and of the Individual Witnesses, 4 volumes, Paris: Paul Geuthner. Margolis, M. L. (1992), The Book of Joshua in Greek According to the Critically Restored Text with an Apparatus Containing the Variants of the Principal Recensions and of the Individual Witnesses, Volume 5, Philadelphia: Annenberg Research Institute. Marshall, P. S. (2007), “A Critical Edition of the Hexaplaric Fragments of Ecclesiastes,” Ph.D. dissertation, Southern Baptist Theological Seminary, Louisville. Martens, P. W. (2012), Origen and Scripture: The Contours of the Exegetical Life, OECS, Oxford: Oxford University Press. Martin, R. A. (1974), Syntactical Evidence of Semitic Sources in Greek Documents, SBLSCS 3, Cambridge: Scholars Press. Martone, C. (2005), “Qumran Readings in Agreement with the Septuagint against the Masoretic Text, Part One: The Pentateuch,” Henoch 27: 53–113. Martone, C. (2007), “Qumran Readings in Agreement with the Septuagint against the Masoretic Text, Part Two: Joshua–Judges,” in A. Hilhorst, E. Puech, and E. Tigchelaar (eds.), Flores Florentino: Dead Sea Scrolls and Other Early Jewish Studies in Honour of Florentino García Martínez, JSJSup 122, 141–6, Leiden: Brill. Martone, C. (2012), “Qumran Readings in Agreement with the Septuagint against the Masoretic Text, Part Three: 1 Samuel,” RevQ 25: 557–73. Masius, A. (1574), Iosvæ imperatoris historia, F. Raphelengius (ed.), Antwerp: Plantin. Mason, S. (2002), “Josephus and His Twenty-Two Book Canon,” in L. M. McDonald and J. A. Sanders (eds.), The Canon Debate, 110–27, Peabody: Hendrickson. Matusova, E. (2018), “The Origins of Translation Theory: The LXX among Jewish Greek Writers,” in M. Meiser, M. Geiger, S. Kreuzer, and M. Sigismund (eds.), Die Septuaginta— Geschichte, Wirkung, Relevanz. 6. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 21.-24. Juli 2016, 557–72, WUNT 405, Tübingen: Mohr Siebeck. Maurais, J. (2020), “The Quest for LXX Deuteronomy’s Translator: On the Use of Translation Technique in Ascertaining the Translator’s Vorlage,” in E. Bons, M. Geiger, F. Ueberschaer, M. Sigismund, and M. Meiser (eds.), Die Septuaginta – Themen, Manuskripte, Wirkungen, 186–203, WUNT 444, Tübingen, Mohr Siebeck. Mayser, E. (1970), Grammatik der griechischen Papyri aus der Ptolemäerzeit: Band I Laut- und Wortlehre: I. Teil Einleitung und Lautlehre, Berlin: De Gruyter. Maximos, M., E. J. Pentiuc, M. Najim, J. N. Sparks, J. Allen, and T. Stylianopoulos, eds. (2008), The Orthodox Study Bible, Nashville: Thomas Nelson.

444 REFERENCES

Mazzinghi, L. (2019), Wisdom, IECOT, Stuttgart: Kohlhammer. McClurg, A. H. (2011), “A Critical Edition of the Hexaplaric Fragments of Numbers 19–36,” Ph.D. dissertation, Southern Baptist Theological Seminary, Louisville. McDonald, L. M. (2017), The Formation of the Biblical Canon, Volume I: The Old Testament: Its Authority and Canonicity, London: Bloomsbury T&T Clark. McDonald, L. M. (2018), “The Reception of the Writings and Their Place in the Biblical Canon,” in D. F. Morgan (ed.), The Oxford University Press Handbook on the Writings, 397–413, Oxford: Oxford University Press. McKendrick, S. and O. A. O’Sullivan, eds. (2003), The Bible as Book: The Transmission of the Greek Text, London: The British Library & Oak Knoll Press. McLay, R. T. (1994), Translation Technique and Textual Studies in the Old Greek and Theodotion Versions of Daniel, Ph.D. dissertation, University of Durham. McLay, R. T. (1998), “Kaige and Septuagint Research,” Text 19: 127–39. McLay, R. T. (2003), The Use of the Septuagint in New Testament Research, Grand Rapids: Eerdmans. McLay, R. T. (2010), “Why Not a Theology of the Septuagint?” in M. Karrer and W. Kraus (eds.), Die Septuaginta—Texte, Theologien, Einflüsse 2. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 23.-27. Juli 2008, 607–20, WUNT 252, Tübingen: Mohr Siebeck. Meade, J. D. (2012), “A Critical Edition of the Hexaplaric Fragments of Job 22–42,” Ph.D. dissertation, 2 volumes, Southern Baptist Theological Seminary, Louisville. Meade, J. D. (2020), A Critical Edition of the Hexaplaric Fragments of Job 22–42, Leuven: Peeters. Meadowcroft, T. J. (1995), Aramaic Daniel and Greek Daniel: A Literary Comparison, JSOTSup 198, Sheffield: Sheffield Academic Press. Meiser, M. (2018), “Present Perspectives of Septuagint Research,” Testimonia Theologica 12: 21–39. Mélèze-Modrzejewski, J. (1996), “Jewish Law and Hellenistic Legal Practice in the Light of Greek Papyri from Egypt,” in N. S. Hecht, B. S. Jackson, S. M. Passamaneck, D. Piattelli, and A. Rabello (eds.), An Introduction to the History and Sources of Jewish Law, 75–100, Oxford: Oxford University Press. Mélèze-Modrzejewski, J. (2008), La Bible d’Alexandrie. XV.3. Troisième Livre des Maccabées, BdA 15.3, Paris: Cerf. Menge, H., A. Thierfelder, and J. Wiesner (1999), Repetitorium der griechischen Syntax, 10th edn., Darmstadt: Wissenschaftliche Buchgesellschaft. Menken, M. J. J. (1996), Old Testament Quotations in the Fourth Gospel: Studies in Textual Form, Kampen: Kok Pharos. Menken, M. J. J. (2004), Matthew’s Bible. The Old Testament Text of the Evangelist, Leuven: Leuven University Press/Peeters. Menzies, G. W. (1996), Joseph’s Bible Notes (Hypomnestikon): Introduction, Translation, and Notes, SBLTT 41.9, Atlanta: Scholars Press. Mercati, G. (1895–1896), “D’un palimpsesto Ambrosiano contenente i Salmi esapli,” Atti della Reale Accademia delle scienze di Torino 31: 663–7. Mercati, G. (1916), “Chi sia l’autore della nuova versione dall’ebraico del codice veneto greco VII,” RB 13: 510–26. Mercati, G. (1941), Nuove Note di letteratura biblica e cristiana antica, Studi e Testi 95, Città del Vaticano: Biblioteca Apostolica Vaticana. Mercati, G., ed. (1958), Psalterii Hexapli Reliquiae cura et studio Iohannis Card, Vatican City: Bybliotheca Vaticana.

REFERENCES 445

Metso, S. and E. Ulrich (2003), “The Old Greek Translation of Leviticus,” in R. Rendtorff and R. A. Kugler (eds.), The Book of Leviticus: Composition and Reception, 247–68, VTSup 93, Leiden: Brill. Metzger, B. M. (1991), Manuscripts of the Greek Bible: An Introduction to Greek Palaeography, 2nd edn., Oxford: Oxford University Press. Meurer, S. and P. Ellingworth, eds. (1992), The Apocrypha in Ecumenical Perspective: The Place of the Late Writings of the Old Testament among the Biblical Writings and Their Significance in the Eastern and Western Church Traditions, Reading: United Bible Societies. Meyer, R. (1971), “Σαδδουκαῖος,” in G. Friedrich (ed.), Theological Dictionary of the New Testament, Volume 7: 35–54, Grand Rapids: Eerdmans. Middeldorpf, H., ed. (1835), Codex Syriaco-Hexaplaris: liber quartus Regum e codice parisiensi, Jesaias, duodecim Prophetae minores, Proverbia, Jobus, Canticum, Threni, Ecclesiastes e codice mediolanensi, Berlin: Enslin. Mihăilă, A. (2018), “The Septuagint and the Masoretic Text in the Orthodox Church(es),” Review of Ecumenical Studies Sibiu 10: 30–60. Millard, A. (1970), “‘Scriptio Continua’ in Early Hebrew: Ancient Practice or Modern Surmise?” JSS 15: 2–15. Miller, J. C. (2010), “The Prophetologion: the Old Testament of Byzantine Christianity?” in P. Magdalino and R. Nelson (eds.), The Old Testament in Byzantium, 55–76, Washington, Dumbarton Oaks Research Library and Collection. Mills, W. E. (1992), “Versions, Ancient (Coptic),” in D. N. Freedman (ed.), The Anchor Bible Dictionary, Volume 6: 803, New York: Doubleday. Miscall, P. D. (1992), “Isaiah: New Heavens, New Earth, New Book,” in D. Nolan Fewell (ed.), Reading Between Texts: Intertextuality and the Hebrew Bible, 41–56, Louisville: John Knox Press. Moatti-Fine, J. (1996), La Bible d’Alexandrie. VI. Jésus (Josué), BdA 6, Paris: Cerf. Moffitt, D. M. (2013), “A Note on Some Προσήλυτοι in P.Duk.inv. 727R,” in M. K. H. Peters (ed.), XIV Congress of the International Organization for Septuagint and Cognate Studies, Helsinki, 2010, SBLSCS 59, 357–60, Atlanta: SBL Press. Montanari, F. (1995), GI: Vocabolario della lingua greca, Turin: Loescher. Montanari, F. (2015), GE: The Brill Dictionary of Ancient Greek, Leiden: Brill. De Moor, J. C., ed. (1998), Intertextuality in Ugarit and Israel: Papers Read at the Tenth Joint Meeting of The Society for Old Testament Study and Het Oudtestamentisch Werkgezelschap in Nederland en België, Held at Oxford, 1997, OtSt 40, Leiden: Brill. Moore, G. F. (1895), A Critical and Exegetical Commentary on Judges, ICC, New York: Scribner. Moore, S. (2015), Jewish Ethnic Identity and Relations in Hellenistic Egypt: With Walls of Iron? JSJSup 171, Leiden: Brill. Morrish, G. (1887), A Handy Concordance of the Septuagint, London: Bagster and Sons. Mortari, L., ed. (1999), La Bibbia dei LXX. 1. Il Pentateuco: Testo greco con traduzione italiana a fronte, Roma: Edizioni Dehoniane. Moscati, S., ed. (1980), An Introduction to the Comparative Grammar of the Semitic Languages: Phonology and Morphology, PLO 6, Wiesbaden: Harrassowitz. Mosegaard Hansen, M.-B. (1998), The Function of Discourse Particles: A Study with Special Reference to Spoken Standard French, Pragmatics & Beyond New Series 53, Amsterdam: John Benjamins. Mosegaard Hansen, M.-B. (2006), “A Dynamic Polysemy Approach to the Lexical Semantics of Discourse Markers (with an Exemplary Analysis of French toujours),” in K. Fischer (ed.), Approaches to Discourse Particles, 21–41, Amsterdam: Elsevier.

446 REFERENCES

Moshavi, A. (2010), Word Order in the Biblical Hebrew Finite Clause, Linguistic Studies in Ancient West Semitic 4, Winona Lake: Eisenbrauns. Moulton, J. H. and G. Milligan (1930), The Vocabulary of the New Testament, London: Hodder & Stoughton. Moyer, I. S. (2011), Egypt and the Limits of Hellenism, Cambridge: Cambridge University Press. Moyise, S. (1995), The Old Testament in the Book of Revelation, LNTS 115, Sheffield: Sheffield Academic Press. Moyise, S. and M. J. J. Menken, eds. (2004), The Psalms in the New Testament, London: Bloomsbury T&T Clark. Mroczek, E. (2016), The Literary Imagination in Jewish Antiquity, Oxford: Oxford University Press. Müller, M. (1996), The First Bible of the Church: A Plea for the Septuagint, Sheffield: Sheffield Academic Press. Mulroney, J. A. E. (2016), The Translation Style of Old Greek Habakkuk, FAT 2.86, Tübingen: Mohr Siebeck. Munday, J. (2008), Introducing Translation Studies: Theories and Applications, 2nd edn., London: Routledge. Munnich, O. (1995), “Les Hexaples d’Origène à la lumière de la tradition manuscrite de la Bible grecque,” in G. Dorival and A. Le Boulluec (eds.), Origeniana Sexta: Origène et la Bible; actes du Colloquium Origenianum Sextum Chantilly, 30 août—3 septembre 1993, 167–85, BETL 118, Leuven: Peeters. Muraoka, T. (1984), A Greek-Hebrew/Aramaic Index to I Esdras, Chico: Scholars Press. Muraoka, T., ed. (1990), Melbourne Symposium on Septuagint Lexicography, SBLSCS 28, Georgia: Scholars. Muraoka, T. (1993), A Greek English of the Septuagint (Twelve Prophets), Leuven: Peeters. Muraoka, T. (1998), “Hebrew/Aramaic Index to the Septuagint,” in E. Hatch and H. A. Redpath (eds.), A Concordance to the Septuagint and the Other Greek Versions of the Old testament (Including the Apocryphal Books), 217–368, 2nd edn., Grand Rapids: Baker Academic. Muraoka, T. (2002), A Greek-English Lexicon of the Septuagint: Chiefly of the Pentateuch and the Twelve Prophets, Leuven: Peeters. Muraoka, T. (2008), “Recent Discussions on the Septuagint Lexicography with Special Reference to the So-called Interlinear Model,” in M. Karrer and W. Kraus (eds.), Die Septuaginta—Texte, Kontexte, Lebenswelten: Internationale Fachtagung Veranstaltet Von Septuaginta Deutsch (LXX.D), Wuppertal 20.-23. Juli 2006, 221–35, WUNT 219, Tübingen: Mohr Siebeck. Muraoka, T. (2010a), A Greek Hebrew/Aramaic Two-way Index to the Septuagint, Leuven: Peeters. Muraoka, T. (2010b), A Greek-English Lexicon of the Septuagint, Leuven: Peeters. Muraoka, T. (2013), “A Methodological Issue in Septuagint Studies with Special Reference to Lexicography,” JSCS 46: 101–8. Muraoka, T. (2015), “With the Publication of NETS,” in J. K. Aitken and T. V. Evans (eds.), Biblical Greek in Context: Essays in Honour of John A. L. Lee, 145–63, BiTS 22, Leuven: Peeters. Muraoka, T. (2016), A Syntax of Septuagint Greek, Leuven: Peeters. Muraoka, T. and B. Porten (1998), A Grammar of Egyptian Aramaic, HO 32, Leiden: Brill.

REFERENCES 447

Murtonen, A. (1981–2), “Methodological Preliminaries to a Study of Greek (and Latin) Transcriptions of Hebrew,” Abr-Nahrain 20: 60–73. Mussies, G. (1976), “Greek in Palestine and the Diaspora,” in S. Saafrai and M. Stern (eds.), The Jewish People in the First Century, Volume 2: 1040–64, CRINT 1, Leiden: Brill. Myers, P. D. (2018), “The Textual Affinities of Sinaiticus’ Correctors in 2 Esdras: An Analysis of Proper Nouns,” BASP 55: 157–93. Myers, P. D. (2019), “A Grammar of Transcriptions in 2 Esdras: Text, Philology, Phonology,” Ph.D. dissertation, University of Cambridge, Cambridge. Najman, H. (2003), Seconding Sinai: The Development of Mosaic Discourse in Second Temple Judaism, JSJSup 77, Leiden/Boston: Brill. Najman, H. and B. G. Wright (2017), “Perfecting Translation: The Greek Scriptures in Philo of Alexandria,” in J. Baden, H. Najman, and E. Tigchelaar (eds.), Sibyls, Scriptures, and Scrolls: John Collins at Seventy, 897–915, JSJSup 175/1–2, Leiden/Boston: Brill. Nautin, P. (1977), Origène: Sa vie et son oeuvre, Paris: Beauchesne. Nautin, P. (1985), “Hieronymus,” TRE 15: 304–16. Nestle, E. (1903), “Septuagint,” in J. Hastings (ed.), A Dictionary of the Bible, Volume 4: 440–1, New York: Charles Scribner’s Sons. Neuschäfer, B. (1983), Origenes als Philologe, Schweizerische Beiträge aur Altertumswissenschaft, Heft 18/1-2, Basel: Friedrich Reinhardt Verlag. New, D. S. (1993), Old Testament Quotations in the Synoptic Gospels, and the Two-Document Hypothesis, SBLSCS 37, Atlanta: Scholars Press. Newman, H. (2009), “How Should We Measure Jerome’s Hebrew Competence,” in A. Cain and J. Lössl (eds.), Jerome of Stridon: His Life, Writings and Legacy, 131–40, Burlington, VT: Ashgate. Ngunga, A. T. (2012), Messianism in the Old Greek of Isaiah: An Intertextual Analysis. FRLANT 245, Göttingen: Vandenhoeck & Ruprecht. Nicholson, E. (1998), The Pentateuch in the Twentieth Century. The Legacy of Julius Wellhausen, Oxford: Oxford University Press. Nicklas, T. (2013), “The Bible and Anti-Semitism,” in M. Lieb, E. Mason, and J. Roberts (eds.), The Oxford Handbook of the Reception History of the Bible, 267–80, Oxford: Oxford University Press. Niehoff, M. R. (2011), Jewish Exegesis and Homeric Scholarship in Alexandria, Cambridge: Cambridge University Press. Nöldeke, T. (1904), Compendious Syriac Grammar, J. A. Crichton (trans.), London: Williams & Norgate. Nogalski, J. D. (1996), “Intertextuality and the Twelve,” in J. W. Watts and P. R. Houe (eds.), Forming Prophetic Literature: Essays on Isaiah and the Twelve in Honor of John D. W. Watts, 102–24, JSOTSup 235, Sheffield: Sheffield Academic Press. Nongbri, B. (2018), God’s Library: The Archaeology of the Earliest Christian Manuscripts, New Haven: Yale University Press. Norberg, M. (1787), Codex Syriaco-Hexaplaris Ambrosiano-Mediolanensis, Lund: Berling. Norton, G. J., O. P. with Carmen Hardin (2005), Frederick Field’s Prolegomena to Origenis Hexaplorum quae Supersunt sive veterum interpretum Graecorum in totum Vetus Testamentum fragmenta, Cahiers de la Revue Biblique 62, Paris: Gabalda. Norton, J. D. H. (2011), Contours in the Text: Textual Variation in the Writings of Paul, Joseph and the Yahad, LNTS 430, London: T&T Clark. Noth, M. (1966), Die israelitischen Personennamen im Rahmen der gemeinsemitischen Namengebung, Hildesheim: Georg Olms.

448 REFERENCES

O’Callaghan, J. (1972), “¿Papiros neotestamentarios en la cueva 7 de Qumrān?” Bib 53: 91–100. O’Connell, S. (2006), From Most Ancient Sources: The Nature and Text-Critical Use of the Greek Old Testament Text of the Complutensian Polyglot Bible, Fribourg: Academic Press. Olley, J. W. (2009), Ezekiel, SEPT, Leiden: Brill. Olofsson, S. (1990), God Is My Rock: A Study of Translation Technique and Theological Exegesis in the Septuagint, ConBOT 31, Stockholm: Almqvist & Wiksell. Olofsson, S. (2001), “Death Shall Be Their Shepherd: An Interpretation of Ps 49:15 in LXX,” in R. Sollamo and S. Sipilä (eds.), Helsinki Perspective on the Translation Technique of the Septuagint, 139–66, PFES 82, Göttingen: Vandenhoeck & Ruprecht. Olszowy-Schlanger, J. (2003), “An Early Hebrew Manuscript from Byzantium,” in S. Berger, M. Brocke, and I. Zwiep (eds.), Zutot 2002, 148–55, London/Dordrecht: Kluwer. Orlinsky, H. (1959), “Studies in the Septuagint of the Book of Job. Chapter III: On the Matter of Anthropomorphisms, Anthropopathisms, and Euphemisms,” HUCA 30: 153–67. Orlinsky, H. (1962), “Studies in the Septuagint of the Book of Job,” HUCA 33: 119–51. Ottley, R. R. (1920), A Handbook to the Septuagint, London: Methuen. Owen, H. (1769), An Enquiry into the Present State of the Septuagint Version of the Old Testament, London: B. White. Palmer, J. K. (2004), “‘Not Made with Tracing Paper’: Studies in the Septuagint of Zechariah,” Ph.D. dissertation, University of Cambridge, Cambridge. Palmer, L. R. (1945), A Grammar of the Post-Ptolemaic Papyri, Volume 1: Accidence and Word-Formation, Part 1: The Suffixes, Publications of the Philological Society 13, London: Oxford University Press. Paltridge, B. (2012), Discourse Analysis: An Introduction, London: Bloomsbury. Panayotov, A. (2014), “Jews and Jewish Communities in the Balkans and the Aegean until the Twelfth Century,” in J. K. Aitken and J. C. Paget (eds.), The Jewish-Greek Tradition in Antiquity and the Byzantine Empire, 54–76, Cambridge University Press: Cambridge. Parsons, P. J. (1990), “The Scripts and Their Date,” in E. Tov (ed.), The Greek Minor Prophets Scroll from Naḥal Ḥever (8ḤevXIIgr) (The Seiyâl Collection 1), 19–26, DJD 8, Oxford: Clarendon Press. Parsons, P. J. (1992), “The Palaeography and Date of the Greek Manuscripts,” in P. W. Skehan, E. Ulrich, and J. E. Sanderson (eds.), Qumran Cave 4 IV, Palaeo-Hebrew and Greek Biblical Manuscripts, 7–13, DJD 9, Oxford: Clarendon Press. Pelletier, A. (1962), Flavius Josèphe, adaptateur de la Lettre d’Aristée: une réaction atticisante contre la Koiné, Études et Commentaires 45, Paris: Klincksieck. Penner, K. Isaiah, SEPT, Leiden: Brill. Penner, K. M., R. Brannan, M. Aubrey, I. Hoogendyk, and I. Loken, eds. (2019), The Lexham English Septuagint, Bellingham: Lexham Press. Pentiuc, E. J. (2014), The Old Testament in Eastern Orthodox Tradition, Oxford: Oxford University Press. Perkins, L. (2007), “‘The Lord Is a Warrior’—‘The Lord Who Shatters Wars’: Exod 15:3 and Jdt 9:7; 16:2,” BIOSCS 40: 121–38. Perkins, L. (2009), “Greek Exodus and Greek Isaiah: Detection and Implications of Interdependence in Translation,” BIOSCS 42: 18–33. Peters, M. K. H. (2010), “Translating a Translation: Some Final Reflections on the Production of the New English Translation of Greek Deuteronomy,” in R. J. V. Hiebert (ed.), “Translation Is Required”: The Septuagint in Retrospect and Prospect, 119–34, SBLSCS 56, Atlanta: SBL Press.

REFERENCES 449

Petersen, N. (2009), “An Analysis of Two Early LXX Manuscripts from Qumran: 4QLXXNum and 4QLXXLeva in the Light of Previous Studies,” BBR 19: 481–510. Pfeiffer, S. (2016), “Ägyptische Elemente im Griechischen der LXX,” in E. Bons and J. Joosten (eds.), Die Sprache der Septuaginta/The Language of the Septuagint, 231–45, LXX.H 3, Gütersloh: Gütersloher Verlagshaus. Pietersma, A. (1984), “Kyrios or Tetragram: A Renewed Quest for the Original Septuagint,” in A. Pietersma and C. Cox (eds.), De Septuaginta: Studies in Honour of John William Wevers on His Sixty-fifth Birthday, 85–101, Mississauga: Benben. Pietersma, A. (1990), “Ra 2110 (P. Bodmer XXIV) and the Text of the Greek Psalter,” in D. Fraenkel, U. Quast, and J. W. Wevers (eds.), Studien zur Septuaginta—Robert Hanhart zu Ehren: Aus Anlass seines 65. Geburtstages, 262–86, MSU 20, Göttingen: Vandenhoeck & Ruprecht. Pietersma, A. (1996), Translation Manual for “A New English Translation of the Septuagint” (NETS), Ada: Uncial Books. Pietersma, A. (1997), “A New English Translation of the Septuagint,” in B. A. Taylor (ed.), IX Congress of the International Organization for Septuagint and Cognate Studies, Cambridge, 1995, 177–87, SBLSCS 45, Atlanta: Scholars Press. Pietersma, A. (1998), “A Prospectus for a Commentary on the Septuagint, Sponsored by the International Organization for Septuagint and Cognate Studies,” BIOSCS 31: 43–8. Pietersma, A. (2001), “A New English Translation of the Septuagint,” in B. A. Taylor (ed.), X Congress of the International Organization for Septuagint and Cognate Studies, Oslo, 1998, 217–28, SBLSCS 51, Atlanta: SBL Press. Pietersma, A. (2002), “A New Paradigm for Addressing Old Questions: The Relevance of the Interlinear Model for the Study of the Septuagint,” in J. Cook (ed.), Bible and Computer. The Stellenbosch AIBI-6 Conference. Proceedings of the Association Internationale Bible et Informatique “From Alpha to Byte.” University of Stellenbosch 17–21 July, 2000, 337–64, Leiden/Boston: Brill. Pietersma, A. (2006a), “Exegesis in the Septuagint: Possibilities and Limitations (The Psalter as a Case in Point),” in W. Kraus and R. Glenn Wooden (eds.), Septuagint Research: Issues and Challenges in the Study of the Greek Jewish Scriptures, 33–46, SBLSCS 53, Atlanta: SBL Press. Pietersma, A. (2006b), “Greek Jeremiah and the Land of Azazel,” in P. W. Flint, E. Tov, and J. C. VanderKam (eds.), Studies in the Hebrew Bible, Qumran, and the Septuagint Presented to Eugene Ulrich, 402–13, VTSup 101, Leiden: Brill. Pietersma, A. (2006c), “LXX and DTS: A New Archimedean Point for Septuagint Studies?” BIOSCS 39: 1–11. Pietersma, A. (2006d), “Messianism and the Greek Psalter: In Search of the Messiah,” in M. A. Knibb (ed.), The Septuagint and Messianism, 49–75, BETL 195, Leuven: Peeters. Pietersma, A. (2007), “To the Reader of Psalms,” in A. Pietersma and B. G. Wright (eds.), A New English Translation of the Septuagint, xiii–xx, New York: Oxford University Press. Pietersma, A. (2008), “Text-Production and Text-Reception: Psalm 8 in Greek,” in M. Karrer and W. Kraus (eds.), Die Septuaginta—Texte, Kontexte, Lebenswelten: Internationale Fachtagung Veranstaltet Von Septuaginta Deutsch (LXX.D), Wuppertal 20.-23. Juli 2006, 487–501, WUNT 219, Tübingen: Mohr Siebeck. Pietersma, A. (2010), “Beyond Literalism: Interlinearity Revisited,” in R. J. V. Hiebert (ed.), “Translation Is Required”: The Septuagint in Retrospect and Prospect, 3–22, SBLSCS 56, Atlanta: SBL Press.

450 REFERENCES

Pietersma, A. (2015), “Context Is King in Septuagint Lexicography—or Is It?” in J. K. Aitken and T. V. Evans (eds.), Biblical Greek in Context: Essays in Honour of John A. L. Lee, 165–75, BiTS 22, Leuven: Peeters. Pietersma, A. (2017), “The Society of Biblical Literature Commentary on the Septuagint: Basic Principles,” in D. Büchner (ed.), The SBL Commentary on the Septuagint: An Introduction, 1–16, SBLSCS 67, Atlanta: SBL. Pietersma, A. (2019), “The IOSCS at Fifty,” JSCS 52: 5–18. Pietersma, A. and B. G. Wright, eds. (2007), A New English Translation of the Septuagint and the Other Greek Translations Traditionally Included under That Title, New York: Oxford University Press. Pietersma, A. and B. G. Wright, eds. (2009), A New English Translation of the Septuagint and the Other Greek Translations Traditionally Included under That Title, 2nd edn., New York: Oxford University Press. Polaski, D. C. (2001), Authorizing an End: The Isaiah Apocalypse and Intertextuality, BibInt 50, Leiden: Brill. Poleg, E. (2013), “The Interpretation of Hebrew Names in Theory and Practice,” in E. Poleg and L. Light (eds.), Form and Function in the Late Medieval Bible, 217–36, Library of the Written Word: The Manuscript World 4, Leiden: Brill. Popowski, R. (2013), Septuaginta czyli grecka biblia starego testamentu: Wraz z księgami deuterokanonicznymi i apokryfami, Warsaw: Vocatio. Porter, S. E. (2011), “The Language(s) Jesus Spoke,” in T. Holmén and S. E. Porter (eds.), Handbook for the Study of the Historical Jesus, Volume 3: 2455–71, Leiden: Brill. Porter, S. E. (2013a), “Commentaries on the Book of Romans,” in S. E. Porter and E. J. Schnabel (eds.), On the Writing of New Testament Commentaries: Festschrift for Grant R. Osborne on the Occasion of His 70th Birthday, 365–404, TENT 8, Leiden: Brill. Porter, S. E. (2013b), How We Got the New Testament: Text, Transmission, Translation, Grand Rapids: Baker. Porter, S. E. (2013c), “The Linguistic Competence of New Testament Commentaries,” in S. E. Porter and E. J. Schnabel (eds.), On the Writing of New Testament Commentaries: Festschrift for Grant R. Osborne on the Occasion of His 70th Birthday, 33–56, TENT 8, Leiden: Brill. Porter, S. E. (2013d), “What Do We Know and How Do We Know It? Reconstructing Early Christianity from Its Manuscripts,” in S. E. Porter and A. W. Pitts (eds.), Christian Origins and Greco-Roman Culture: Social and Literary Contexts for the New Testament, 41–70, ECHC 1, TENT 9, Leiden: Brill. Porter, S. E. (2014), “Septuagint,” in G. K. Giannakis (ed.), Encyclopedia of Ancient Greek Language and Linguistics, 3: 287–90, Leiden: Brill. Porter, S. E. (2015), Constantine Tischendorf, London: Bloomsbury. Porter, S. E. (2016a), “Big Enough Is Big Enough: A Review Article,” TrinJ n.s. 37: 31–40. Porter, S. E. (2016b), “History of Scholarship on the Language of the Septuagint,” in E. Bons and J. Joosten (eds.), Die Sprache der Septuaginta/The Language of the Septuagint, 15–38, LXX.H 3, Gütersloh: Gütersloher Verlagshaus. Porter, S. E. (2018), “Composite Citations in Luke-Acts,” in S. A. Adams and S. Ehorn (eds.), Composite Citation in Antiquity, Volume 2: New Testament Uses, 62–93, LNTS 593, London: Bloomsbury T&T Clark. Porter, S. E. and B. W. R. Pearson (1997), “Isaiah through Greek Eyes: The Septuagint of Isaiah,” in C. C. Broyles and C. A. Evans (eds.), Writing and Reading the Scroll of Isaiah: Studies of an Interpretive Tradition, Volume 2: 531–46, VTSup 70, Leiden: Brill.

REFERENCES 451

Porter, S. E. and J. C. Robinson (2011), Hermeneutics: An Introduction to Interpretive Theory, Grand Rapids: Eerdmans. Porter, S. E. and B. M. Stovell (2012), “Introduction: Trajectories in Biblical Hermeneutics,” in S. E. Porter and B. M. Stovell (eds.), Biblical Hermeneutics: Five Views, 9–26, Downers Grove: InterVarsity. Pouchelle, P. (2019), “Document(s) et épître(s): Nommer ‘la lettre’ dans le Texte Massorétique et la Septante,” EstBib 77: 65–82. Preisigke, F. (1915), Fachwörterbuch des öffentlichen Verwaltungsdienstes Ägyptens in den griechischen Papyrusurkunden der ptolemäisch-römischen Zeit, Göttingen: Vandenhoeck & Ruprecht. Prestel, P. (2016), “Die Diversität des Griechischen in der Septuaginta,” in E. Bons and J. Joosten (eds.), Die Sprache der Septuaginta/The Language of the Septuagint, 39–68, LXX.H 3, Gütersloh: Gütersloher Verlagshaus. Prijs, L. (1948), Jüdische Tradition in der Septuaginta, Leiden: Brill. Pritchard, J. B., ed. (1969), Ancient Near Eastern Texts Relating to the Old Testament, 3rd edn., Princeton: Princeton University Press. Provan, I. (2017), The Reformation and the Right Reading of Scripture, Waco: Baylor University Press. Pym, A. (1998), Method in Translation History, Manchester: St. Jerome Publishing. Quast, U. (1990), “Der rezensionelle Charakter einiger Wortvarianten im Buche Numeri,” in D. Fraenkel, U. Quast, and J. W. Wevers (eds.), Studien zur Septuaginta—Robert Hanhart zu Ehren: Aus Anlass seines 65. Geburtstages, 230–252, MSU 20, Göttingen: Vandenhoeck & Ruprecht. Quast, U., ed. (2009), Septuaginta. Vetus Testamentum Graecum. IV.3. Ruth, 2nd edn., Göttingen: Vandenhoeck & Ruprecht. Qimron, E. (1986a), The Hebrew of the Dead Sea Scrolls, Harvard Semitic Studies 29, Atlanta: Scholars Press. Qimron, E. (1986b), “‫ּפתּח ּבעברית המקראית‬/‫חילּופי צירי‬,” Lĕš 50: 77–102. Qimron, E. (1991), “‫פתח ו׳׳חוק פיליפי׳׳‬/‫חילופי צירי‬,” Lĕš 56 111–15. Qimron, E. (2006), “The Pausal Patah. in Biblical Hebrew,” in S. E. Fassberg and A. Hurvitz (eds.), Biblical Hebrew in Its Northwest Semitic Setting: Typological and Historical Perspectives, 305–14, Publication of the Institute for Advanced Studies 1, Jerusalem/Winona Lake: Magnes/Eisenbrauns. Rabinowitz, I. (1956), “Aramaic Inscriptions of the Fifth Century B.C.E. from a North-Arab Shrine in Egypt,” JNES 15: 1–9. Rahlfs, A. (1914), Verzeichnis der griechischen Handschriften des Alten Testaments, für das Septuaginta-Unternehmen aufgestellt, Nachrichten der Königlichen Gesellschaft der Wissenschaften zu Göttingen. Philologisch-historische Klasse, MSU 2, Berlin: Weidmann. Rahlfs, A., ed. (1931), Psalmi cum Odis, Göttingen: Vandenhoeck & Ruprecht. Rahlfs, A., ed. (1935), Septuaginta id est vetus testamentum graece iuxta LXX interpretes, 2 volumes, Stuttgart: Deutsche Bibelgesellschaft. Rahlfs, A., ed. (1979), Septuaginta. Vetus Testamentum Graecum. X Psalmi cum odis, 3rd edn., Göttingen: Vandenhoeck & Ruprecht. Rahlfs, A. (2012), Offizielles Verzeichnis der Rahlfs-Sigeln: Herausgegeben vom SeptuagintaUnternehmen der Akademie der Wissenschaften zu Göttingen, Stand: December 2012, Göttingen: Lagarde-Haus. Rahlfs, A. and D. Fraenkel, eds. (2004), Verzeichnis der griechischen Handschriften des Alten Testaments, SVTGSup 1.1: Die Überlieferung bis zum VIII, Jahrhundert, Göttingen: Vandenhoeck & Ruprecht.

452 REFERENCES

Rahlfs, A., and R. Hanhart, eds. (2006), Septuaginta: Id est Vetus Testamentum graece iuxta LXX interpretes edidit Alfred Rahlfs. Editio altera quam recognovit et emendavit Robert Hanhart, Stuttgart: Deutsche Bibelgesellschaft. Rajak, T. (2009), Translation & Survival: The Greek Bible of the Ancient Jewish Diaspora, Oxford: Oxford University Press. Rechenmacher, H. (2012), Althebräische Personennamen, Lehrbücher orientalischer Sprachen/ Textbooks of Near Eastern Languages II/1, Münster: Ugarit-Verlag. de Regt, L. J. (2016), “Canon and Biblical Text in the Slavonic Tradition in Russia,” BT 67: 223–39. Rehkopf, F. (1989), Septuaginta-Vokabular, Göttingen: Vandenhoeck & Ruprecht. Reider, J. (1916), Prolegomena to a Greek-Hebrew & Hebrew-Greek Index to Aquila, Oxford: University Press. Reider, J. (1966), An Index to Aquila: Greek-Hebrew, Hebrew-Greek, Latin Hebrew: With the Syriac and Armenian Evidence, compiled and revised by N. Turner, VTSupp 12, Leiden: Brill. Reif, S. C. (2000), A Jewish Archive from Old Cairo: The History of Cambridge University’s Genizah Collection, Richmond, Surrey: Curzon. Reilly, W. (1911), “Polyglot Bibles,” in The Catholic Encyclopedia, n.p., New York: Robert Appleton Company, Online: http://www.newadvent.org/cathen/12222a.htm (accessed October 21, 2019). Rendsburg, G. A. (2013), “Phonology: Biblical Hebrew,” in G. Khan (ed.), Encyclopedia of Hebrew Language and Linguistics, Volume 3: 100b–9b, Leiden: Brill. Renkema, J. (2004), Introduction to Discourse Studies, Amsterdam/Philadelphia: John Benjamins Publishing Company. Rex, R. (1992), “St John Fisher’s Treatise on the Authority of the Septuagint,” JTS 43: 56–116. Richey, M. (2012), “The Use of Greek at Qumran: Manuscript and Epigraphic Evidence for a Marginalized Language,” DSD 19: 177–97. Robar, E. (2016), “The Historical Present in NT Greek: An Exercise in Interpreting Matthew,” in S. E. Runge and C. J. Fresch (eds.), The Greek Verb Revisited: A Fresh Approach for Biblical Exegesis, 329–52, Bellingham: Lexham Press. Rösel, M. (1994), Übersetzung als Vollendung der Auslegung: Studien zur Genesis-Septuaginta, BZAW 223, Berlin: De Gruyter. Rösel, M. (2018a), “A Theology of the Septuagint? Clarifications and Definitions,” in M. Rösel (ed.), Tradition and Innovation: English and German Studies on the Septuagint, 273–90 SBLSCS 70, Atlanta: SBL Press. Rösel, M. (2018b), “Towards a ‘Theology of the Septuagint’,” in M. Rösel (ed.), Tradition and Innovation: English and German Studies on the Septuagint, 253–72, SBLSCS 70, Atlanta: SBL Press. Previously published in 2006 in W. Kraus and R. G. Wooden (eds.), Septuagint Research: Issues and Challenges in the Study of the Greek Jewish Scriptures, 239–52, SBLSCS 53, Atlanta: SBL Press. Rösel, M. (2018c), “Projekt: Theologie der Septuaginta-Theology of the Septuagint I. Forschungsgeschichte,” Online: https://www.academia.edu/41203456/ Martin_R%C3%B6sel_Projekt_Theologie_der_Septuaginta-Theology_of_the_Septuagint_I._ Forschungsgeschichte (accessed March 25, 2020). Rösel, M. (2018d), “Projekt: Theologie der Septuaginta-Theology of the Septuagint Teil II. Methodische Grundlegung,” Online: https://www.academia.edu/41203457/ Martin_R%C3%B6sel_Projekt_Theologie_der_Septuaginta-Theology_of_the_Septuagint_ Teil_II._Methodische_Grundlegung (accessed March 25, 2020).

REFERENCES 453

Rösel, M. and C. Schlund (2011), “Arithmoi/Numeri/Das Vierte Buch Mose,” in M. Karrer and W. Kraus (eds.), Septuaginta Deutsch. Erläuterungen und Kommentare zum griechischen Alten Testament, Volume 1: Genesis bis Makkabäer, 431–522, Stuttgart: Deutsche Bibelgesellschaft. Romeny, R.B. ter H. (1997), A Syrian in Greek Dress: The Use of Greek, Hebrew, and Syriac Biblical Texts in Eusebius of Emesa’s Commentary on Genesis, TEG 6, Leuven: Peeters. Ros, J. S. J. (1940), De Studie van het Bijbelgrieksch van Hugo Grotius tot Adolf Deissmann, Nijmegen: Dekker & van de Vegt. Rosenthal, F. (2006), A Grammar of Biblical Aramaic, 7th edn., PLO 5, Wiesbaden: Harrassowitz. Ross, W. A. (2017), “Style and Familiarity in Judges 19,7 (Old Greek): Establishing Dependence within the Septuagint,” Bib 98: 25–36. Ross, W. A. (2018a), “The Septuagint as a Catalyst for Language Change in the Koine: A Usage-Based Approach,” in M. Meiser, M. Geiger, S. Kreuzer, and M. Sigismund (eds.), Die Septuaginta—Geschichte, Wirkung, Relevanz. 6. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 21.-24. Juli 2016, 382–97, WUNT 405, Tübingen: Mohr Siebeck. Ross, W. A. (2018b), “Septuagint Lexicography and Language Change in Greek Judges,” Ph.D. dissertation, University of Cambridge, Cambridge. De’ Rossi, A. (2001), The Light of the Eyes, J. Weinberg (ed. and trans.), New Haven: Yale University Press. Roth-Gerson, L. (1987), The Greek Inscriptions from the Synagogues in Eretz-Israel, Jerusalem: Yad Ben Zvi [Hebrew]. Ruark, J. D. (2020), “‘To Shine’ or ‘To Show’? A Conceptual and Theological Exploration of Light in the LXX,” in J. Cook and M. Rösel (eds.), Toward a Theology of the Septuagint: Stellenbosch Congress on the Septuagint, 2018, 245–60, SBLSCS 74, Atlanta: SBL Press. Rüger, H. P. (1959), “Vier Aquila-Glossen in einem hebräischen Proverbien-Fragment aus der Kairo-Geniza,” ZNW 50: 275–7. Runge, S. E. (2010), Discourse Grammar of the Greek New Testament: A Practical Introduction for Teaching and Exegesis, Lexham Bible Reference Series, Peabody: Hendrickson. Rupprecht, H.-A. (1994), Kleine Einführung in die Papyruskunde, Darmstadt: Wissenschaftliche Buchgesellschaft. Rutgers, L. V. (2003), “Justinian’s Novella 146 between Jews and Christians,” in R. Kalmin and S. Schwartz (eds.), Jewish Culture and Society under the Christian Roman Empire, 385–407, Interdisciplinary Studies in Ancient Culture and Religion, Leuven: Peeters. Sáenz-Badillos, Á. (1990), La filología bíblica en los primeros helenistas de Alcalá, Estella: Verbo Divino. Sagarin, J. L. (1987), Hebrew Noun Patterns (Mishqalim): Morphology, Semantics, and Lexicon, Riga: Scholars Press. Saley, R. J. (2010), “Reconstructing 4QJerb According to the Text of the Old Greek,” DSD 17: 1–12. Salvesen, A. (1991), Symmachus in the Pentateuch, JSS Monographs 15, Manchester: University of Manchester. Salvesen, A. (2005), “Review of Adam Kamesar, Jerome, Greek Scholarship and the Hebrew Bible,” BIOSCS 38: 129–31 Salvesen, A. (2009), “The Relationship of LXX and the Three in Exodus 1–24 to the Readings of Fb,” in N. de Lange, J. Krivoruchko, and C. Boyd-Taylor (eds.), Jewish Reception of Greek Bible Versions, 103–27, TSMEMJ 23, Tübingen: Mohr Siebeck.

454 REFERENCES

Salvesen, A. (2015), “The ‘Three’ in Early Christian Commentary: The Case of the ‘Song of the Vineyard’ (Isaiah 5:1–7),” JSCS 48: 72–85. Salvesen, A. and T. M. Law, eds. (forthcoming), Oxford Handbook of the Septuagint, Oxford Handbooks, Oxford/New York: Oxford University Press. Sanders, J. A. (1965), The Psalms Scroll of Qumrân Cave 11 (11QPsa), DJD 4, Oxford: Clarendon Press. Dos Santos, E. C. (1973), An Expanded Hebrew Index for the Hatch-Redpath Concordance to the Septuagint, Jerusalem: Dugith Publishers. Sarna, N. M. (1989), The JPS Torah Commentary: Genesis, Philadelphia: The Jewish Publication Society. Satterthwaite, P. E. (2015), “Judges,” in J. K. Aitken (ed.), T&T Clark Companion to the Septuagint, 102–17, London: Bloomsbury T&T Clark. Scanlin, H. P. (1996), “The Old Testament Canon in the Orthodox Churches,” in B. Nassif (ed.), New Perspectives on Historical Theology: Essays in Memory of John Meyendorff, 300–12, Grand Rapids: Eerdmans. Scanlin, H. P. (2016), “Review of Nicholas King, translator, The Bible: A Study Bible freshly translated by Nicholas King. Buxhall, Stowmarket, Suffolk: Kevin Mayhew, 2019. xiii + 2395 pp., ISBN 978-1-84867-667-1 (hardback),” JSCS 49: 119–22. Scarlata, M. W. (2015), “Genesis,” in J. K. Aitken (ed.), The T&T Clark Companion to the Septuagint, 13–28, London: Bloomsbury T&T Clark. Schäfer, C. (2012a), Benutzerhandbuch zur Göttinger Septuaginta, Band 1: Die Edition des Pentateuch von John William Wevers, Göttingen: Vandenhoeck & Ruprecht. Schäfer, C. (2012b), Benutzerhandbuch zur Göttinger Septuaginta, Band 2: Die Edition des Buches Ruth von Udo Quast, Göttingen: Vandenhoeck & Ruprecht. Schäfer, C. (2016), Alfred Rahlfs (1865–1935) und die kritische Edition der Septuaginta. Eine biographisch-wissenschaftsgeschichtliche Studie, BZAW 489, Berlin: De Gruyter. Schaper, J. (1995), Eschatology in the Greek Psalter, WUNT 2/76, Tübingen: Mohr Siebeck. Schaper, J. (2006), “Messianism in the Septuagint of Isaiah and Messianic Intertextuality in the Greek Bible,” in M. A. Knibb (ed.), The Septuagint and Messianism, 371–80, BETL 195, Leuven: Peeters. Schaper, J. (2014), “The Concept of the Translator(s) in the Contemporary Study of the Septuagint,” in K. De Troyer, T. Michael Law, and M. Liljeström (eds.), In the Footsteps of Sherlock Holmes: Studies in the Biblical Text in Honour of Anneli Aejmelaeus, 31–46, CBET 72, Leuven: Peeters. Scheck, T. P., trans. (2010), St. Jerome’s Commentaries on Galatians, Titus, and Philemon, Notre Dame: University of Notre Dame Press. Schiffman, L. H. (1992), “The Septuagint and the Temple Scroll: Shared ‘halakhic’ Variants,” in G. J. Brooke and B. Lindars (eds.), Septuagint, Scrolls and Cognate Writings, 277–97, SBLSCS 33, Atlanta: Scholars Press. Schironi, F. (2012), “The Ambiguity of Signs: Critical σημεῖα from Zenodotus to Origen,” in M. R. Niehoff (ed.), Homer and the Bible in the Eyes of Ancient Interpreters, 87–112, JSRC 16, Leiden: Brill. Schironi, F. (2015), “P.Grenf. 1.5, Origen and the Scriptorium of Caesarea,” BASP 52: 181–223. Schironi, F. (2018), The Best of the Grammarians: Aristarchus of Samothrace on the Iliad, Ann Arbor: University of Michigan Press. Schleusner, J. F. (1820–21), Novus Thesaurus philologico-criticus, sive Lexicon in LXX et reliquos interpretes graecos ac scriptores apocryphos Veteris Testamenti, 5 volumes, Leipzig: Weidmann.

REFERENCES 455

Schmitt, A. (1974), “Interpretation der Genesis aus hellenistischem Geist,” ZAW 86: 137–65. Schürer, E. (1979), The History of the Jewish People in the Age of Jesus Christ (175 B.C.–A.D. 135), Volume 2, revised edn., Edinburgh: T&T Clark. Schultz, R. L. (1999), The Search for Quotation: Verbal Parallels in the Prophets, JSOTSup 180, Sheffield: Sheffield University Press. Schwyzer, E. (1938), Griechische Grammatik auf der Grundlage von Karl Brugmanns Griechischer Grammatik: Erster Band, Handbuch der Altertumswissenschaft II.1.1, München: C. H. Beck’sche. Schwyzer, E. (1953), Griechische Grammatik auf der Grundlage von Karl Brugmanns Griechischer Grammatik: Dritter Band: Register, Handbuch der Altertumswissenschaft II.1.3, München: C. H. Beck’sche. Scialabba, D. (2019), “Travel and Wisdom—An Interpretation of Sirach 31[34],9-12; 39,4; 51,13 against its Jewish and Hellenistic Background,” Bib 100: 373–85. Scouteris, C. and C. Belezos (2015), “The Bible in the Orthodox Church from the Seventeenth Century to the Present Day,” in J. Riches (ed.), The New Cambridge History of the Bible, Volume 4: From 1750 to the Present, 523–56. Cambridge: Cambridge University Press. Screnock, J. (2017), Traductor Scriptor: The Old Greek Translation of Exodus 1–14 as Scribal Activity, VTSup 174, Leiden: Brill. Screnock, J. (2018), “A New Approach to Using the Old Greek in Hebrew Bible Textual Criticism,” Text 27: 229–57. Seeligmann, I. L. (1948), The Septuagint Version of Isaiah: A Discussion of its Problems, MVEOL 9, Leiden: Brill. Seeligmann, I. L. (1990), “Problems and Perspectives in Modern Septuagint Research,” Text 15: 169–232. Seeligmann, I. L. ([1948] 2004), The Septuagint Version of Isaiah and Cognate Studies, FAT 40, Tübingen: Mohr Siebeck. Segal, M. H. (1927), A Grammar of Mishnaic Hebrew, Oxford: Clarendon Press. “Seleznev, M. G. (2008), “The Hebrew text of the Bible and the LXX: two base texts, two translations?” in Proceedings of the XVIII theological conference of the Saint Tihkon’s Orthodox University, Volume 1, 56–61, Moscow: Publishing House of the  the Saint Tihkon’s Orthodox University. Sgherri, G. (1977), “Sulla valutazione origeniana dei LXX,” Bib 58: 1–28. Shamir, I. A. (2009), “Translating the Bible into Hebrew,” Online: http://www.israelshamir.net/ English/Bible_to_Hebrew.htm (accessed May 16, 2019). Sharf, A. (1971), Byzantine Jewry from Justinian to the Fourth Crusade, London: Routledge. Shreve, G. M., C. Schäffner, J. H. Danks, and J. Griffin (1993), “Is There a Special Kind of ‘Reading’ for Translation?: An Empirical Investigation of Reading in the Translation Process,” Target 5: 21–41. Sieg, U. (2007), Deutschlands Prophet: Paul de Lagarde und die Ursprünge des modernen Antisemitismus, Munich: Hanser. Siegert, F. (2001), Zwischen hebräischer Bibel und Altem Testament: Eine Einführung in die Septuaginta, Münsteraner Judaistische Studien 9, Münster: Lit. Simonetti, M. (1994), Biblical Interpretation in the Early Church: An Historical Introduction to Patristic Exegesis, Edinburgh: T&T Clark. Sipilä, S. (1998), “Max Leopold Margolis and the Origenic Recension in Joshua,” in A. Salvesen (ed.), Origen’s Hexapla and Fragments: Papers Presented at the Rich Seminar on the Hexapla, Oxford Centre for Hebrew and Jewish Studies, 25th July–3rd August 1994, 16–38, TSAJ 58, Tübingen: Mohr Siebeck.

456 REFERENCES

Skarsaune, O. (1996), “The Question of Old Testament Canon and Text in the Early Greek Church,” in M. Sæbø (ed.), Hebrew Bible/Old Testament: The History of Its Interpretation, Volume 1: From the Beginnings to the Middle Ages (Until 1300), Part 1: Antiquity, 443–50, Göttingen: Vandenhoeck & Ruprecht. Skeat, T. C. (2004), “The Use of Dictation in Ancient Book-Production,” in J. K. Elliott (ed.), The Collected Biblical Writings of T. C. Skeat, 1–32, VTSup 113, Leiden: Brill. Skehan, P. W. (1975a), “The Qumran Manuscripts and Textual Criticism,” in F. M. Cross and S. Talmon (eds.), Qumran and the History of the Biblical Text, 212–25, Cambridge: Harvard University Press. Skehan, P. W. (1975b), “The Biblical Scrolls from Qumran and the Text of the Old Testament,” in F. M. Cross and S. Talmon (eds.), Qumran and the History of the Biblical Text, 264–77, Cambridge: Harvard University Press. Skehan, P. W. (1977), “4QLXXNum: A Pre-Christian Reworking of the Septuagint,” HTR 70: 39–50. Skehan, P. W. (1980), “The Divine Name at Qumran, in the Masada Scroll, and in the Septuagint,” BIOSCS 13: 14–44. Skehan, P. W. and A. A. Di Lella (1987), The Wisdom of Ben Sira, AB 39, New York: Doubleday. Skehan, P. W., E. Ulrich, and J. E. Sanderson (1992), Qumran Cave 4 IV, Palaeo-Hebrew and Greek Biblical Manuscripts, DJD 9, Oxford: Clarendon Press. Smelik, W. (2007), “Code-Switching: The Public Reading of the Bible in Hebrew, Aramaic and Greek,” in L. Morenz and S. Schorch (eds.), Was ist ein Text? – Âgyptologische, altorientalistische und alttestamentliche Perspektiven, 123–47, BZAW 362, Berlin: De Gruyter. Smyth, H. W. (1920), Greek Grammar for Colleges, New York: American Book Company. Soisalon-Soininen, I. (1987), “Beobachtungen zur Arbeitsweise der Septuaginta-Übersetzer,” in A. Aejmelaeus and R. Sollamo (eds.), Ilmari Soisalon-Soininen: Studien zur SeptuagintaSyntax. Zu sinem 70. Geburstag am 4. Juni 1987, 28–39, AASF, Ser. B, Tom. 237, Helsinki: Suomalainen Tiedeakatemia. Sokoloff, M. and J. Yahalom (1978), “Christian Palimpsests from the Cairo Genizah,” Revue d’Histoire des Textes 8: 109–32. Sollamo, R. (1979), Renderings of Hebrew Semiprepositions in the Septuagint, Annales ASFDHL 19, Helsinki: Suomalainen Tiedeakatemia. Sollamo, R. (1985), “The LXX Renderings of the Infinitive Absolute Used with a Paronymous Finite Verb in the Pentateuch,” in N. Fernández Marcos (ed.), La Septuaginta en la investigación contemporánea: (V Congreso de la IOSCS), 101–13, Madrid: CSIC. Sollamo, R. (2001), “Prolegomena to the Syntax of the Septuagint,” in R. sollamo and S. Sipilä (eds.), Helsinki Perspectives on the Translation Technique of the Septuagint: Proceedings of the IOSCS Congress in Helsinki 1999, 23–41, PFES 82, Göttingen: Vandenhoeck & Ruprecht. Sollamo, R. (2016), “The Study of Translation Technique,” in E. Bons and J. Joosten (eds.), Die Sprache der Septuaginta/The Language of the Septuagint, 143–53, LXX.H 3, Gütersloh: Gütersloher Verlagshaus. Sophocles, E. A. (1914), Greek Lexicon of the Roman and Byzantine Periods (from B.C. 146 to A.D. 1100), New York: Scribner’s sons. Speiser, E. A. (1926), “The Pronunciation of Hebrew According to the Transliterations in the Hexapla,” JQR 16: 343–82. Spicq, C. (1994), Theological Lexicon of the New Testament, Peabody: Hendrickson Publishers.

REFERENCES 457

Spottorno, V. (2013), “The Status of the Antiochene Text in the First Century A.D.,” in S. Kreuzer and M. Sigismund (eds.), Der Antiochenische Text der Septuaginta in seiner Bezeugung und seiner Bedeutung, 74–83, DSI 4, Göttingen: Vandenhoeck & Ruprecht. Stanley, C. D. (1992), Paul and the Language of Scripture: Citation Technique in the Pauline Epistles and Contemporary Literature, SNTSMS 74, Cambridge: Cambridge University Press. Stanley, C. D. (2004), Arguing with Scripture: The Rhetoric of Quotations in the Letters of Paul, New York & London: T&T Clark. Staples, W. E. (1927), “The Hebrew of the Septuagint,” AJSLL 44: 6–30. Starr, J. (1939), The Jews in the Byzantine Empire 641–1204, Athens: Verlag der ByzantinischNeugriechischen Jahrbücher. Steiger, J. A. (2008), “The Development of the Reformation Legacy: Hermeneutics and Interpretation of the Sacred Scripture in the Age of Orthodoxy,” in M. Sæbø, M. Fishbane, and J. L. Ska (eds.), Hebrew Bible/Old Testament: The History of its Interpretation. Volume 2: From the Renaissance to the Enlightenment, 691–757, Göttingen: Vandenhoeck und Ruprecht. Steinberg, J. and T. J. Stone (2015), “The Historical Formation of the Writings in Antiquity,” in J. Steinberg and T. J. Stone with R. Stone (eds.), The Shape of the Writings, 1–58, Siphrut 16, Winona Lake: Eisenbrauns. Steiner, R. C. (2005), “On the Dating of Hebrew Sound Changes (*ḫ > ḥ and *ġ > ʿ) and Greek Translations (2 Esdras and Judith),” JBL 124: 229–67. Stern, F. (1961), The Politics of Cultural Despair: A Study in the Rise of the Germanic Ideology, Berkeley: University of California Press. Steyn, G. J. (1995), Septuagint Quotations in the Context of the Petrine and Pauline Speeches of the Acta Apostolorum, CBET 12, Kampen: Kok Pharos. Steyn, G. J. (2011), A Quest for the Assumed LXX Vorlage of the Explicit Quotations in Hebrew, FRLANT 235, Göttingen: Vandenhoeck & Ruprecht. Stolze, R. (2010), “Hermeneutics and Translation,” in Y. Gambier and L. van Doorslaer (eds.), Handbook of Translation Studies, Volume 1, 141–6, Amsterdam: John Benjamins. Stone, M. E. (1979), The Armenian Version of IV Ezra, University of Pennsylvania Armenian Texts and Studies 1, Missoula: Scholars Press. Strazicich, J. (2007), Joel’s Use of Scripture and Scripture’s Use of Joel: Appropriation and Resignification in Second Temple Judaism and Early Christianity, BibInt 82, Leiden: Brill. Stuckenbruck, L. and S. Weeks (2015), “Tobit,” in J. K. Aitken (ed.), The T&T Clark Companion to the Septuagint, 237–60, London: Bloomsbury T&T Clark. Sundberg, Jr., Albert C. (1964), The Old Testament of the Early Church, Cambridge: Harvard University Press. Swete, H. B. (1887), The Old Testament in Greek according to the Septuagint, Volume 1: Genesis–IV Kings, Cambridge: Cambridge University Press. Swete, H. B. (1887–94), The Old Testament in Greek according to the Septuagint, 3 volumes, Cambridge: Cambridge University Press. Swete, H. B. (1900), An Introduction to the Old Testament in Greek, Cambridge: Cambridge University Press. Swete, H. B. (1914), An Introduction to the Old Testament in Greek, 2nd edn., Cambridge: Cambridge University Press. Taube, M. (2012), “Jewish-Christian Collaboration in Slavic Translations from Hebrew,” in V. Izmirlieva and B. Gasparov (eds.), Translation and Tradition in “Slavia Orthodoxa,” 26–45, Slavische Sprachgeschichte 5, Zürich: Lit.

458 REFERENCES

Taylor, B. A., ed. (2001), X Congress of the International Organization for Septuagint and Cognate Studies, Oslo, 1998, SBLSCS 51, Atlanta: SBL Press. Taylor, B. A. (2009), Analytical Lexicon to the Septuagint, expanded edn., Peabody: Hendrickson. Taylor, C. (1900), Hebrew-Greek Cairo Genizah Palimpsests from the Taylor-Schechter Collection, Including a Fragment of the Twenty-Second Psalm According to Origen’s Hexapla, Cambridge: Cambridge Univeristy Press. Tchernetska, N. (2002), “Greek-Oriental palimpsests in Cambridge: problems and prospects,” in C. Holmes and J. Waring (eds.), Literacy, Education and Manuscript Transmission in Byzantium and Beyond, 243–56, The Medieval Mediterranean 42, Leiden: Brill. Tchernetska, N., J. Olszowy-Schlanger, and N. de Lange (2007), “An Early Hebrew-Greek Biblical Glossary from the Cairo Genizah,” RJ 166: 91–128. Teodorsson, S.-T. (1977), The Phonology of Ptolemaic Koine, Studia Graeca et Latina Gothoburgensia 36, Goteborg: Acta Universitatis Gothoburgensis. Teodorsson, S.-T. (1978), The Phonology of Attic in the Hellenistic Period, Studia Graeca et Latina Gothoburgensia 40, Göteborg: Acta Universitatis Gothoburgensis. Thackeray, H. St J. (1903a), “The Greek Translators of Jeremiah,” JTS 4: 245–66. Thackeray, H. St J. (1903b), “The Greek Translators of Ezekiel,” JTS 4: 398–411. Thackeray, H. St J. (1903c), “The Greek Translators of the Prophetical Books,” JTS 4: 578–85. Thackeray, H. St J. (1907), “The Greek Translators of the Four Books of Kings,” JTS 8: 262–78. Thackeray, H. St J. (1908), “Renderings of the Infinitive Absolute in the LXX,” JTS 9: 597– 601. Thackeray, H. St J. (1909), A Grammar of the Old Testament in Greek according to the Septuagint, Volume 1: Introduction, Orthography and Accidence, Cambridge: Cambridge University Press. Thackeray, H. St J. (1921), The Septuagint and Jewish Worship: A Study in Origins, London: British Academy. Theocharous, M. (2012), Lexical Dependence and Intertextual Allusion in the Septuagint of the Twelve Prophets: Studies in Hosea, Amos and Micah, LHBOTS 570, London: T&T Clark. Thomson, C. (1808), The Holy Bible, Containing the Old and New Covenant, Commonly Called the Old and New Testament: Translated from the Greek, 4 volumes, Philadelphia: Jane Aitken. Thomson, C. (1960), The Septuagint Bible: The Oldest Text of the Old Testament in the Translation of Charles Thomson, Secretary of the Continental Congress of the United States of America, 1774–1789, Edited, revised, and enlarged by C. A. Muses, Indian Hills, CO: Falcon’s Wing. Thomson, F. J. (1999), “The Slavonic Translation of the Old Testament,” in J. Krašovec (ed.), The Interpretation of the Bible: The International Symposium in Slovenia, 605–920, JSOTSup 289, Sheffield: Sheffield Academic. Thompson, E. M. (1883), Facsimile of the Codex Alexandrinus: Old Testament, Volume 2: Hosea–4 Maccabees, London: British Museum. Thompson, E. M. (1912), An Introduction to Greek and Latin Palaeography, Oxford: Clarendon Press. Threatte, L. (1980), The Grammar of Attic Inscriptions, Volume 1: Phonology, Berlin: De Gruyter. Threatte, L. (1982), “The Alleged Conservatism of Attic Epigraphical Documents: A Different View,” Hesperia Supplements 19: 148–56.

REFERENCES 459

Tilly, M. (2005), Einführung in die Septuaginta, Einführung Theologie, Darmstadt: Wissenschaftliche Buchgesellschaft. Toomer, G. J. (1996), Eastern Wisdome and Learning. The Study of Arabic in SeventeenthCentury England, Oxford: Clarendon Press. Torczyner, H. (1937), “Die Aussprache der Begad-kefat in der Geschichte der hebräischen Sprache,” Monatsschrift für Geschichte und Wissenschaft des Judentums 81: 340–51. Toury, G. (1995), Descriptive Translation Studies and Beyond, Benjamins Translation Library 4, Amsterdam: John Benjamins. Toury, G. (2012), Descriptive Translation Studies and Beyond, 2nd revised edn., Benjamins Translation Library 100, Amsterdam: John Benjamins. Tov, E. (1973), “Some Corrections to Reider-Turner’s Index to Aquila,” Text 8: 164–74. Tov, E. (1976), “Three Dimensions of LXX Words,” RB 83: 529–44. Tov, E. (1980), A Classified Bibliography of Lexical and Grammatical Studies on the Language of the Septuagint and Its Revisions, Jerusalem: Academon. Tov, E. (1981), “The Impact of the LXX Translation of the Pentateuch on the Translation of the Other Books,” in P. Casetti, O. Keel, and A. Schenker (eds.), Mélanges Dominique Barthélemy: Études Bibliques Offertes à l’Occasion de son 60e Anniversaire, 577–93, OBO 38, Göttingen: Vandenhoeck & Ruprecht. Tov, E. (1984), “The Rabbinic Tradition concerning the ‘Alterations’ Inserted into the Greek Pentateuch and Their Relation to the Original Text of the LXX,” JSJ 15: 65–89. Tov, E. (1986), “Jewish Greek Scriptures,” in R. A. Kraft and G. W. E. Nickelsburg (eds.), Early Judaism and Its Modern Interpreters, 223–37, Philadelphia: Fortress Press. Tov, E. (1987), “Die Septuatinta in ihrem theologischen und traditionsgeschichtlichen Verhältnis zur hebraischen Bibel,” in M. A. Klopfenstein, U. Luz, S. Talmon, and E. Tov (eds.), Mitte der Schrift? Ein jüdisch-christliches Gespräch. Texte des Berner Symposions vom 6.-12. Januar 1985, 237–68, Judaica et Christiana 11, Bern: Peter Lang. Tov, E. (1990a), The Greek Minor Prophets Scroll from Naḥ al Ḥever (8ḤevXIIgr) (The Seiyâl Collection 1), DJD 8, Oxford: Clarendon Press. Tov, E. (1990b), “Theologically Motivated Exegesis Embedded in the Septuagint,” in D. M. Goldenberg (ed.), Translation of Scripture: Proceedings of a Conference at the Annenberg Research Institute, 1989, 215–33, JQRSup, Philadelphia, PA: Annenberg Research Institute. Tov, E. (1992), “The Contribution of the Qumran Scrolls to the Understanding of the LXX,” in G. J. Brooke and B. Lindars (eds.), Septuagint, Scrolls and Cognate Writings, SBLSCS 33, 11–47, Atlanta: Scholars Press. Tov, E. (1993), “Some Reflections on the Hebrew Texts from Which the Septuagint was Translated,” JNSL 19: 107–122. Tov, E. (1997), “The ‘Accordance’ Search Program for the MT, LXX, and the CATSS Database,” BIOSCS 30: 36–44. Tov, E. (1991), “The CATSS Project—A Progress Report,” in C. E. Cox (ed.), VII Congress of the International Organization for Septuagint and Cognate Studies, Leuven 1989, 157–63, SBLSCS 31, Atlanta: Scholars Press. Tov, E. (1999a), “Did the Septuagint Translators Always Understand Their Hebrew Text?” in E. Tov (ed.), The Greek and Hebrew Bible: Collected Essays on the Septuagint, 203–18, VTSup 72, Leiden: Brill. Tov, E. (1999b), “Loan-words, Homophony, and Transliterations in the Septuagint,” in E. Tov (ed.), The Greek and Hebrew Bible: Collected Essays on the Septuagint, 165–82, VTSup 72, Leiden: Brill.

460 REFERENCES

Tov, E. (1999c), “The Contribution of the Qumran Scrolls to the Understanding of the Septuagint,” in E. Tov (ed.), The Greek and Hebrew Bible: Collected Essays on the Septuagint, 285–300, VTSup 72, Leiden: Brill. Tov, E. (1999d), The Greek and Hebrew Bible: Collected Essays on the Septuagint, VTSup 72, Leiden: Brill. Tov, E. (1999e), “Theologically Motivated Exegesis Embedded in the Septuagint,” in E. Tov (ed.), The Greek and Hebrew Bible: Collected Essays on the Septuagint, 257–69, VTSup 72, Leiden: Brill. Tov, E. (2005), “The Evaluation of the Greek Scripture Translations in Rabbinic Sources,” in F. García Martínez and M. Vervenne (eds.), Interpreting Translation: Studies on the LXX and Ezekiel in Honour of Johan Lust, 385–99, BETL 192, Leuven: Peeters. Tov, E. (2008), “The Greek Biblical Texts from the Judean Desert,” in E. Tov (ed.), Hebrew Bible, Greek Bible, and Qumran: Collected Essays, 339–64, TSAJ 121, Tübingen: Mohr Siebeck. Tov, E. (2010), “Reflections on the Septuagint with Special Attention Paid to the PostPentateuchal Translations,” in M. Karrer and W. Kraus (eds.), Die Septuaginta—Texte, Theologien, Einflüsse 2. Internationale Fachtagung veranstaltet von Septuaginta Deutsch (LXX.D), Wuppertal 23.-27. Juli 2008, 3–22, WUNT 252, Tübingen: Mohr Siebeck. Tov, E. (2012), Textual Criticism of the Hebrew Bible, 3rd edn., Minneapolis: Fortress Press. Tov, E. (2015a), The Text-Critical Use of the Septuagint in Biblical Research, 3rd edn., Winona Lake: Eisenbrauns. Tov, E. (2015b), “The Qumran Hebrew Texts and the Septuagint: An Overview,” in E. Tov (ed.), Textual Criticism of the Hebrew Bible, Qumran, Septuagint: Collected Essays, 353–67, VTSup 167, Leiden: Brill. Tov, E. (2016), “The Septuagint Translation of the Torah as a Source and a Resource for the Post-Pentateuchal Translators,” in E. Bons and J. Joosten (eds.), Die Sprache der Septuaginta/ The Language of the Septuagint, 316–28, LXX.H 3, Gütersloh: Gütersloher Verlagshaus. Tov, E. (2020a), “Der eine Gott und sein Verständnis: Pentateuch,” in H. Ausloos and B. Lemmelijn (eds.), Die Theologie der Septuaginta/The Theology of the Septuagint, 47–58, LXX.H 4, Gütersloh: Gütersloher Verlagshaus. Tov, E. (2020b) “The Interaction between Theological and Text-Critical Approaches,” in J. Cook and M. Rösel (eds.), Towards a Theology of the Septuagint: Stellenbosch Congress on the Septuagint 2018, 23–46, SBLSCS 74, Atlanta: SBL Press. Tov, E. and F. Polak (2009), The Revised CATSS Hebrew/Greek Parallel Text (Tov-Polak, 2008): Manual and Introduction, 2nd edn., Jerusalem. Trapp, E. (2001–2017), Lexikon zur byzantinischen Gräzität besonders des 9.-12. Jahrhunderts, Wien: Verlag der österreichischen Akademie der Wissenschaften. Trebolle Barrera, J. (2005), “The Text-Critical Value of the Old Latin and Antiochean Greek Texts in the Books of Judges and Joshua,” in F. García Martinez and M. Vervenne (eds.), Interpreting Translation. Studies on the LXX and Ezekiel in Honour of Johan Lust, 401–14, BETL 192, Leiden: Brill. Trebolle Barrera, J. (2016), “Vetus Latina,” in A. Lange and E. Tov (eds.), Textual History of the Bible, Volume 1A: 319–31, Leiden: Brill. Trommius, A. (1718), Concordantiae graecae versionis vulgo dictae LXX interpretum, 2 volumes, Amsterdam/Utrecht: Sumptibus societatis. Troxel, R. L. (2008), LXX-Isaiah as Translation and Interpretation: The Strategies of the Translator of the Septuagint of Isaiah, JSJSup 124, Leiden: Brill.

REFERENCES 461

De Troyer, K. (2012), “The Seventy-Two and Their Many Grandchildren: A Review of Septuagint Studies from 1997 Onward,” CBR 11: 8–64. Tsourkas, C. (1967), Les Débuts de l’enseignement philosophique et de la libre pensée dans les Balkans, Thessaloniki: Institute for Balkan Studies. Turner, N. (1955), “The Unique Character of Biblical Greek,” VT 5: 208–13. Turner, N. (1965), Grammatical Insights into the New Testament, Edinburgh: T&T Clark. Twining, T. (2020), “The Early Modern Debate over the Age of the Hebrew Vowel Points: Biblical Criticism and Hebrew Scholarship in the Confessional Republic of Letters,” Journal of the History of Ideas 81: 337–58. Ulrich, E. (1978), The Qumran Text of Samuel and Josephus, Missoula: Scholars Press. Ulrich, E. (1992), “The Septuagint Manuscripts from Qumran: A Reappraisal of Their Value,” in G. J. Brooke and B. Lindars (eds.), Septuagint, Scrolls and Cognate Writings, 49–80S, BLSCS 33, Atlanta: Scholars Press. Ulrich, E. (1999), The Dead Sea Scrolls and the Origins of the Bible, Grand Rapids, MI: Eerdmans. Ulrich, E. (2012), “The Evolutionary Composition of the Bible,” in J. Kloppenborg and J. H. Newman (eds.), Editing the Bible: Assessing the Task Past and Present, 23–40, SBLRBS 69, Atlanta: SBL Press. Ulrich, E. (2015), The Dead Sea Scrolls and the Developmental Composition of the Bible, VTSup 169, Leiden: Brill. United Bible Societies (n.d.), “Septuagint Bibliography,” Online: http://www.ubs-translations. org/bibliographies/#c519 (accessed February 28, 2020). Usener, K. (2011), “Zur Sprache der Septuaginta,” in M. Karrer, W. Kraus (eds.), Septuaginta Deutsch. Erläuterungen und Kommentare zum griechischen Alten Testament, Volume 1: Genesis bis Makkabäer, 40–52, Stuttgart: Deutsche Bibelgesellschaft. Utzschneider, H. (2001), “Auf Augenhöhe mit dem Text: Überlegungen zum wissenschaftlichen Standort einer Übersetzung der Septuaginta ins Deutsche,” in H.-J. Fabry and U. Offerhaus (eds.), Im Brennpunkt: Die Septuaginta: Studien zur Entstehung und Bedeutung der griechischen Bibel, Volume 1: 11–50, BWANT 8, Stuttgart: Kohlhammer. VanderKam, J. C. (2001), “Greek at Qumran,” in J. J. Collins and G. E. Sterling (eds.), Hellenism in the Land of Israel, 175–81, CJA 13, Notre Dame: University of Notre Dame Press. VanderKam, J. C. (2012), The Dead Sea Scrolls and the Bible, Grand Rapids: Eerdmans. van der Kooij, A. (1998), “Perspectives on the Study of the Septuagint: Who Were the Translators?” in F. G. Martínez and E. Noort (eds.), Perspectives on the Study of the Old Testament and Judaism: A Symposium in Honour of Adam S. van der Woude on the Occasion of his 70th Birthday, 214–29, VTSup 73, Leiden: Brill. van der Kooij, A. (2001), “The Septuagint of Psalms and the First Book of Maccabees,” in R. J. V. Hiebert, C. E. Cox, and P. J. Gentry (eds.), The Old Greek Psalter: Studies in Honour of Albert Pietersma, 229–47, JSOTSup 332, Sheffield: Sheffield Academic. van der Kooij, A. (2002), “The Textual Criticism of the Hebrew Bible Before and After the Qumran Discoveries,” in E. D. Herbert and E. Tov (eds.), The Bible as Book: The Hebrew Bible and the Judaean Desert Discoveries, 167–77, London: The British Library/Oak Knoll Press. van der Kooij, A. (2003), “On the Use of βωμός in the Septuagint,” in M. F. J. Baasten and W. Th. van Peursen (eds.), Hamlet on a Hill: Semitic and Greek Studies Presented to Professor T. Muraoka on the Occasion of His Sixty-Fifth Birthday, 600–7, OLA 118, Leuven: Peeters.

462 REFERENCES

van der Kooij, A. (2007), “The Septuagint of the Pentateuch and Ptolemaic Rule,” in G. N. Knoppers and B. M. Levinson (eds.), The Pentateuch as Torah: New Models for Understanding Its Promulgation and Acceptance, 289–300, Winona Lake: Eisenbrauns van der Kooij, A. (2008), “Isaiah and Daniel in the Septuagint: How Are These Two Books Related?” in H. Ausloos, B. Lemmelijn, and M. Vervenne (eds.), Florilegium Lovaniense: Studies in Septuagint and Textual Criticism in Honour of Florentino García Martínez, 465–73, BETL 224, Leuven: Peeters. van der Kooij, A. (2017), Review of Seulgi L. Byun, The Influence of Post-Biblical Hebrew and Aramaic on the Translator of Septuagint Isaiah, RBL, Online: http://www.bookreviews.org. van der Louw, T. A. W. (2007), Transformation in the Septuagint: Towards an Interaction of Septuagint Studies and Translation Studies, CBET 47, Leuven: Peeters. van der Louw, T. A. W. (2008a), “Linguistic or Ideological Shifts? The Problem-Oriented Study of Transformations as a Methodological Filter,” in A. Voitila and J. Jokiranta (eds.), Scripture in Transition: Essays on Septuagint, Hebrew Bible, and Dead Sea Scrolls in Honour of Raija Sollamo, 107–26, JSJSup 126, Leiden: Brill. van der Louw, T. A. W. (2008b), “The Dictation of the Septuagint Version,” JSJ 39: 211–29. van der Louw, T. A. W. (2008c), “Translation and Writing in 4QLXXLeva,” in T. Römer (ed.), The Books of Leviticus and Numbers, 383–97, BETL 215, Leuven: Peeters. van der Louw, T. A. W. (2010), “The Duration of the LXX-Pentateuch Project,” BIOSCS 43: 41–51. van der Louw, T. A. W. (2012), “Review of Natalio Fernández Marcos, and M.a Victoria Spottorno Díaz-Caro (coordinatores). La Biblia griega: Septuaginta, II: Libros históricos. Biblioteca de Estudios Bíblicos 126. Salamaca: Ediciones Sígueme, 2011,” JSCS 45: 136–8. van der Louw, T. A. W. (2019), “The Unity of LXX Genesis and Exodus,” VT 70: 270–84. van der Meer, M. (2016), “Anthropology in the Ancient Greek Versions of Gen 2:7,” in J. van Ruiten and G. van Kooten (eds.), Dust of the Ground and Breath of Life (Gen 2:7): The Problem of a Dualistic Anthropology in Early Judaism and Christianity, 36–57, TBN 20, Leiden: Brill. van der Merwe, C. H. J., J. A. Naudé, and J. H. Kroeze (2017), A Biblical Hebrew Reference Grammar, 2nd edn., London: Bloomsbury. van der Vorm-Croughs, M. (2001), “Intertextuality in the Septuagint: The Case of Isaiah 19,” BIOSCS 34: 81–94. van der Vorm-Croughs, M. (2014), The Old Greek of Isaiah: An Analysis of Its Pluses and Minuses, SBLSCS 61, Atlanta: SBL Press. van Emde Boas, E., A. Rijksbaron, L. Huitink, and M. de Bakker (2019), Cambridge Grammar of Classical Greek, Cambridge: Cambridge University Press. van Esbroeck, M. (1998), “Les versiones orientales de la Bible. Une orientation bibliographique,” in J. Krašovec (ed.), The Interpretation of the Bible. The International Symposium in Slovenia, 399–509, JSOTSup 289, Sheffield: Sheffield Academic Press. VanGemeren, W.A. (2012), New International Dictionary of Old Testament Theology and Exegesis, Grand Rapids: Zondervan. van Miert, D. (2018), The Emancipation of Biblical Philology in the Dutch Republic, 1590– 1670, Oxford: Oxford University Press. van Minnen, P. (1998), “Boorish or Bookish? Literature in Egyptian Villages in the Fayum in the Greco-Roman period,” Journal of Juristic Papyrology 27: 99–184. Vaporis, N. M. (1994), Translating the Scripture into Modern Greek, Brookline: Holy Cross Orthodox Press.

REFERENCES 463

Veltri, G. (1994a), “Die Novelle 146 Περὶ Ἑβραίων: das Verbot des Targumsvortrags in Justinian’s Politik,” in M. Hengel and A. M. Schwemer (eds.), Die Septuaginta zwischen Judentum und Christentum, 116–30, WUNT 72, Tübingen: Mohr Siebeck. Veltri, G. (1994b), Eine Torah für den König Talmai: Untersuchungen zum Übersetzungsverständnis in der jüdisch-hellenistischen und rabbinistischen Literatur, TSAJ 41, Tübingen: Mohr Siebeck. Veltri, G. (2002), Gegenwart der Tradition. Studien zur jüdischen Literatur und Kulturgeschichte, JSJSup 69, Leiden: Brill. Veltri, G. (2006), Libraries, Translations and “Canonic” Texts—The Septuagint, Aquila and Ben Sira in the Jewish and Christian Traditions, JSJSup 109, Leiden: Brill. Veltri, G. (2009), “The Septuagint in Disgrace: Some Notes on the Stories on Ptolemy in Rabbinic and Medieval Judaism,” in N. de Lange, J. Krivoruchko, and C. Boyd-Taylor (eds.), Jewish Reception of Greek Bible Versions, 142–54, TSMEMJ 23, Tübingen: Mohr Siebeck. Vessella, C. (2018), Sophisticated Speakers: Atticistic Pronunciation in the Atticist Lexica, Trends in Classics – Supplementary Volumes 55, Berlin: De Gruyter. Vetus Latina (1954-), Vetus Latina: Die Reste der altlateinischen Bibel, Freiburg: Herder. Vianès, L. (2011), La Bible d’Alexandrie. XXIII.12. Les Douze Prophètes. Malachie, BdA 23.12, Paris: Cerf. Vinel, F. (2002), La Bible d’Alexandrie. XVIII. L’Ecclésiaste, BdA 18, Paris: Cerf. von Tischendorf, F. C. (1850), Vetus Testamentum Graece iuxta LXX Interpretes, 3 volumes, Leipzig: Brockhaus. de Vries, J. and M. Karrer, eds. (2013), Textual History and the Reception of Scripture in Early Christianity/Textgeschichte und Schriftrezeption im frühen Christentum, SBLSCS 60, Atlanta: SBL Press. Wagner, J. R. (2002), Heralds of the Good News: Isaiah and Paul “In Concert” in the Letter to the Romans, NovTSup 101, Leiden: Brill. Wagner, J. R. (2013), Reading the Sealed Book: Old Greek Isaiah and the Problem of Septuagint Hermeneutics, FAT 88, Tübingen: Mohr Siebeck; Waco, TX: Baylor University Press. Wakker, G. C. (2009), “‘Well I Will Now Present My Arguments’. Discourse Cohesion Marked by οὖν and τοίνυν in Lysias,” in S. Bakker and G. Wakker (eds.), Discourse Cohesion in Ancient Greek, 63–81, Leiden: Brill. Waldman, N. M. (1989), The Recent Study of Hebrew: A Survey of the Literature with Selected Bibliography, Bibliographica Judaica 10, Winona Lake: Eisenbrauns. Walser, G. (2001), The Greek of the Ancient Synagogue. An Investigation on the Greek of the Septuagint, Pseudepigrapha and the New Testament, Stockholm: Almqvist & Wiksell International. Walser, G. (2008), “The Greek of the Bible: Translated Greek or Translational Greek?” in A. Voitila and J. Jokiranta (eds.), Scripture in Transition: Essays on Septuagint, Hebrew Bible, and Dead Sea Scrolls in Honour of Raija Sollamo, 449–61, JSJSup 126, Leiden: Brill. Walser, G. (2012), Jeremiah, SEPT, Leiden: Brill. Walser, G. (2016), “Statistical Differences between Translation Greek and Non-Translation Greek Texts,” in E. Bons and J. Joosten (eds.), Die Sprache der Septuaginta/The Language of the Septuagint, 221–8, LXX.H 3, Gütersloh: Gütersloher Verlagshaus. Walters (formerly Katz), P. (1973), The Text of the Septuagint: Its Corruptions and Their Emendation, D. W. Gooding (ed.), Cambridge: Cambridge University Press.

464 REFERENCES

Ward, W. A. (1992), “Goshen (Place),” in D. N. Friedman (ed.), The Anchor Bible Dictionary, Volume 2: 1076–7, New York: Doubleday. Ware, K. (1963) The Orthodox Church, Harmondsworth: Penguin. Ware, T. (1964), Eustratios Argenti. A Study of the Greek Church under Turkish Rule, Oxford: Clarendon Press. Wasserstein, A. and D. Wasserstein (2006), The Legend of the Septuagint: From Classical Antiquity to Today, Cambridge: Cambridge University Press. Watts, R. (1997), Isaiah’s New Exodus and Mark, WUNT 2/88, Tubingen: Mohr Siebeck. Weber, R., ed. (1994), Biblia Sacra iuxta Vulgatam Versionem, 4th edn., Stuttgart: Deutsche Bibelgesellschaft. Wellhausen, J. (1871), Der Text der Bücher Samuelis, Göttingen: Vandenhoeck & Ruprecht. Wenham, G. J. (2000), Genesis 16-50, WBC 2, Grand Rapids: Zondervan. Westermann, W. L. (1929), Upon Slavery in Ptolemaic Egypt, New York: Columbia University Press. Wevers, J. W. (1954), “Septuaginta-Forschungen,” TRu 22: 85–138. Wevers, J. W. (1968), “Septuaginta Forschungen seit 1954,” TRu 33: 18–76. Wevers, J. W. (1970), “Ḥeth in Classical Hebrew,” in J. W. Wevers and D. B. Redford (eds.), Essays on the Ancient Semitic World, 101–12, Toronto Semitic Texts and Studies, Toronto: University of Toronto Press. Wevers, J. W., ed. (1974a), Septuaginta. Vetus Testamentum Graecum. I. Genesis, Göttingen: Vandenhoeck & Ruprecht. Wevers, J. W. (1974b), Text History of the Greek Genesis, MSU 11, Göttingen: Vandenhoeck & Ruprecht. Wevers, J. W., ed. (1982a), Septuaginta. Vetus Testamentum Graecum. III.1. Numeri, Göttingen: Vandenhoeck & Ruprecht. Wevers, J. W. (1982b), “An Early Revision of the Septuagint of Numbers,” ErIsr 16: 235*–9*. Wevers, J. W. (1985), “The Use of the Versions for Text Criticism: The Septuagint,” in N. Fernández Marcos (ed.), La Septuaginta en la investigación contemporánea: (V Congreso de la IOSCS), 15–24, TECC 34, Madrid: Instituto Arias Montano. Wevers, J. W., ed. (1986), Septuaginta. Vetus Testamentum Graecum. II.2. Leviticus, Göttingen: Vandenhoeck & Ruprecht. Wevers, J. W. (1990), Notes on the Greek Text of Exodus, SBLSCS 30, Atlanta: Scholars Press. Wevers, J. W., ed. (1991), Septuaginta. Vetus Testamentum Graecum. II.1. Exodus, Göttingen: Vandenhoeck & Ruprecht. Wevers, J. W. (1992), Text History of the Greek Exodus, MSU 21, Göttingen: Vandenhoeck & Ruprecht. Wevers, J. W. (1993a), “A Secondary Text in Codex Ambrosanus of the Greek Exodus,” in R. Gryson (ed.), Philologia Sacra: Biblische und patristische Studien für Hermann J. Frede und Walter Thiele zu ihrem siebzigsten Geburtstag, 36–48, AGLB 24, Freiburg: Herder. Wevers, J. W. (1993b), Notes on the Greek Text of Genesis, SBLSCS 35, Atlanta: Scholars Press. Wevers, J. W. (1995), Notes on the Greek Text of Deuteronomy, SBLSCS 39, Atlanta: Scholars Press. Wevers, J. W. (1997), Notes on the Greek Text of Leviticus, SBLSCS 44, Atlanta: Scholars Press. Wevers, J. W. (1998), Notes on the Greek Text of Numbers, SBLSCS 46, Atlanta: Scholars Press.

REFERENCES 465

Wevers, J. W. (2003), “The Future of Septuagint Textual Studies,” in S. McKendrick and O. O’Sullivan (eds.), The Bible as Book: The Transmission of the Greek Text, 209–19, London: British Library. Wevers, J. W. (2005), “The Dead Sea Scrolls and the Septuagint,” BIOSCS 38: 1–24. Wevers, J. W., ed. (2006), Septuaginta. Vetus Testamentum Graecum. III.2. Deuteronomium, 2nd edn., Göttingen: Vandenhoeck & Ruprecht. Whiteford, F. J. (2015), “An Orthodox Look at English Translations of the Bible,” Online: http://orthochristian.com/81240.html (accessed May 16, 2019). Williams, F., trans. (2013), The Panarion of Epiphanius of Salamis, Books II and III. De Fide, 2nd edn., NHMS 79, Leiden: Brill. Williams, P. J. (2011), “The Bible, the Septuagint, and the Apocrypha: A Consideration of Their Singularity,” in G. Khan and D. Lipton (eds.), Studies on the Text and Versions of the Hebrew Bible in Honour of Robert Gordon, 169–80, VTSup 149, Leiden: Brill. Williams, P. J. (2018), “Semitic Long/i/Vowels in the Greek of Codex Vaticanus of the New Testament,” in N. Vidro, R. Vollandt, E.-M. Wagner, and J. Olszowy-Schlanger (eds.), Studies in Semitic Linguistics and Manuscripts: A Liber Discipulorum in Honour of Professor Geoffrey Khan, 15–26, SSU 30, Uppsala: Uppsala Universitet. Williams, T. F. (2001), “Towards a Date for the Old Greek Psalter,” in R. J. V. Hiebert, C. E. Cox, and P. J. Gentry (eds.), The Old Greek Psalter: Studies in Honour of Albert Pietersma, 248–76, JSOTSup 332, Sheffield: Sheffield Academic. Williamson, H. G. M. (1985), Ezra, Nehemiah, WBC 16, Waco: Word. Wilk, F. (2006). “The Letters of Paul as Witnesses to and for the Septuagint Text” in W. Kraus and R. Glenn Wooden (eds.), Septuagint Research: Issues and Challenges in the Study of the Greek Jewish Scriptures, 253–71, SBLSCS 53, Atlanta: SBL Press. Wirth, R. (2016), Die Septuaginta der Samuelbücher: Untersucht unter Einbeziehung ihrer Rezensionen, DSI 7, Göttingen: Vandenhoeck & Ruprecht. Wolters, A., Proverbs, SEPT, Leiden: Brill. Woods, N. T. (2009), “A Critical Edition of the Hexaplaric Fragments of Job, Chapters 1–21,” Ph.D. dissertation, Southern Baptist Theological Seminary, Louisville. Wright, B. G. (1987), “The Quantitative Representation of Elements: Evaluating ‘Literalism’ in the LXX,” in C. E. Cox (ed.), VI Congress of the International Organization for Septuagint and Cognate Studies, Jerusalem, 1986, 311–35, SBLSCS 23, Atlanta: Scholars Press. Wright, B. G. (2003), “Access to the Source: Cicero, Ben Sira, the Septuagint and Their Audiences,” JSJ 34: 1–27. Wright, B. G. (2006), “The Letter of Aristeas and the Reception History of the Septuagint,” BIOSCS 39: 47–67. Wright, B. G. (2008), “The Septuagint and Its Modern Translators,” in M. Karrer and W. Kraus (eds.), Die Septuaginta—Texte, Kontexte, Lebenswelten: Internationale Fachtagung Veranstaltet Von Septuaginta Deutsch (LXX.D), Wuppertal 20.-23. Juli 2006, 103–14, WUNT 219, Tübingen: Mohr Siebeck. Wright, B. G. (2011), “The Letter of Aristeas and the Question of Septuagint Origins Redux,” JAJ 2: 304–26. Wright, B. G. (2013), “Pseudonymous Authorship and Structures of Authority in the Letter of Aristeas,” in G. Xeravits, I. Kalimi, and T. Niklas (eds.), Scriptural Authority in Early Judaism and Ancient Christianity, 43–62, DCLS 16, Berlin: De Gruyter. Wright, B. G. (2015), The Letter of Aristeas: “Aristeas to Philocrates” or “On the Translation of the Law of the Jews,” CEJL, Berlin: De Gruyter.

466 REFERENCES

Wright, N. T. (2013), Paul and the Faithfulness of God, London: SPCK. Wright, W. (2004), Catalogue of the Syriac Manuscripts in the British Museum, 3 volumes, Piscataway: Gorgias. Würthwein, E. (1995), The Text of the Old Testament, 2nd edn., E. F. Rhodes (trans.), Grand Rapids: Eerdmans. Würthwein, E. (2014), The Text of the Old Testament: An Introduction to the Biblia Hebraica, 3rd edn., A. A. Fischer (ed.), E. F. Rhodes (trans.), Grand Rapids: Eerdmans. Wutz, F. (1925–1933), Die Transkriptionen von der Septuaginta bis zu Hieronymus, 2 volumes, Stuttgart: Kohlhammer. Yardney S. (2018), “The Use of Glossaries by the Translators of the Septuagint,” Textus 28: 157–77. Yeivin, I. (1985), ‫מסורת הלשון העברית המשתקפת בניקוד הבבלי‬, Jerusalem: Academy of the Hebrew Language. Yuditsky, A. E. (2013), “Transcription into Greek and Latin Script: Pre-Masoretic Period,” in G. Khan (ed.), Encyclopedia of Hebrew Language and Linguistics, Volume 3: 803a–22b, Leiden: Brill. Yuditsky, A. E. (2017), ‫דקדוק העברית של תעתיקי אוריגמס‬, Jerusalem: The Academy of the Hebrew Language. Zadok, R. (1988), The Pre-Hellenistic Israelite Anthroponomy and Prosopography, OLA 28, Leuven: Peeters. Ziegert, C. (2008), “Das Buch Ruth in der Septuaginta als Modell für eine integrative Übersetzungstechnik,” Bib 89: 221–51. Ziegert, C. (2017), “Kultur und Identität. Wörtliches Übersetzen in der Septuaginta,” VT 67: 648–65. Ziegler, J. (1934), Untersuchungen zur Septuaginta des Buches Isaias, ATA 12.3, Münster: Aschendorffsche Verlagsbuchhandlung. Ziegler, J., ed. (1939), Septuaginta. Vetus Testamentum Graecum Auctoritate Academiae Scientiarum Gottingensis editum, XIV Isaiah, Göttingen: Vandenhoeck & Ruprecht. Ziegler, J., ed. (1957), Septuaginta. Vetus Testamentum Graecum Auctoritate Academiae Scientiarum Gottingensis editum, XV, Jeremias, Baruch, Threni, Epistula Jeremiae, Göttingen: Vandenhoeck & Ruprecht. Ziegler, J. (1962), Die Septuaginta: Erbe und Auftrag, WU 33, Würzburg: Julius-MaximiliansUniversität. Ziegler, J., ed. (1980), Septuaginta. Vetus Testamentum Graecum. XII.1. Sapientia Salomonis, 2nd edn., Göttingen: Vandenhoeck & Ruprecht. Ziegler, J., ed. (1982), Septuaginta. Vetus Testamentum Graecum. XI.4. Iob, Göttingen: Vandenhoeck & Ruprecht. Ziegler, J., ed. (1983), Septuaginta. Vetus Testamentum Graecum. XIV. Isaias, 3rd edn., Göttingen: Vandenhoeck & Ruprecht. Ziegler, J., ed. (1999), Septuaginta. Vetus Testamentum Graecum. XVI.2. Susanna, Daniel, Bel et Draco, 2nd edn., Göttingen: Vandenhoeck & Ruprecht. Ziegler, J., ed. (2013), Septuaginta. Vetus Testamentum Graecum. XV. Ieremias, Baruch, Threni, Epistula Ieremiae, 4th edn., Göttingen: Vandenhoeck & Ruprecht. Ziegler, J., ed. (2014), Septuaginta. Vetus Testamentum Graecum. XIII. Duodecim Prophetae, 4th edn., Göttingen: Vandenhoeck & Ruprecht. Ziegler, J., ed. (2015), Septuaginta. Vetus Testamentum Graecum. XVI.1. Ezechiel, 4th edn., Göttingen: Vandenhoeck & Ruprecht.

REFERENCES 467

Ziegler, J., ed. (2016), Septuaginta. Vetus Testamentum Graecum. XII.2. Sapientia Jesu Filii Sirach, 3rd edn., Göttingen: Vandenhoeck & Ruprecht. Ziegler, J., ed. (2017), Septuaginta. Vetus Testamentum Graecum. XII.1. Sapientia Salomonis, 3rd edn., Göttingen: Vandenhoeck & Ruprecht. Zipor, M. (2005), ‫תרגום השבעים לשפר בראשית‬/The Septuagint Version of Genesis, Ramat Gan: BarIlan University Publications. Zohrapian, H., ed. ([1805] 1984), Astuatsashunch‘ Matean ew Nor Ktakaranats‘ (Scriptures of the Old and New Testaments), Venice, Reprint: A facsimile reproduction with an introduction by C. Cox, Classical Armenian Text Reprint Series, Delmar: Caravan Books. Zuckermann, G. (2003), Language Contact and Lexical Enrichment in Israeli Hebrew, Palgrave Studies in Language History and Language Change, Basingstoke: Palgrave Macmillan. Zuurmond, R. (1992), “Versions, Ancient (Ethiopic),” in D. N. Freedman (ed.), The Anchor Bible Dictionary, Volume 6: 808–10, New York: Doubleday.

ANCIENT SOURCES

A Abbahu, R., of Caesarea 270 Alcuin of York 264 Alexander Polyhistor 240 Aquila 64, 66, 115, 124, 151, 201, 259, 271 Aristarchus 193, 202 Aristeas the Exegete 238, 257, 359 Aristobulus 9, 231, 233, 256 Aristophanes 193 Artapanus 237–8 Athanasius 220, 258, 265 Augustine of Hippo 4, 227–8, 255, 257, 259, 263, 265, 270, 294, 296, 306 B Bar Qappara 270 Basil of Neopatrae 220 C Cicero 193 Clement of Alexandria 219, 233, 237–8, 257 Cratander, Andreas 304 Cyprian 170 Cyril of Jerusalem 226, 288–9 D Demetrius of Phalerum 232, 234, 239 Demetrius the Chronographer 232–3, 236–7, 240 Drusius, Johannes 305 E Eliezer, Tobias ben 272 Epiphanius 203–4, 217, 226, 257, 347–8, 359 Erasmus 304–5 Eupolemus 238 Euripedes 237 Eusebius of Caesarea 4, 167, 191, 194, 199, 200, 202–3, 205, 233, 237–8, 240, 258–9, 306 Eusebius of Emesa 259, 261

Euthalius of Rhodes 64 Ezekiel the Tragedian 232, 237 F Fisher, John 306 G Gamaliel, R. Simon ben 270 Gennady, abp. of Novgorod 290 Gregory of Nyssa 261 H Hilarion 287, 294 Hilary of Poitiers 128 Hippolytus 170, 220 Homer 99 I Irenaeus of Lyons 4, 170, 219, 255, 257, 259–61 Isocrates 24 J Jerome 93–4, 165, 168, 176, 194, 201, 203–4, 217–18, 255–9, 262–5, 270, 289, 295, 306 John Chrysostom 169, 184, 220, 257 John of Damascus 217 Josephus 4, 10, 68, 170, 207, 209, 218–19, 231, 235–6, 238, 240–57, 304, 306, 308 Julian 184 Julius Africanus 260 Justin Martyr 4, 175, 226, 251, 255–7, 259–61, 306 K Kapsali, Elijah 273 L Lucian 64, 168

ANCIENT SOURCES

M manuscripts 1QIsaiah 251 11QPs 141 4Q196-200 152 4QJeremiah 136, 156–7 4QLXXDeut 10 4QLXXLev 10, 154–5, 159 4QLXXNum 154, 156–9 4QMMT C 208 4QpapLXXLev 154, 159 4QSamuel 135, 170 8HevXII 151, 162 G 180, 195 MSS 19 129, 130 MSS 22 130 MSS 36 130 MSS 48 130 MSS 51 130 MSS 82 129 MSS 93 129–30 MSS 96 130 MSS 108 129–30 MSS 121 130 MSS 127 129 MSS 153 195 MSS 231 130 MSS 319 130, 152 MSS 341 179 MSS 454 195 MSS 575 182 MSS 637 182 MSS 700 129 MSS 763 130 MSS 848 178 MSS 854 182 MSS 857 182 MSS 922 195 MSS 928 195 MSS 955 182 MSS 974 182 MSS Z 129 P.Fouad 266 (Ra848) 125, 178 P.Ryl. 1458 10 P.Ryl. 458 (Ra957) 125 P46 250 Manutius, Aldus 304 Masius, Andreas 305 Methodius 288–9

469

O Olympiodorus of Alexandria 220 Origen 64, 124, 130, 150, 166–8, 175, 185, 187, 194, 221, 258–63, 265–6, 270, 303 P Pace, Richard 306 Pamphilus 167, 197, 199, 202–3 Philo 19, 208–9, 231–2, 234, 240, 256–7, 259, 304, 347–8, 359 Philo the Epic Poet 238 Plato 24, 30, 234 Polychronius of Apamea 220 Pseudo Justin 257 Pythagoras 234 R Rufinus 194, 201, 203, 218, 255 S Sibylline Oracles 303 Soncino, Eliezer 273 Sophocles 67 Strabo 99 Symmachus 64, 259 T Tertullian 170 Theodoret of Cyrrhus 169, 194, 203, 220, 257, 261, 264 Theodotian 64, 101, 103, 107, 124, 130, 185–6, 201, 259 Theodulf of Orle’ans 264 Thucydides 24 V Vergara, Juan de 304 Virgil 193 Vives, Juan Luis 306 Y Yehiel, Nathan bar 276 Z Zenodotus 193

BIBLICAL/EXTRA-BIBLICAL REFERENCES

1 Chronicles – 1 Clement 55 – 1 Corinthians 1:24 1:26 2:16 14:21 15:54-55 15:55 1 Enoch 1 Esdras 1 Ezra 1 Kingdoms 1 Kings 1 1–2 2:22 15:23 16:28 16:34 22 22:41-52 1 Maccabees 1 Peter 2:6 5:5 1 Samuel 1:23 15:11 2 Chronicles 3:10 28 2 Corinthians 5:19 2 Esdras 12:19 16:1-2 2 Ezra

60, 64, 70 220 258 253 249 252, 256 250 248 69 38–9, 221, 368, 374 132 70 163 49 132 132 132 132 163 132 127, 223 252 244 135 166 60, 64, 70, 127, 133, 284, 290 142 251 251 38–9, 42, 44–5,  54, 60, 205, 217, 368, 373 353 353 132

2 John 2 Kingdoms 11:7 2 Kings 16:2 25 2 Maccabees 1:10 2:13-15 2 Samuel 3:8 5:4-5 10:6 15:2 18:28 3 Esdras 3 John 3 Kingdoms 3 Maccabees 4 Ezra 4 Kingdoms 1:10 8:12 4 Maccabees A Acts 1:3 2:14-41 7:28 7:53 15:16-18 28:26-27 Amos 1:3 1:13 4:13 5:8 5:24 6 6:4

221 70 343 251 163 127, 223, 231, 374 233 208 132 351 169–70 163 169 248 291 221 70 65, 223, 231, 290 188, 207 70 246 113 65, 69, 223, 231, 286, 290 182 264 30 244 247 244, 254 245

113 113 322 107 116 117 104

BIBLICAL/EXTRA-BIBLICAL REFERENCES

6:6 6:11 7:1 9:11-12 9:13

116 105 117 243–4, 254, 322 113–14

B Baruch 3:29–4:1 4:36–5:9 Bel and the Dragon 1:1-2

217, 223, 284, 335, 368 219 219 217, 224, 285 220

C Chronicles

128–9, 132–3, 262, 368

D Daniel 2:35 2:45 3:18 11:45 Daniel (OG) 2:14 3 3:9-14 3:13 3:13-14 3:95 4:37 Daniel, additions to Deutero-Zechariah Deuteronomy 5:4 5:29 7:18 10:24 11:4 24:1-4 32:35 32:6 Deuteronomy (OG) 33:2

65, 130, 133, 163, 227, 284, 286, 292, 368 247 246 102 43 107, 201, 217, 368 102 220 100–3 102 103 101 247 220–1, 223 113 10, 127–8, 130, 178, 243, 370 21 99 236 236 150 12 252 105 149, 252 247

471

E Ecclesiastes Ecclesiastes (OG) Ecclesiasticus Enoch Epistle of Jeremiah 43–44 Esdras Esther Esther, additions to Exodus 1:1 1:9-10 1:10 1:15 1:22 2:2-5 2:5-6 2:14 3:14 3:20 4:1 8:9 9:3 10:24 12:10 13:9 13:18 13:20 15:3 16:20 17 18:2 24:17 27:11 28:4-7 30:28 32:34 33:12 34:7 35–40 Exodus (OG) Ezekiel

127, 205, 208, 272, 292, 324–5, 333, 368 69 65 338 223, 284, 335 150 284, 290 130, 133, 178, 197, 208, 217, 220, 227, 284, 289–90, 333, 368, 374 223 127, 130, 240, 278, 325, 370 138 87–8 139, 143 141 91 237 90 244 324 236 100 99 236 76 46, 253 236 237 353 320, 321 101 237 281 29 76 150 40 25 29 29 136 46, 144, 149 113, 163, 205, 241, 292, 300

472

BIBLICAL/EXTRA-BIBLICAL REFERENCES

5:12–6:3 36:20-27 47:5 Ezra 2:2 2:56 8:7 10:24

195 253 116 217, 284, 290 44 48 39 47–8

G Galatians 3:15 3:15-16 3:19 Genesis 1:1 1:2 1:9 1:14 1:27 2–3 2:2 2:4 2:5 2:8-10 2:17 3:17 3:23-24 4:1 4:8 4:12 4:16 5:24 7:11 8:10 8:21 9:22 10:10 10:22 11:1 11:1-9 11:9 11:10 11:25-27 11:26-27 11:27 12:1 12:4 12:13

254 247 247 127, 175, 178, 243, 292, 315, 346–7, 352, 359, 370 24, 31 95, 105, 323 95 179 165 270 236 80 99 353 349 323 353 258 294 349 353 258 294 350 323 145 353 81 348, 355 348 353, 354 80, 355 79–80 81–2 88 81 354 16, 323

14:13 14:14 17:1 17:17 18:26-31 19:7 21:22 21:33 22:17 24:22 25:10 25:19 26:24 26:26 28:19 31:13 31:53–36:18 34 34:25 35:1 37:28 40:15 41:3 41:34 42:2-3 43:2 45:10 46:28 46:28-29 46:34 47:1-6 49:11 50:2 50:26 Gospel of Thomas 52 H Habakkuk 2:3-4 2:16 3:5 Hebrews 3:7-11 8:8-12 10:5 10:30 10:37-38 13:6 Hosea 2:20 8:1

258 258 209 354 323 114 32, 351–2 357–8 348 358 99 80 323 351–2 354, 360 354 180 355–6 356–7 7, 354 358 349 17, 348 76 10, 99 20, 99 353 180 353 353 27, 353 280 357 357 207

151 252 105 318 245 245 248–9, 264 252 252 244 94 321 324

BIBLICAL/EXTRA-BIBLICAL REFERENCES

11:1 13:14

263, 290–2 248, 250

I Isaiah 6:9 6:9-10 7:14 8:14 9:1 10:22 11:10, 12 19:11 26:8 26:20 28:9-11 28:11-12 28:16 30:17 31:9 32:6 36:11 40:3 40:4 40:13 42:1 42:4 45:22-23 52:5 52:5 52:7 58:2 59:8 59:20 61:1 Isaiah (OG) 7:14 25:8 40:3

75, 94, 178, 199, 205, 243, 271, 292, 315, 324, 373 249 245 251, 256–7, 259–61, 319 252 107 249 244 105 250 253 258 252 252 244 244 105 115 248 249 244, 249 291 250 246 244 7, 253 250 373 248 248 373 112 249 250 249

J Jeremiah 1:6 2:6 10

94, 113, 131, 133, 135, 217, 219, 290, 292, 368 324 106–8 157

473

13:16 17:6 18:20 28:43 31:31-34 39:7-9 44:10 Jeremiah, additions to Job 1 1:20-22 2:1 2:3 2:4 3:5 3:12 4:1 5:22 6:1 9:4 11:4 12:2 13:16 14:4-5 15:14 15:15 15:16 18:3 19:15 25:5 28:16 31:15-17 32:1 36:28 36:38 37–38 37:12 37:5-7 37:18 42:14 Job (OG) Joel 2:19 John 6:31 7:38 10:34 12:13 12:15

107 106 107 107 245 99 101 220–1, 223 75, 96, 128, 132, 163, 188, 199, 205, 262, 368 183 86–7 86–8 104 101 107 184 101 105 101 184 28 115 245–6 209 184 28 184 28 28 28 279–80 185 28 183 183 182 185–6 183 184 30 69, 140, 181 100 253 253 256 253 253

474

12:40 13:18 19:36-37 21:9 Jonah 3:4 3:9 Joshua 4:14 5:12 7:20 13:21 24:17 Joshua (OG) Judges 2:6 5:8 6:2-24 6:32 8:5 14:11-20 14:20 15:2 19:2 Judges (OG) Judith 9:7 16:2

BIBLICAL/EXTRA-BIBLICAL REFERENCES

253 248, 253 253 245 151, 273, 276, 286 263 166 162, 168 25 164–5 100 244 167 125 70, 127, 130–1, 133, 162–3, 168, 171, 286, 333, 339, 367, 368 163 145–6 287 171–2 164 351 351 6, 351 114 124 69–71, 132, 178, 210, 220–3, 227, 262, 284, 285, 290 321 321

K Kingdoms Kings

197, 289, 341, 368 128–30, 132–3, 162–3, 168, 171

L Lamentations Letter of Jeremiah Leviticus 1:11–6:5

163, 217, 219, 292, 335, 368 130, 149, 368 10, 127–8, 130, 144, 371 150

3:12 4:27 16:22 26:2-16 26:5 26:9 26:11 Leviticus (OG) Luke 3:5 4 5:2 M Maccabees Malachi Mark 1:3 Matthew 1:21 1:23 2:6 2:15 4:16 12:18 12:21 13:14 18:20 28:20 Micah 5:2 Minor prophets. See Twelve Prophets, the N Nahum Nehemiah 2:19 6:1-2 7:58 Numbers 3:40 3:40-51 4:1-16 6:25 9:12 12:8 25:4 Numbers (OG)

155–6 155–6 107 150 113, 114 154–5 155 149 249 373 245

132, 221–2, 284–5, 368, 373 279 248–9 167, 319, 367, 373 251 249, 251, 257 248, 250 263, 291 249 291 250 249 251 251 151 248

151 217, 290 353 353 48 130, 325, 371 156 150 150 29 253 29 75 149

BIBLICAL/EXTRA-BIBLICAL REFERENCES

O Octateuch Odes P Pentateuch Pentateuch (OG) Philippians 1:19 2:10-11 Prayer of Manasseh Prophets Proverbs 3:4 3:34 4:3 7:20 8:10 8:22 19:4 19:7 23:10 24:27 24:30 30:32 31:10-31 31:16 Psalms Psalms 3:8 5:2 5:9 15:9 18:3

127, 132 335, 338–9, 341

46, 60, 71, 74–5, 208–9, 234, 273, 315, 321, 325, 341, 347, 359, 367, 372 114–15, 127, 133 245 246 65, 290, 333 127, 132, 208, 210, 341 64, 68, 128, 195, 199, 205, 273, 324–5, 368 209 244 209 99 99 258–9 351 351 100 99 99 99 98 98–100 27–8, 30, 74–5, 96, 127, 144, 163, 179, 191, 208, 217, 243, 262, 273, 286,  292–3, 305,  314–15, 332–3 261 106 146 30 106

475

21:7 22:4 26:4 33:13 34:20 36:16 39:7 40:6 41:9 48:11 49:13 60:10 62:2 76:4 85:5 86:8-10 91:7 93:8 95:7-14 103:27 104:2 104:10 118:6 118:22 151 Psalms of Solomon 2:12 R Revelation 4:1 15:3-5 17:14 20:11 Romans 2:24 2:24 2:25-29 3:17 9:27 9:33 10:15 11:26 11:34 12:19 15:12 Ruth 3:2-3 4:10

98 107 98 104 253 104 264 248, 251 248 106 105 141 106 321 97 246 105–6 105 245 104 144 137–8 244 252 223, 284, 290, 373 65, 69, 222–3, 339, 341 75

246 246 247 246–7 253 27, 244 253 249 249 252 250 248, 250 244 252 244 127, 132, 368 89 99

476

S Samuel Sirach 18:18 44:16 51 51:13 Song of Songs Susanna T Tobit

BIBLICAL/EXTRA-BIBLICAL REFERENCES

128, 130–3, 162–3, 168, 171 21, 132, 144, 195, 205, 208, 220–3, 227, 284, 285 105 258 141 105 163, 205, 208, 262, 276, 292, 368 220, 224, 285

152, 178, 210, 220–3, 227, 262, 284–6, 290, 339, 368

Twelve Prophets, the

96, 113–14, 151, 162, 292, 314, 319, 324, 335

W Wisdom of Solomon 7:17-21 16:2

93, 132, 195, 205, 209, 218,  220–3, 227, 231, 262, 284–5, 290, 318, 368 221 74

Z Zechariah 9:15 11:17 14:20 Zephaniah

151 116 117 116 151

MODERN AUTHORS

A Adams, Sean 253 Aejmelaeus, Anneli 84, 115, 165, 179, 316 Aitken, James K. 66, 131, 324, 369, 393–4 Aland, Kurt 187 Albl, Martin 251 Allen, Willoughby 57, 367 Aslanov, C. 275, 277 Ausloos, H. 316–17, 319–20, 325 Avemarie, Friedrich 118

Caro, Spottorno Díaz 66 Chacon, Pedro 306 Chamberlain, G. A. 67, 388 Chastoupis, Athanasois 288 Cignelli, L. 369, 388 Cimosa, M. 383 Collins, John 238 Collins, Nina 239 Conybeare, F. C. 369, 388 Cook, Johann 324

B Bakhtin, Mikhail 109 Barone, Francesca 217 Barr, James 40–1, 84, 114, 116 Barrera, Julio Trebolle 369 Barrington, Shute 307 Barthélemy, Dominique xiii, 130, 162–4, 301, 335 Barton, John 208 Beale, Gregory 247 Belleli 277 Bengel, J. A. 366 Bertram, G. 67, 315 Bickerman, Elias 239 Biel, J. C. 66 Blondheim, D. S. 276, 278 Bons, E. 387, 393–4 Boyd-Taylor, Cameron 1, 70–1, 278, 334 Brannan, R. 329, 337–9, 390 Brenton, Lancelot C. 187, 293, 345, 390 Brønno, Einar 41–2, 49 Brock, S. P. 382 Brooke, A. E. 1, 45, 375, 384 Brooke, George 118 Brucker, R. 394 Brunello, A. 391 Busto Sáiz, J. R. 64

D Dafni, E. 316–17 de Crom, D. 276 de Lange, N. 274, 276 de Troyer, K. 382 de Vries, Johannes 249 Deissmann, Albert 66, 314, 351–2, 369 Delicostopoulos, A. 295 Dempster, Stephen 208 Dhont, M. 71, 394 Díaz-Caro, Spottorno 341, 389, 391 Dickey, E. 193 Dines, Jennifer 112, 117, 373, 383 Dogniez, Cécile 113, 381, 393 Donaldson, Amy 267 Dooley, R. A. 86 Dorival, Gilles 208, 264, 335, 337, 369, 383 Dos Santos, E. C. 65 Driver, S. R. 135 Drozdov, Filaret 291, 295

C Cacciari, A. 329 Cadwallader, Alan 249–50 Caird, G. B. 66, 352

E Ehorn, Seth 253 Evans, Trevor 369, 394 Even-Zohar, I. 71 Evseev 289 Eynikel, E. 67, 388 F Fabry, H.-J. 155, 157

478

Fernández-Marcos, Natalio 64, 66, 157, 179, 239, 274–8, 340–1, 383–4, 387, 389, 391 Field, Frederick xiii, 1, 66, 166, 387 Fincati 278 Flashar, M. 314 Frankel, Z. 314 Freudenthal, J. 314 Friedrich, G. 67 Fritsch, Charles 2, 381 G Gehman, Henry 2, 350–1, 369 Geiger, A. 314 Gentry, Peter 387 Gignac, F. T. 59 Grabe, Johann 305 Graves, M. 265 Greenspoon, L. 163, 382 Grinfield, E. W. 300, 302–3 Guillaume, Philippe 210 H Hanhart, Robert 3, 44–5, 64, 100, 113–14, 118, 133, 315, 333, 335, 375, 385, 387, 390 Hansen, Mosegaard 86 Harl, Marguerite 323, 335, 364, 369, 383 Harris, Rendel 251 Hatch, E. 140, 264, 389 Hauspie, K. 67, 388 Hays, Richard 110, 245–6, 254 Heine, R. E. 265 Helbing, R. 369, 388 Hengel, M. 394 Hess, Richard 364 Hesseling, D. C. 276 Hiebert, R. J. V. 199 Hody, Humphrey 238–9, 307 Holladay, C. R. 240 Holmes, Robert 177–8, 300, 305, 307, 384 Honigman, S. 239 Horrocks, G. 57, 59 Horsley, G. H. R. 66 Hughes, Rebecca 83 Hwang, Shin 83 J Jacques, X. 65 Jarick, John 364 Jellicoe, Sidney 2, 300, 381, 383 Jenkins, G. 199, 202

MODERN AUTHORS

Jobes, Karen 112, 178, 247, 384, 387 Jones, H. S. 66 Joosten, Jan 210, 217, 221–2, 315, 317, 322, 324, 369, 393–4 K Kahle, Paul xiii, 130, 303 Kaminka, A. 314 Kantor, B. P. 60 Karo, G. 126 Karrer, Martin 249, 339–40, 364, 390 Kauhanen, T. 172 Kennedy, H. A. A. 369 Kennicott, Benjamin 303 Khan, Geoffrey 51 Kharanauli, Anna 193 Kielsmeier-Jones, Abram 387 Kiessling, E. 66 Kircher, Conrad 65, 304 Kittel, G. 67 Knoblach, F. W. 60 Konstantinou, Miltiades 296 Korais, Adamantios 287 Kotzé, G. 382 Kraft, R. 393 Krasôvec, J. 60 Kraus, Wolfgang 337, 339–40, 364, 390, 393 Kreuzer, Sigfried 172, 339–40, 384, 393–4 Kristeva, Julia 109 Krivoruchko, J. G. 276 Kroeze, J. H. 87 L Lagarde, Paul de xiii, 2, 64, 130, 133, 158, 161, 172, 289, 300 Lampe, G. W. H. 67 Langacker, Ronald 83 Lange, A. 157 Lange, Armin 221 Lanier, Greg 387 Law, Timothy M. 172, 247, 339, 393–4 Le Quien, Michel 306 Léonas, A. 382 Leaney, A. R. C. 157–8 Lee, John A. L. 66, 131, 350, 369, 393 Leemans, Johan 220 Lemmelijn, B. 316–17, 319–20, 325 Levinsohn, S. 86 Liddell, H. G. 66 Lietzmann, J. 126 Lightfoot, J. B. 366 Lim, T. H. 157

MODERN AUTHORS 479

Longacre, Robert 83 Louw, J. P. 71 Lust, J. 67, 342, 382, 388 M Mandelbrote, Scott 1 Margolis, M. L. 2 Martens, P. W. 265 Martianay, Jean 306 Mason, Steve 219 McCarthy, Michael 83 McLay, Tim 163, 318–20, 394 McLean, N. 1, 45, 375, 384 Mélèze-Medrzejewski, J. 13 Meiser, M. 382 Menken, Maarten 248, 250–1 Miháilá, A. 296 Milligan 66 Miscall, Peter 110 Moore, George Foot 367 Morin, Pierre 300 Morrish, G. 65 Mortari, L. 391 Moulton, Jame H. 66, 369 Mountfaucon, Bernard de 65, 304 Müller, Mogens 247 Munnich, Olivier 194, 335, 364, 369, 383 Muraoka, Takamitsu xiii, 65–7, 70–1, 338, 369, 388–9, 393 Muses, C. A. 390 Myers, P. D. 60 N Najman, H. 234 Naudé, J. A. 87 Nautin, P. 202 Nestle, E. 178–9, 187 Neuschäfer, B. 193 Nida, E. 71 Nobili, Flaminio 306 O Oikonomos, Constantinos 288 Olofsson, S. 315 Orlinsky, Harry 2, 180 Orsini, Fulvio 306 Ottley, R. R. 330, 383 Owen, Henry 307 P Parsons, James 177–8, 300, 384 Penner, K. M. 329, 337–8, 390

Petersen, N. 155 Pezron, Paul-Yves 306 Pierri, R. 369, 388 Pietersma, A. 70, 84, 316–17, 329, 332–4, 345, 364, 390 Plantin, Christopher 302 Porter, Stanley 364 Preisigke, F. 66 Prijs, L. 315 R Rahlfs, Alfred xiii, 2, 44–5, 64, 66, 70, 100, 131, 133, 173, 221–2, 248–50, 264, 289, 333, 335, 375, 385, 387, 390–1 Rajak, Tessa 239, 342 Redpath, H. A. 140, 389 Rehkopf, F. 66–7, 388 Reíllo, Canas 66, 341, 389, 391 Reider, Joseph 166 Rösel, M. 156, 315–18, 320–1, 324–5 Ross, William 114, 387 Runge, Steven 83 S Sáiz, B. 387 Salvesen, A. 165, 393 Scarlata, Mark 80 Schäfer, C. 386 Schaper, Joachim 112, 315–16 Schiffman, L. H. 154 Schironi, F. 199 Schleusner, J. F. 66 Schlund, C. 156 Scott, R. 66 Screnock, J. 142 Seeligmann, Isaac xiii, 315 Sgherri, G. 193 Siegert, F. 383 Silva, Moisés 112, 178, 247, 369, 384, 387 Simon, Richard 303 Skarsaune, Oskar 219 Skeat, T. C. 197, 205 Skehan, P. W. 155–7 Smend, Rudolf 2 Sokoloff, M. 275 Solsalon-Soininen, Ilmari 369 Stanley, Christopher 253–4 Steiner, R. C. 60 Stock, St. G. 369, 388 Sundberg, Albert 209, 218 Swete, Henry xiii, 1, 2, 64, 258, 264, 338, 375, 383–4

480

MODERN AUTHORS

T Tampellini, S. 329 Taylor, B. 388 Tchernerska, N. 275 Teodorsson, S. T. 57, 59 Thackeray, Henry St. John xiii, 112, 162, 369, 384, 389 Thenius, O. 135 Thomson, Charles 345, 390 Thrax, Dionysius 192 Thumb, J. Albert 369 Tilly, M. 384 Tischendorff, Constantin 64, 286, 299 Toury, Gideon 24, 31, 70, 348, 350 Tov, Emanuel 23–6, 38, 66, 114–15, 118, 142, 155–7, 313, 315–16, 319–20, 323–4, 382, 384 Trapp, E. 67 Trommius, Abraham 64, 304 Turner, Nigel 166, 369

Vamvas, Neophytos 287, 295 Vossius, Isaac 303

U Ulrich, E. 155–7 Ussher, James 300

Y Yahalom, J. 275 Young, Patrick 300, 305 Yuditsky, A. E. 60

V van Dale, Anthonie 307 van der Kooij, Arie 239 van der Louw, T. A. W. 115–16, 155, 342 van der Merwe, C. H. J. 87 VanderKam, J. C. 208

W Wagner, Ross 244, 248, 250, 254 Walser, Georg 369 Walters, P. 126, 301, 394 Wasserstein, Abraham 233 Wasserstein, David 233 Wellhausen, Julius 2, 135, 170, 308 Wettstein, Jakob 366 Wevers, John 3, 133, 155–6, 175–6, 191, 194, 278, 346, 351, 367, 370, 372, 382 Williams, Carrin 253 Wooden, G. 393 Wright, Benjamin G. 1, 70, 84, 234, 345, 364, 390 Wright, N. T. 246, 253–4 Würthwein, E. 186

Z Ziegert, C. 324 Ziegler, Joseph xiii, 3, 183, 186–7, 249–50, 315 Zohrapian, H. 178

SUBJECTS

A Akkadian 52 Alexandria 9, 14, 202, 209, 232, 234, 238, 303, 372 alliteration 97, 104–7 allophones 52, 54, 56–7 allusions 110, 245 anthropomorphism 29 anti-Semitism 67–8 apocryphal writings 4, 284 Apostles 257 approximants 54–5, 57 Aramaic 11, 37, 43, 49, 52, 70, 72, 94–5, 101, 130, 141–2, 149, 246 Egyptian 52, 54 square script 113 Aramaisms 70 Association for the Study of the Septuagint in South Africa 394 attestation 47, 110 B Babylonian (language) 55 base text 69–70 Bible authority of 299 Bible, Latin. See Vulgate Bibles Church Slavonic 285, 288–91, 295–6, 373 Gennady 290 King James Version 293, 350, 372 London Polyglot 304 Moscow Greek 285, 290 New King James Version 293 New Revised Standard Version 333, 364 Orthodox Study Bible 293, 329 Ostroh 290 Russian Synodal 291, 295 Zoe 286 bibliographies 381–2 British and Foreign Bible Society 287, 291, 295 Byzantium 271

C Caesarea 269–70 Calvinism 285 canons 68–9, 207–28, 372–5 Athanasius 220 comparisons 211–16 Eastern Orthodox 283–5 Hebrew 218–20, 317 Jewish 207–8 lists 210–17 catchwords 119 Centre for Septuagint Studies and Textual Criticism 395 Christological controversies 283 church confessions 299 church fathers 208, 220, 227, 305–6, 346 Church of England 307 Church Slavonic 284–5, 292–3 codices 48, 68, 70, 126, 210, 221–2 Alexandrinus 45, 64–5, 69, 127, 131, 146, 150, 155, 171, 182, 191, 193, 222, 247–9, 285–7, 292, 304–5, 375 Ambrosianus 278 Bezac 249 Graecus Venetus VII, 273 Leningrad 137 Lugdunensis 172 Sinaiticus 45, 65, 136, 146, 150, 152, 186, 205, 222, 247–9, 299, 304, 375 Vaticanus 1, 3, 41, 45, 124, 130–1, 146, 150, 155, 163, 167, 171, 177–8, 182, 222, 247–9, 286–7, 302, 304, 307, 338 Vaticanus Graecus 343, 273 colophons 194–200 commentaries 3, 345–61, 363–76, 392–3 Patristic 126, 128, 180 concordances 1, 63–6, 389–90 consonants 52–5 Constantinople 269, 273 Council of Ephesus (431) 176–7 Council of Jamnia (Yavneh) 207–8 Council of Laodicea 285

482 SUBJECTS

D Dead Sea Scrolls 14, 54, 113, 125, 135–6, 149–59, 179, 187, 243, 251, 265, 296, 313 Temple Scroll 154 deuterocanonical writings 4, 127, 217, 219–23, 227, 284–5, 288, 291, 374 diachrony 41, 50 dictation 16–17 dictionaries 51, 63 Arukh 276 Bauer, Leander, and Kahle 51 Borée 51 Donner 51 Fox 51 Koehler and Baumgartner 51 Moscati 51 Noth 51 Rechenmacher 51 Zadok 51 discourse grammar 79–92, 371 definition 83 discourse markers 86–8 discourse units 81 dittography 126, 193 E editions 63–4, 123–33, 384–7. See also recensions Aldine (1518) 1, 64, 65, 124, 223 Antiochene 64, 66, 341 Aquila 3, 124, 161, 164–6, 176, 178, 191, 205, 244, 249, 258–61, 270, 275–7, 279–81 Arabic 176–7, 184, 304 Armenian 133, 175–9, 182–3, 186, 188 Athens Septuagint 286 Basel (1545) 64 Bohairic 175, 178 Brenton (1844) 329 Cambridge (1906–40) 1, 69, 125, 166, 177–8, 375, 385 Complutensian (1517) 1, 63, 124, 131, 186, 223, 225–7, 301, 304 Coptic 175–9, 184–5 Ethiopic 133, 175–6, 178, 184, 304 Frankfort (1597) 64, 300 Georgian 176–7 Göttingen 2, 64, 69–70, 113, 125, 133, 137, 146, 150, 152, 166, 175, 177–8, 181, 187, 194, 220, 248–9, 265, 317, 333, 335, 339, 341, 346, 375, 385

Grabe (1707–20) 64 Gutenberg, 124 Holmes and Parsons (1798–1827) 1, 64, 223 Jerome 223 La Bibbia dei Settanta 391 La Bible d’Alexandrie 329, 335–7, 340, 345, 364–5, 390 La Biblia griega. Septuaginta 329, 340–3, 391 Lexham English Septuagint 329, 337–9, 343, 390 London Polyglot (1657) 64 New English Translation of the Septuagint 329, 332–5, 338–40, 342–3, 345, 347, 352, 364–5, 375, 390 Old Greek 79–80, 100–3, 107, 135–47, 150–1, 155–7, 161, 164–5, 167, 170–1, 173, 176–9, 182–8, 202, 231, 236, 239–41, 243–54, 264, 271, 319, 346, 359–60 Old Latin 129, 146, 175–9, 303 Palestinian Aramaic 178 Palestinian Syriac 175 Polyglot of Antwerp (1572) 64 Ra46 225–6 Ra68 225–6 Ra106 224–5 Ra130 224–5 Rahlfs and Hanhart 64, 223, 248, 286, 293, 317, 339, 341, 375, 385 Samaritan 304 secondary 128–9, 132–3, 175–88 Septuaginta Deutsch 329, 338–40, 343, 345, 364–5, 390 Sixtine (1587) 1, 65, 223 Slavonic 177 Strasbourg (1526) 64 Swete (1887–94) 64 Symmachus 124, 161, 164–6, 176, 178, 191, 205, 244, 258–61, 270, 275–7, 281, 304 Syro-Hexapla 175, 178, 184, 195, 205 Theodotion 101, 124, 161, 164–6, 176, 178, 185–7, 191, 205, 244, 246, 258–261, 270, 277, 281, 304 Thomson (1808) 329 Tischendorf (1850) 1, 64 Vetus Latina 223, 249 Zohrapian 182–4 Egypt 9–10, 13–14, 17, 19, 75–6 encyclopedias 63 Epistle of Aristeas 68

SUBJECTS 483

eschatology 30 Ethiopian Manuscript Microfilm Library 133 Ethiopic 48 etymology 38, 48–51, 68 Europe, early modern 299–309 exegesis 37, 60, 131, 258–9, 264–5, 317, 324 Antiochene school 259 contextual 25 Hellenistic 25–33 lexical 25 multicausality 27, 29, 32 theological 25–6 F Forschungskommission zur Edition und Erforschung der Septuginta 395 fragments 274–5, 281 fricatives 54–5, 57 G Geniza texts 272, 274–5 glosses 126, 275, 278, 280–1, 290 grammar 16, 350 Semitic 48 grammars 3, 51, 192, 369, 388–9 Abel 369 Bauer and Leander 51 Beyer 51 Fox 50 Helbing 1, 369 Joüon and Muraoka 51 Nöldeke 51 Rosenthal 51 Sagarin 51 Thackeray 1, 369 graphemes 42, 50, 52 Greek 37, 64, 149 Attic 56, 132, 169, 368 Jewish dialect 70, 350–1 koine 57, 70, 87, 89, 91–2, 94, 274, 331, 334, 336, 369 literature 127, 131, 368 style 93–108 syntax 11, 89–90 Greek, post classical 27, 31 Greek fathers 70 Groupe de Recherche Septante et Critique Textuelle 395 H Halbturn Amulet 60 handbooks 393

haplography 126, 193 harmonization 113 Hasmonean dynasty 49 Hebraisms 70, 94–5, 132, 157, 168–9, 305, 355 Hebrew 10–11, 37, 43, 50, 63, 70, 92, 94, 130, 149, 301, 314 late Biblical 11, 72 liturgical use 272 Middle 280 Mishnaic 54 pre-Septuagint 52 Samaritan 52, 54 syntax 12 Tiberian 52, 54–5, 60 Hellenic Bible Society 288, 296 hermeneutics 111–12, 265, 316, 363 Hexapla 1, 43, 65, 124, 127, 129, 161, 166–70, 172, 176–7, 203, 259–60, 262–3, 270, 277, 279–80, 291, 296, 302–4, 306 Hexapla, Syro 175, 178, 184, 195, 205, 302, 304–5 historic present 90–1 history of scholarship 1–3 homoioteleuton 105 homophones 38 I idioms 11, 16, 22, 72, 75, 91, 305, 331, 350, 354, 360 inflection 47 information structure 89–90 inspiration, divine 257, 261, 265 International Organization for Septuagint and Cognate Studies 2, 67, 329, 345, 364, 395 intertextuality, Biblical 109–19 introductions 2, 3, 383–4 J Jews bilingualism 10–11 diaspora 10, 367 John William Wevers Institute for Septuagint Studies 395 journals 393–4 Judaism ancient 367 education 11 liturgy 11–12

484 SUBJECTS

Judaism, Byzantine 269–81 Judaism, Hellenistic 11, 23, 30, 63, 233, 240, 272, 274, 301, 313–14, 321, 336, 342, 346 Judaism, rabbinic 271, 281 Justinian, I, Emperor 271, 274 K Karaite movement 272, 281 L language, source 70–1 language, target 70–1 Latin 64 Latin Christianity 262–4 Letter of Aristeas 4, 9–10, 12–13, 19, 231–2, 234–6, 238–40, 256, 294, 306–7, 347, 352, 372 lexemes 37, 72, 116 lexicography 60, 68–70, 96, 144, 344, 353 lexicons 3, 63, 66–7, 76, 387–8 Liddell, Scott, and Jones 66 Lust, Eynikel, and Hauspie 67 linguistics 18, 24, 131, 305, 323, 331, 350, 369 literature surveys 382 loan words 10, 38–9, 60, 72 Lucianic tradition 39, 64 Lutheran churches 302, 305 M majuscules 44, 49, 274 manuscripts 45–6, 125, 150, 194–200, 222, 249, 279 Greek 180, 184, 188, 222, 224–6, 253, 346 Latin 152, 184 Slavonic 289 marginal notes 193 Masoretes 41, 43 Masoretic text 66, 125, 131, 150–3, 155, 157–8, 163, 168, 170, 210, 243–4, 248, 250, 263, 266–7, 287–96, 300–1, 303, 320, 323, 337, 353, 374 Mennonites 307 messianism 31 Midrash 132 miniscules 45, 49, 126, 222 Mishnah 52, 54 monograph series 393–4 morphemes 51

morphology 40–1, 42, 44, 46–7, 49–51, 139, 144 N Names divine 28 personal 39–41 place 353 proper 65, 259 nasals 55, 57 neologisms 72–3 New Testament 30 allusions to Old Testament 245–7 citations of Old Testament 243–5, 252–3 relation to Septuagint 243–4 verbal parallels to Old Testament 245–7 New Testament, Greek Nestle-Aland 177, 187, 249 UBS 187, 243, 249 nouns, proper 72 O Octateuch 290 onosmatics 60 oral tradition 37 oral transmission 12 orality 12, 38, 41 Orthodox Church diaspora 293–4 Eastern 283–7, 367, 373–4 Georgian 283 Greek 255, 284–7, 295 Russian 283–4, 290–3 orthography 39, 42–3, 49, 193, 307 ossuaries 60 Ottoman Empire 273 P paleography 49, 126 Palestinian (language) 55 Pan-Orthodox Conference (1962) 285 papyri 19, 66, 68, 72, 76, 125–6, 369 Chester Beatty 182 Egyptian 57, 70, 99–100, 131 Greek 99, 184, 187, 351 Greek magical 60 Zenon 54 parallelism 76 paranomasia 97, 104–7 Patristic writings 255–67, 336, 346 Peshitta 29

SUBJECTS 485

Philippi’s law 43, 49 philology 17–18, 300, 351 phonemes 42, 49, 52, 56 phonetics 72 phonology 37–60 phrasing 97 plosives 52, 54, 57 polyglots 177, 187 Antwerp 177 Complutensian 177 Constantinople Pentateuch 273, 281 Hamburg 177 Heidelberg 177 London 177 Paris 177 Protestant Reformation 302 Proto-Semitic 42–3, 51, 55 Pseudipigrapha 66, 69, 284 Ptolemaic era 256, 271, 336, 340, 351 Ptolemy I, 9, 235 Ptolemy II, 9, 13, 232, 234–5, 306 Ptolemy IV, 236 Ptolemy VI, 233 publications 2 Q Qumran. See Dead Sea Scrolls R recensions 64, 124, 130, 161–73. See also editions Antiochene 128, 131–3, 169 Aquila 64, 66, 115, 124, 151, 161 Constantinopolitan 125, 289 Hesychian 125 Kaige 127–31, 151, 157, 161–4, 171–2 Lucianic 124, 127–9, 131–2, 161, 163–4, 167–72, 182–4, 186, 250, 289 Origen 125 Palestinian 125 Syrian 125 reception 231, 370–2 references 110 Reformed churches 299, 302–3 register (language) 31, 70 research centers 394–5 retroversion 139, 158 revelation, verbal 324 rhetoric 94, 96–7 Roman Catholic Church 299 post-Tridentine 302–3

roots 50, 72 Russian Bible Society 291, 293, 295 S Sadducees 49 scribal choices 247–8 scribal error 247, 321–2 scribes 13–15, 47, 141, 152, 221, 247 scrolls 16, 48 Torah 150 Second Temple period 30, 60, 136, 157–8, 207–10, 231–41 Secunda 60 semantics 72, 79, 82, 131, 143–4, 334, 348, 350, 358 Septuaginta: Vetus Testamentum 2 Seventy(-Two), the 226–8, 232, 256, 261, 263, 265, 271, 291, 347, 351 translation legend 255, 260 sibilants 57 social context 19 societies 2 Society of Biblical Literature 2, 345, 364 software 44–5, 395–6 Accordance 395 CATSS 66, 140, 395–6 Logos 337, 395 STEP Project 45 stirps 50–1 symbols Aristarchan 133, 194, 204 hexaplaric 182 Synod of Constantinople (1642) 285 Synod of Jassy (1642) 285 Synod of Jerusalem (1672) 285 syntax 16, 50, 72, 79, 82–3, 94, 96–104, 108, 116, 139, 143, 305, 336, 338, 348, 353 Syriac 48, 259 T Tannaitic tradition 43, 274, 279 targums 12, 29, 64, 273 testimonia 251 Tetrapla 199, 201, 204 texts ancient Near Eastern 110 texts, source 112, 330–3, 336, 344, 355 texts, target 330 textual additions 96 textual analysis 43–9 textual criticism 48, 123–33, 179, 247, 342 Cambridge school 131

486 SUBJECTS

Hebrew Bible 135–47 Madrid school 131 textual dating 37 textual omissions 96, 126 textual sources 123, 125–6, 136 textual transmission 124, 126, 145, 193, 248–50, 262, 274–5, 279, 316, 321 textus receptus 124 Theological Dictionary of the New Testament 67–8 theology, Christian 313 theology of the Septuagint 313–26 history of 313–17 Thesaurus Linguae Graecae 66 Tiberian tradition 41, 43, 48–9 toponyms 38 transcription 37–9, 44, 59 translation 3, 12 anaphoric 32 consistency 23–4, 276–7, 349 divergence 106–7, 116–17 equivalency 247, 348–9, 353, 357, 360, 371 free 23–5, 31–2, 85, 107, 111, 128, 368 funding 14 harmonization 32 history of 9–20 homophonous 39–40 interference 22, 31, 71–2, 74–5, 348–9, 359 interlinear 18, 84, 333–4, 338, 371 isomorphic 139, 144, 355 Kaige approach 32 legal basis 12–13 literal 23–5, 31–2, 85, 107, 111, 334, 368 literary 359 oral 74 order 114 religious basis 12 segmentation 23–4 technique 9–11, 15–17, 19, 21–33, 85, 94–5, 115, 136, 140, 143, 278–9, 315–16, 322, 330–2, 348, 368

testimony hypothesis 251–2 theories 63, 165 translation influences sociocultural 22, 26, 30, 75–7, 128, 331, 355 theological 22, 25–7, 91, 112–13, 117, 127, 144–5, 247, 260, 263–4, 267, 297, 300, 316, 326, 352, 357, 366 translation studies 18–19, 23 translations 128–9, 150–1, 359–60, 390–2 translators 110–12, 316 Jewish 94 transliteration 38, 142 trills 55 U uncials 126 United Bible Societies 177, 293, 382 V variant readings 124, 267 Vatican II 301 Vetus Latina 130, 133 vocabulary 17–18, 27, 94, 97–104, 114–15, 131–2, 139, 323, 331 vocalization 41, 48–59, 140–1 voice (linguistic) 111–12 Vorlagen 37, 40, 47–8, 65, 75, 77, 79, 96–7, 103, 107, 112, 115, 119, 125, 127, 130, 133, 135–7, 140–2, 146, 153, 156–7, 251, 257, 261, 265, 314–16, 320–5, 337, 340, 346, 354 vowels 56–9 Vulgate 76, 165, 255, 262–4, 284, 290, 294, 296, 301–2, 306 W word choice 24 word equivalency 24 word frequency 247 word order 24, 79, 89, 97, 331, 336 word stems 104