The Qatal//Yiqtol (Yiqtol//Qatal) Verbal Sequence in Semitic Couplets: A Case Study in Systemic Functional Grammar with Applications on the Hebrew Psalter and Ugaritic Poetry 9781463214883

Citation preview

The Qatal//Yiqtol (Yiqtol//Qatal) Verbal Sequence in Semitic Couplets

Gorgias Ugaritic Studies 3   General Editor N. Wyatt

The Qatal//Yiqtol (Yiqtol//Qatal) Verbal Sequence in Semitic Couplets A Case Study in Systemic Functional Grammar with Applications on the Hebrew Psalter and Ugaritic Poetry

Silviu Tatu

Gorgias Press 2008

First Gorgias Press Edition, 2008 Copyright © 2008 by Gorgias Press LLC All rights reserved under International and Pan-American Copyright Conventions. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise without the prior written permission of Gorgias Press LLC. Published in the United States of America by Gorgias Press LLC, New Jersey ISBN 978-1-59333-958-6 ISSN 1935-388X

Gorgias Press

180 Centennial Ave., Suite A, Piscataway, NJ 08854 USA www.gorgiaspress.com Library of Congress Cataloging-in-Publication Data Tatu, Silviu. The qatal//yiqtol (yiqtol//qatal) verbal sequence in Semitic couplets : a case study in systemic functional grammar with applications on the Hebrew Psalter and Ugaritic poetry / Silviu Tatu. -- 1st Gorgias press ed. p. cm. -- (Gorgias Ugaritic studies, 1935-388X) Includes bibliographical references and indexes. 1. Hebrew poetry--History and criticism. 2. Hebrew language--Grammar. 3. Hebrew language--Discourse analysis. 4. Ugaritic poetry--History and criticism. I. Title. PJ5022.T38 2008 892.4’1009 2008003083 The paper used in this publication meets the minimum requirements of the American National Standards. Printed in the United States of America

TABLE OF CONTENTS Table of Contents Preface Acknowledgments Abbreviations and notations General abbreviations Biblical books (when chapter/verse reference is given) Hebrew characters List of tables and figures 1 SETTING OUT THE PROBLEM 1.1. Differentiating Poetry from Prose 1.2. Particularities of the Prose/Poetry Debate in Hebrew Literature 1.3. Features of Hebrew Poetry 1.4. The Verbal Sequence Variation in Hebrew Poetry 1.5. The Psalter as a Corpus of Poetic Texts 1.6. Terminology 1.7. Method 1.7.1. Verse-line Delimitation 1.7.2. Systemic Functional Analysis 1.8. Outline of this study 1.9. Relevance of this Study 1.9.1. Relevance to the Literary Criticism of Hebrew Poetry 1.9.2. Relevance to the Grammar of Classical Hebrew 1.9.3. Relevance to the Exegesis of the Hebrew Psalms as Preserved by the Masoretes 1.9.4. Relevance to Systemic Functional Grammar 1.9.5. Relevance to Ugaritic studies 2 PREVIOUS CONSIDERATIONS OF THE QYYQ VERBAL SEQUENCE 2.1. The Rhetorical Solution 2.1.1. Moshe Held 2.1.2. Mitchell Dahood v

v xi xiii xv xv xvii xviii xix 1 1 3 5 8 12 16 17 17 19 19 21 21 22 22 23 23 25 25 25 27

vi

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

2.1.3. Stanley Gevirtz 2.1.4. Frank C. Fensham 2.1.5. Randall Buth 2.1.6. Wilfred G.E. Watson 2.2. The Pragmatic Solution 2.2.1. Talmy Givón 2.2.2. Michael Rosenbaum 2.2.3. Walter Gross 2.2.4. Nicholas P. Lunn 2.3. Summary

29 29 30 32 34 35 37 40 41 46

3 THE STRUCTURE OF POETIC VERSE AS PRESERVED BY THE MASORETES 3.1. The Masoretic Tradition of Verse-line Delimitation 3.1.1. The relevance of the Masoretic accents 3.1.2. Rules of Engagement with the Masoretic Accents 3.2. The ‘Delimitation Theory’ 3.2.1. Marjo C.A. Korpel and Johannes C. de Moor 3.2.2. William T. Koopmans 3.2.3. Raymond de Hoop 3.2.4. Ernst J. Revell 3.2.5. Jan P. Fokkelman 3.3. Methodological Steps to Approach the Masoretic Text of the Psalms 3.4. The Masoretic Text and Some of Its Modern Interpreters 3.3.1. Julius Wellhausen 3.3.2. Franz Wutz 3.3.3. Oswald Loretz 3.3.4. Hans-Joachim Kraus 3.3.5. Peter C. Craigie, Marvin E. Tate and Leslie C. Allen 3.3.6. Pierre Auffret 3.3.7. Raymond J. Tournay 3.3.8. Tertulian Langa 3.4. Summary

47 48 48 52 55 56 58 58 59 60 64 66 66 66 67 67 69 71 72 73 74

4 SYSTEMIC FUNCTIONAL GRAMMAR: A READER 4.1. Principles of the Systemic Theory 4.2. Thematic structure 4.2.1. The System of Information 4.2.2. The System of theme 4.2.3. Theme in declarative clauses 4.2.4. Theme in interrogative clauses 4.2.5. Theme in volitive clauses 4.2.6. Textual, Interpersonal and Topical Themes 4.2.7. Equative Constructions and Predicated Themes

75 76 85 85 87 90 90 91 91 93

TABLE OF CONTENTS

vii

4.2.8. Theme in bound, minor and elliptical clauses 4.3. Mood Structure 4.3.1. The Mood Element 4.3.2. The Residue Element 4.3.3. Mood in Declarative Clauses 4.3.4. Mood in Interrogative Clauses 4.3.5. Mood in Volitive Clauses 4.3.6. Mood in Minor and Elliptical Clauses 4.3.7. Extensions of Mood Analysis 4.4. Transitivity Structure 4.4.1. Transitivity Structure of Material Clauses 4.4.2. Transitivity Structure of Mental Clauses 4.4.3. Transitivity Structure of Relational Clauses 4.4.4. Transitivity Structure of Other Types of Clauses 4.4.5. Circumstantial Elements 4.4.6. Ergativity as an Alternative Interpretation to Transitivity 4.5. Logical Structure 4.5.1. Interdependency 4.5.1.1. Parataxis 4.5.1.2. Hypotaxis 4.5.2. Logico-semantic Types: Expansion 4.5.3. Logico-semantic Types: Projection 4.6. Summary

94 95 96 100 103 104 105 106 108 109 113 115 116 122 125 128 132 133 134 135 136 144 151

5 A SYSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW 5.1. Thematic Structure 5.1.1. Theme in declarative clauses 5.1.2. Theme in interrogative clauses 5.1.3. Theme in volitive clauses 5.1.4. Textual and interpersonal Themes 5.1.5. Equative Constructions 5.1.6. Theme in Bound, Minor and Elliptical Clauses 5.2. Mood structure 5.2.1. The Mood Element 5.2.2. The Residue element 5.2.3. Mood in declarative clauses 5.2.4. Mood in interrogative clauses 5.2.5. Mood in volitive clauses 5.2.6. Mood in minor and elliptical clauses 5.2.7. Extensions of Mood analysis 5.3. Transitivity Structure 5.3.1. Transitivity structure of material clauses 5.3.2. Transitivity structure of mental clauses

153 153 156 162 165 167 176 178 181 183 189 191 194 198 200 203 205 205 209

viii

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

5.3.3. Transitivity Structure of Relational Clauses 5.3.4. Transitivity structure of other types of clauses 5.3.5. Circumstantial elements 5.3.6. Ergativity as an alternative interpretation to transitivity 5.4. Logical Structure 5.4.1. Interdependency (taxis) 5.4.2. Logico-semantic type: expansion 5.4.3. Logico-semantic type: projection

212 218 223 231 240 241 249 265

6 PSALM COUPLETS WITH QYYQ VERBAL SEQUENCE: A SFG INVESTIGATION 6.1. Thematic Structure of Psalm Couplets with a QYYQ Verbal Sequence 6.1.1. Psalm 38 6.1.2. Evaluation of the database 6.2. Mood structure of Psalm couplets with QYYQ verbal sequence 6.2.1. Psalm 63 6.2.2. Evaluation of the database 6.3. Transitivity structure of Psalm couplets with QYYQ verbal sequence 6.3.1. Psalm 26 6.3.2. Evaluation of the database 6.4. Logical structure of Psalm couplets with QYYQ verbal sequence 6.4.1. Psalm 7 6.4.2. Evaluation of the database

277 277 277 289 291 291 298 301 301 307 310 310 316

7 THE CASE OF UGARITIC POETRY 7.1. The Ugaritic text 7.2. The poetic quality of Ugaritic narrative poems 7.3. The Ugaritic database 7.4. The SFG analysis of UPy couplets with QYYQ verbal sequence 7.4.1. Thematic f-structure of Ugaritic couplets with QYYQ verbal sequence 7.4.2. Mood f-structure of Ugaritic couplets with QYYQ verbal sequence 7.4.3. Transitivity f-structure of Ugaritic couplets with QYYQ verbal sequence 7.4.4. Logical structure of Ugaritic couplets with QYYQ verbal sequence 7.5. Summary

319 322 323 327 333 333 334 336 336 337

8 CONCLUSIONS 8.1. Contributions of this research 8.2. Resources for further research 8.3. Recommendations for further research

339 339 343 343

Bibliography APPENDICES Appendix 1: Functional structure of Hebrew Psalms couplets with q*f~l//y]qf{l verbal sequence

345 379

TABLE OF CONTENTS

ix

Appendix 2: Functional structure of Hebrew Psalms couplets with y]qf{l//q*f~l verbal sequence 439 Appendix 3: Authorities Consulted for Verse-lines Delimitation in Psalms 481 Appendix 4: Relations among verb forms, Subject and Marked Theme 491 Appendix 5: Functional structure of Ugaritic couplets with QYYQ verbal sequence 499 (a) q*f~l//y]qf{l 499 (b) y]qf{l//q*f~l 506 Appendix 6: Archaic HPy couplets with QYYQ verbal sequence 515 (a) q*f~l//y]qf{l 515 variant 515 (b) y]qf{l//q*f~l 516 Appendix 7: Distribution of couplets in Psalms 517 Appendix 8: Variations of verses with the QYYQ verbal sequence 531 Distribution of couplets with a QYYQ verbal sequence on one of its lines 531 Distribution of tercets with a QYYQ verbal sequence base 531 Appendix 9: An overview of verse-line delimitation in Psalms as preserved by the Masoretes and compared with Ancient, Medieval and modern authorities 533 Index of subjects 541 Index of authorities 545 Index of biblical references 549 index of ugaritic references 561

PREFACE The substance of this book is my PhD thesis defended in July 2006 in Oxford at Oxford Centre for Mission Studies. The idea of this research took shape during many years of interaction with the poetic literature of the Semites in general and the Hebrew Psalms in particular, in the context of translation work prompted by the involvement in a new project of providing a fresh translation of the Bible into Romanian by International Bible Society. The object of this study is represented by the q*f~l//y]qf{l (y]qf{l//q*f~l) verbal sequence, previously known as ‘tense shifting’, as found in couplets of the Hebrew Psalter. One hundred and two couplets of this sort were found in the emerged database. This study attempts to offer an innovatory explanation for such an irregular use of verbal forms in the Hebrew Psalter by means of the M.A.K. Halliday’s Systemic Theory. The data displays evidence both of similarity and of dissimilarity between the two verse-lines of the couplets. Arguably, similarity is achieved at multiple levels. Both verse-lines seem to prefer a similar structure in terms of thematic markedness, mood, Process type, and clause type. It is argued that dissimilarity occurs at various levels. At the thematic level, a qatal line with a regular Theme is more likely to trigger a Marked Theme in the yiqtol parallel line than vice versa, regardless of which line it occurs in the couplet. At the interpersonal level, an explicit Subject in the yiqtol line is most likely to be marked. A Marked Theme with a yiqtol in line A is more likely to trigger an explicit Subject functioning as Marked Theme in line B than vice versa. At the experiential level, the functional structure of these couplets occasionally displays verse-lines with a middle clause as pairs for lines realized by operative clauses, yiqtol verbs being more likely to accompany the former. In the couplets investigated from Ugaritic poetry, yiqtol verbal forms promote stability and familiarity, whereas qatal verbal forms are closely related to marked Themes, Subject change, Process type variation, and irregular clause complex structure.

xi

ACKNOWLEDGMENTS I express my gratitude to my wife Cornelia for all the emotional support during many years of part-time research, as well as for the financial support she provided diligently during the years of full-time research. Uprooted from Romania, leaving behind friends and family in order to share the desert experience of an immigrant with their father, our daughters were always supportive. This study is warmly dedicated to them. My thanks are due to Revd Dr Emil Bartoş for endorsing my research venture since an undergraduate student, motivating me along the way, frequently expressing his faith in my abilities to complete it. The Principal of Institutul Teologic Penticostal (Bucureşti, Romania), Revd Dr John Tipei, was instrumental in providing the two-year leave from the academic life, thus enabling me to complete my research. Revd Dr Paul Fiddes, Principal at Regent’s Park College in Oxford, had provided for our family whilst in the beginning years of research in Oxford. Our thanks to him are long overdue. We are grateful to God for the Brethren community at James Street Church in Oxford, who offered us a home, a sanctuary and a family abroad during the last two years. Mr Martin Daubney, Mr Brian Davies on behalf of Philadelphia Trust, and Langham Trust proved themselves as very good friends in times of need, covering all academic and travel expenses over the years. Their contribution stands as a friendship memorial. Oxford Centre for Mission Studies, with its research oriented facilities and always-supportive personnel, generously provided the context, the means, and the resources to enhance the sharpness of my argument and bring to completion an exhausting research. Sitting with researchers such as Dr Carl Armerding, Dr Bernard Farr, and Dr Ben Knighton always generated stimulating discussions. Professor John Day (St Margaret’s Hall, Oxford) opened my eyes for the great world of critical studies on the Psalms. Dr Nicholas Wyatt (New College, Edinburgh) was responsible for introducing me to the language and literature of Ugarit and for initiating the xiii

xiv

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

contact with Dr Wilfred G.E. Watson (Newcastle upon Tyne). I am indebted to Dr Watson, who has midwifed my development as a researcher, for painstakingly reading and correcting hundreds of pages, sometimes at short notice, and for graciously putting up with my avalanche of emails. Dr John Callow, Kathleen Callow, and Eileen Edmondson took over the tiring experience of thoroughly proofreading what seemed to be an English manuscript of this thesis with a keen eye to linguistics which they master. Whatever good comes out of this thesis, I owe it to all of them.

tyb hnby-al hwhy-~a wb wynwb wlm[ awv ry[-rmvy-al hwhy-~a (Psalm 127.1) rmwv dqv awv

ABBREVIATIONS AND NOTATIONS

GENERAL AB BREVIATIONS A

a

Adj aff AH ANE Beh Benef ch./chs. cog Compl ed./eds. Exis Fin GHB GNV Gr HB Heb HPy Hi Hit Ho infa infc Interp JPS á Lat LH Mat

Aleppo Codex Codex Alexandrinus Adjunct affection, subtype of behavioural and mental Process Ancient Hebrew Ancient Near East Behavioural, Process type Beneficiary chapter/chapters cognition, subtype of behavioural and mental Process Complement editor/editors Existential, Process type Finite Ginsburg Hebrew Bible Geneva Bible Version, 1599 Greek Hebrew Bible Hebrew Hebrew Poetry h]pu'l h]tP~u}l h"[u~l infinitivum absolutum infinitivum constructum Interpersonal Theme Jewish Publication Society version of the Old Testament, 1917 The Old Latin version of the Old Testament Latin Late Hebrew Material, Process type xv

xvi Men Mo-od MS/MSS MT MTheme NAB NAS Ni NIB NJB NKJV NLT nom. NRSV OT p (p) per Phenom Pi Pred Pro-cess ps./pss. ptcp Pu ä QYYQ Rel:Att Rel:Id Res Res-idue RFSO RSV ã SH Ö Ä SFG Subj å Text TNK Top

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS Mental, Process type Split Mood manuscript/manuscripts The Masoretic Text MARKED Theme New American Bible, 1991 New American Standard, 1977 n]pu~l New International Version - British edition, 1984 New Jerusalem Bible, 1985 New King James Version, 1982 New Living Translation, 1996 nominal clause New Revised Standard Version, 1989 Old Testament page (s) perception, subtype of behavioural and mental Process Phenomenon P]u}l Predicator Split Process psalm/psalms participial clause (nominal clause with a participle predicate) P%u~l Manuscripts found at Qumran and in the Judaean Desert q*f~l//y]qf{l (y]qf{l//q*f~l) Relational:Attributive, Process type Relational:Identifying, Process type Residue Split Residue Rhythmic Fluency Structurally Obscured Revised Standard Version Peshitta, the Syriac version of the Old Testament Standard Hebrew Septuagint, the Greek translation of the Old Testament (knowing many versions) The Samaritan Pentateuch Systemic Functional Grammar Subject Targums, the Aramaic version of the Old Testament Textual Theme The Holy Scriptures according to the traditional Hebrew Text (Masoretic), 1985 Topical Theme

ABBREVIATIONS

trans Ug UPy ç v./vv. ver Ver Voc / // ║ ║│ [] * ÆØ > > √ = + × “ ‘ ^

xvii

translated, translator Ugaritic (language) Ugaritic poetry Vulgate verse/verses verbal, subtype of behavioural and mental Process Verbal, Process type Vocative half line parallel line end of clause simplex end of clause complex reconstructed text reconstructed form embedded clause embedded clause transitional form deriving into grammatical root elaborating, sub-type of extension expanding, sub-type of extension enhancing, sub-type of extension locution, sub-type of projection idea, sub-type of projection adjacency of constituents

BIBLICAL BOOKS (WHEN CHAPTER/VERSE REFERENCE IS GIVEN) 1-2 Chr 1-2 Kgs 1-2 Sam Cant Dan Deut Eccl Esth Ezek Exod Gen Hab Hag Hos Isa Jer Josh Judg

1-2 Chronicles 1-2 Kings 1-2 Samuel Song of Songs Daniel Deuteronomy Ecclesiastes Esther Ezekiel Exodus Genesis Habakkuk Haggai Hosea Isaiah Jeremiah Joshua Judges

xviii

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Lam Lev Mal Mic Nah Neh Num Obad Ps Prov Zech Zeph

Lamentations Leviticus Malachi Micah Nahum Nehemiah Numbers Obadiah Psalms Proverbs Zechariah Zephaniah

HEBREW CHARACTERS Character

a b B g G d D h w z x j y k K l

Transliteration

a b B g G d D h w z j f y k K l

Character

~m !n s [ p P c q r f v t T : } h'

Transliteration

m n s u p P x q r c v t T ^ * >

Character

, E y,yE i yÅ ' ™ A u W . / | }

Transliteration

# @ ? ! ' " ) o % W + $ ( &

ABBREVIATIONS

xix

LIST OF TABLES AND FIGURES Table 3.1: Table 3.2: Table 4.1: Table 4.2: Table 4.3: Table 4.4: Table 4.5: Table 4.6: Table 5.1: Table 5.2: Table 6.1: Table 6.2: Table 6.3: Table 6.4: Table 6.5: Table 6.6: Table 6.7: Table 7.1: Table 7.2: Table 7.3: Table 8.1:

Hierarchy of Masoretic accents (Price 1990, 168) Master disjunctive accents and their slave conjunctives (Price 1990, 169) The unmarked choice of mood in English Textual and interpersonal Themes (IFG, 79 Table 3(3)) Speech functions and responses (IFG, 108, Table 4(1)) Marking person and polarity by intonation in imperative clauses (IFG, 138-9) Categories of relational clauses (IFG, 216, Table 5(12)) Nouns of projection and nouns of fact in English (IFG, 469, Table 7(25) The unmarked choice of mood in SH The order of constituents according to the mood type of clauses in SH Thematic f-structure of Psalm couplets with QYYQ verbal sequence Distribution of Marked Theme in relation to Mood constituents in Psalm 63 Distribution of Subject in couplets with QYYQ verbal sequence Distribution of Process type in Psalm 26 Distribution of voice in Psalm 26 Distribution of middle voice in couplets with QYYQ verbal sequence Distribution of clause nexuses in our database according to taxis and logico-semantic type Distribution of Ugaritic couplets with QYYQ verbal sequence according to their thematic f-structure Distribution of clauses in Ugaritic couplets with QYYQ verbal sequence according to the quality of their Subjects Distribution of clause nexuses in Ugaritic couplets with QYYQ verbal sequence according to their logico-semantic relationships Distribution of couplets with QYYQ verbal sequence in the Hebrew Psalter

52 53 89 91 95-6 106 119 150 156 182-3 289-90 297 299-302 306 306 308 316-7 333 335 337 339

xx

Figure 4.1: Figure 4.2: Figure 4.3: Figure 4.4: Figure 5.1:

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

The system network of Mood in English (IFG, 23 fig. 1-9) Stratification (IFG, 25 fig. 1-10) Word classes recognized in a functional grammar of English (IFG, 52 fig. 2-8) Process types in English (IFG, 172 fig. 5-2) The network of systems of Agency, Ranging, Voice and Agentivity

79 79 82 111 240

1 SETTING OUT THE PROBLEM Reading Hebrew literature intelligibly, and for that matter any literature, is a painstaking exercise, involving interdisciplinary knowledge, exquisite sensitivity and elevated literary ability, to name just a few qualities. One does not perceive form without content, meaning being a natural outcome of the balanced observation of the two.1 Nevertheless, besides the text itself (utterance), the hermeneutical circle comprises an encoder (writer), a decoder (reader), and the proper response of the decoder to the encoder’s stimuli. All these elements are important for the interpretation of the text. This theory of the process of communication guarantees that our expectations of finding meaning in literary forms may be met, if our response to the original stimuli is sensitive to the author’s intention.

1.1. D IFFERENTIATING POETRY FROM PROSE The general distinction between POETRY and PROSE leaves disregarded a large diversity of texts. Critics generally agree on the matter that if there is such a thing as literature, there have to be at least two main literary genres, PROSE and POETRY. Both of them, whether oral or written, are responsible for creating a specific effect on the audience. Judging them on the level of aesthetics, POETRY is generally associated with subjectivity, vivid expression and atemporality, whereas PROSE is completely the opposite. Theoretically, however, defining POETRY as opposed to PROSE proves to be rather a difficult task. This delimitation proves to be extremely difficult in the postmodern context, where relativism replaced standardization, and deconstruction became the norm (cf. Jakobson 1987, 368 ff.). Aviram avoids the traditional definition of aesthetics, refraining from defining pure categories of POETRY and PROSE. Instead, he promotes 1

This section was previously published in Journal for the Study of Religions and Ideologies 17 (2007): 47-68. 1

2

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

and PROSE as hypothetical directions (Aviram 1994, 44f). The two opposite extremes are by themselves theoretical probabilities and factual impossibilities. Both of them represent asymptotic ends.1 Consequently, at one end of the spectrum lies ordinary language which is supposed to be transparent, i.e., it draws the reader’s attention not to its rhetorical features (style, images, figures), nor to its formal features (how it sounds), but rather to its thematic content. Expository and scientific writing share this quality. At the other end of the spectrum is opaque language, i.e. language with a strong interest in sounds, and therefore very difficult to understand. One can discover its meaning only in the larger context of a conventional system of signs with potential meaning (Aviram 1994, 49).2 Although he admits a diachronic evolution of POETRY, Aviram is promoting a unique organizing principle that of rhythm, with particular embodiments throughout history, because the concept of POETRY is conditioned temporally. Jakobson militates for a more liberal, though holistic, approach to POETRY, according to which poeticity is not to be reduced mechanically to its components. Instead, he invokes emotivity and opacity as the main characteristics of such texts (Jakobson 1987, 378). It is obvious that both ‘emotivity’ and ‘opacity’ display a high degree of subjectivity. As we relate these observations to ANE literature, if the means to stir emotions in the mind of an Ancient Semite were different from those of a modern human, what were they? If rhythm as a state of mind was embodied in a particular way in ANE literature, how was it done? The matter of the conscious use of structures by the ancient poets, noted by modern scholars, remains with the purposive value of poetic language (Jakobson 1987, 250-61). Spontaneous creation is not excluded, particularly where revelatory texts are concerned, but this cannot restrict us from consciously and painstakingly undertaking the mission of careful analysis of the texts. POETRY

1 By doing so, Aviram hopes ‘to isolate the qualities that serve as criteria for what is more poetic or less poetic,’ so that he may formulate a definition of POETRY (Aviram 1994, 46). 2 One can render the binary opposition transparent : opaque as the two extremes of a bipolar spectrum as an expression of the dual nature of the linguistic sign, as first described by de Saussure. Transparent language focuses our attention on the signified, whereas opaque language focuses it on the signifiers, which always mean more than just one thing. Signifiers are bound together into a network of contexts, and can easily lose their meaning, reverting to meaninglessness.

SETTING OUT THE PROBLEM

1.2. PARTICULARITIES OF THE PROSE/POETRY DEBATE IN HEBREW LITERATURE One scholar who applied the idea of binary scale to biblical criticism was Tremper Longman III (1987, 120-21), who concluded that POETRY is characterized by a higher level of artistry than PROSE.1 Robert Alter subscribes to the idea that Hebrew literature consists of two main genres, PROSE and POETRY, to which he actually dedicated two parallel volumes, The Art of Biblical Narrative (1981), and The Art of Biblical Poetry (1985). In the case of HPy, he noticed a particular lack of epic works extant in all ANE literatures except Hebrew.2 Alter believes in two pre-existing literary forms, PROSE and POETRY, of which PROSE represented a continuous attraction for the Hebrew bards ever trying to render its characteristics into poetic lines. Narrativity lies latent in the Hebrew poems, in their minute articulations, from line to line, from unit to unit, articulation mainly generated by synonymity. This sense of articulation may be defined as consequentiality both in logical and temporal terms.3 Alter also tried to propose a few elements marking POETRY as opposed to PROSE. Dialogue is considered a standard practice of PROSE, which defines the action and relation between the actors. In POETRY, the characters have a rather emblematic presence without interacting with one Longman defines artistry in terms of rhetoric. He identified three synthetic characteristics of HP as present in all works of Hebrew literature, i.e. parallelism, imagery and terseness. One may reproach the school of New Criticism for mixing rhetorical devices with stylistics and structuralism (O’Connor 1980a, 10). 2 The historical Psalms (e.g. 78, 105, 106) are just a few exceptions catalogued more precisely as ‘catechistic rehearsals of Israelite history’, namely doctrinal material which has no meaning in itself unless the historical version is known (Alter 1990, 27). However, Alter is more interested in tracing the characteristics that unite them, namely parallelism, narrativity and the use of keywords. 3 See also Alter’s explanations on David’s Song (2 Sam 22), Joel 2 and Deborah’s Song (Judg 5). Although narrative is ignored in POETRY, a sense of narrativity is still extant by the fact that metaphors are given a strong narrative realization (Alter 1990, 40). For example, ‘the right hand’ is a metaphor for authority but in Judges it is acted by Jael, who with her ‘right hand’ killed Sisera. Similarly, ‘the way’, a stereotypical metaphor for the moral life in wisdom literature, is performed literally in Proverbs 7 by a young man going astray to the prostitute (Alter 1990, 45, 61). 1

3

4

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

another in terms of dialogue (Alter 1990, 48-9).1 POETRY is denser than PROSE stylistically, but the distinctive marks of the poetic sub-genres refer to other criteria: the quality of the hearers (fictitious vs realistic), tone (hortative vs vocative), hermeneutic value (individualistic vs archetypal). Wilfred G.E. Watson admits that poetic devices can be met in PROSE too (Watson 1984a, 44-62), but their presence in POETRY is higher in density and they even interplay there with one another. POETRY can be defined in opposition to PROSE (negative indicators) or by an analysis on its own merits (positive indicators, of which there are three categories: broad, structural and other, see infra). Unexpectedly, at the end of a very convincing discourse on the multiplicity of facets HPy has, Watson refers to the factor of content in ANE POETRY. Thus, PROSE is the vehicle used to render the language of letters, treaties, economic documents, etc., and POETRY is preferred only for religious and mythological texts (Watson 1984a, 60).2 Despite the detailed technique promoted in his books, he is far from being dogmatic about the procedure proposed. Therefore, he declares: Ultimately, the decision owes a great deal to mature reflection which will consider content as well as form, with an eye on traditions both in Classical Hebrew and in ancient Near Eastern literature generally. (Watson 1984a, 55)

Alternatively, one can investigate HPy in comparison with POETRY of Northwest Semitic origin or from world literature, looking for common features, provided the poetic features of that literature are clearly understood in their particular context. Invariably, one has to rely on the ability to identify and explain poetic devices specific to each piece of literature. The following section surveys several of the most significant contributions to the literary criticism of poetic devices specific to Hebrew literature. 1 Although imaginery, a short dialogue can still be traced in Judg 5 as if maids are responding to their mistress, Sisera’s mother. Furthermore, dialogue can be encountered in prophetic POETRY. For types and discussion of this ‘technique of style’, see Alonso-Schökel (1988, 170-9). The personal participation of the characters in Job and Song of Songs are not considered literary dialogue, but rather monologues. Similarly, there are some liturgical responsorial pieces in Psalms. 2 A similar association between form and social function is rendered later, but asserting that myths, wisdom literature, and liturgical texts are cast in forms that better reflect the interest of their respective authors in a more ornamented use of language (de Moor & Watson 1993, xvii).

SETTING OUT THE PROBLEM

1.3. FEATURES OF HEBREW POETRY Modern research has promoted various poetic devices as significantly important, even decisive, for the description of HPy. Robert Lowth’s theory of parallelism in the prophetic Hebrew literature (Lowth 1753), so influential for centuries, has lost its grip on literary studies of Hebrew literature.1 With George B. Gray (1915), parallelism received a sharper definition, making room for several specific categories, such as complete/ incomplete and alternative parallelism. Hrushovski (1971, 1202a) redefined it as ‘free-rhythm’, loose enough to accommodate various categories. The syntactical theory of parallelism O’Connor (1980a) promoted does not seem to have gained overall scholarly support. It would be very difficult to prove that parallelismus membrorum functions at the syntactic level in HPy. His theory was applied first to prophetic texts, which are well known as being inspired by the true HPy found outside the Canonical oracles. Despite the severe criticism received (Kugel 1981, 317; Miller, P.D. 1983; Geller 1982, 69-70), O’Connor’s idea that Hebrew verse is organized according to some principles that surpass structural and poetic language is of value. Although they escape the reader’s grasp, we are still able to locate them and even apprehend them, without speculating on their formal definition and function (O’Connor 1980a, 13-4). Lowth’s theory underwent a severe asault by James L. Kugel. Although he declared himself anti-Lowthian,2 others give him credit for producing a newer, more refined, definition of parallelism (Berlin 1985, 64), so much so that Landy (1984, 62-3) even sees in him the ‘saviour’ of the most critized section of Lowth’s theory, that on synthetic parallelism. What Kugel really did was to prove that parallelism is not a poetic device but rather a mental paradigm, which stimulated the creative energies of the authors, under the pressure of tradition and cultural interchange.3 1 A complementary reading of O’Connor (1980), Kugel (1981) and Berlin (1985) can offer a panoramic history of parallelism. 2 Alter evaluates Kugel’s contribution as ‘a bold step forward, together with a giant step backward, in understanding the nature of biblical POETRY.’ Kugel’s sarcasm, enjoyed by Landy, is registered as stubborn resistance by Alter (see Alter 1998, 226-7). Kugel agrees with Lowth only where he was most vehemently criticised, on synthetic parallelism, but only after preliminary redefinitions had taken place (Alter 1985, 12ff). 3 This perception coheres with Meschonnic’s distinction between the Hellenistic perception of POETRY, where reason has priority over rhythm and the

5

6

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Kugel (1981, 26-8) admits that the three main characteristics of UPy, namely frequent ternary rhymes, repetition of words and/or phrases, and word-pairs, represent fertile soil for searching for poetic devices in HPy, but he left it as a mere theory. Once believed to be uniquely characteristic of HPy, parallelism is nowadays accepted not only as a traditional rhetorical device present in some classic literature, but also as a common feature of literary writing. Whether quantitative in terms of metre and elements of clause structure, or qualitative in terms of phonetics and semantics, parallelism is traced in much of world literature. Since Hebrew verse can hardly exist without it, it was inferred that in ancient literatures such as Hebrew and Ugaritic, parallelism is not a mere rhetorical device but an organizing principle that ties together two clauses in a meaningful unit. Upon admitting a connection between POETRY and music and an interconnection between neighbouring cultures, Abraham M. Habermann (1972, 670-93) agreed that HPy owes much to Canaanite aesthetics. As main characteristics, Habermann listed action, imagery, simplicity, vigour, and concreteness, which are portrayed, due to the extreme concreteness of the Hebrew language, by parallelism, strophe structure, metre, genre, and style. Although Habermann’s approach to Hebrew poetic devices is diachronic, it resembles more of an outsider’s approach offering a good and cautious resumé of the status quo in research. Although parallelism of thought is the highly esteemed device of HPy, there were many scholars along the years who suggested that metre or rhythm described HPy, be it number of syllables or number of words in each line.1 Christensen (1986, 62-3) promotes an even older method, that of counting ‘morae’ (i.e. vowels) and disjunctive accents. As Korpel (1998) exemplifies from the book of Isaiah, apparently the Hebrew verse-line is made of two or three words, which give an average of eight syllables. This process seems to be hindered by some irregularities of the Masoretic system, such as the frequent use of the conjunctive m*Q}p, Jewish perception of POETRY, which is intrinsically rhythmical, as codified by the Masoretic accents (Meschonnic 1985, 466-75). Kugel, however, holds a contradictory opinion on the role of Masoretic accents in Judaism, namely, that it obscures the structure of the Hebrew verse, contributing to the forgetting of parallelism (Kugel 1984, 111-6). 1 For a criticism of the classical theories of stress counting of Ley-Sievers see W.H. Cobb, A Criticism of Systems of Hebrew Metre (Oxford: Clarendon; 1905, 111-28) or Gray (1915, 123-54).

SETTING OUT THE PROBLEM which permitted them to connect two or more words into one single stressed unit. Again, far from being a tool enabling the scholar to emend the transmitted text metri causa, the rhythmical balance may sometimes function as an additional argument in the delimitation of cola and verse-lines. (Korpel 1998, 12)

In terms of syllable counting, Freedman and Geoghegan discovered that, in the case of the first three chapters of Lamentations, the average couplet length is 12 syllables, while line A has an average of 7.2 or 7.3 and line B an average of 5.7 or 5.8 (Freedman & Geoghegan 1999, 241).1 Christensen himself concluded that the careful assessment of verselines in relation to their ‘morae’ and disjunctive accents indicates a balanced and rhythmical construction of verses (1986, 68 and again 1988, 35). Watson’s contribution to the description of POETRY in ANE literature is extremely valuable. The numerous indicators categorised into broad and structural types are more or less members of two general categories, namely, conciseness and parallelism, and the result of their interplay. Although they vary in terms of the grammatical level they appear at, from phonetics, to morphology, word order and semantics, parallelism is assumed as the fundamental principle that binds smaller units into larger ones. Watson’s arsenal of poetic devices includes ellipsis, unusual vocabulary, conciseness, unusual word-order, archaisms, metre and rhythm, regularity and symmetry (as broad indicators), parallelism, word-pairs, chiastic patterns, envelope figures, break-up of stereotyped phrases, gendermatched parallelism, tercets (as structural indicators), rhyme and sound patterns (as other) (Watson 1984a, 44-62), as well as numerous figures of speech (Watson 1984a, 251-348). Some of these features resurface intermittently in his later studies. Of particular interest for the diachronic study of POETRY in ANE literature is the recognition of particular markers of poetical units or subunits. Among them, de Moor and Watson (1993, xv) include colometric orthography of some Ugaritic and Akkadian documents, or the accents, the traditional division into P=tWjot and s=tWmot, and the use of s=l> in the Hebrew documents preserved by the Masoretes. 1 The search was done on a peculiar type of text, with a specific rhythm (3 + 2), also known as Qina, but the measures of the first verse-line match the average length of a verse-line as discovered by others. This theory is not unanimously accepted among scholars, though (cf. de Hoop 2000c).

7

8

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

More recently, Alviero Niccacci (1997) proposed three characteristics as specific to HPy: (1) segmented communication, (2) parallelism of similar bits of information, and (3) a non-detectable verbal system. Niccacci’s overall conclusion is that HPy is structured at the most basic level in parallel lines, arranged mainly grammatically in direct order (a b // a´ b´) or in chiastic order (a b // b´ a´).1 So far, sufficient evidence could not be found to support the promotion of a particular poetic device as definitive for HPy. Therefore, the poetic qualities of texts in Hebrew literature are still promoted heuristically, and the categories of POETRY and PROSE are used only as qualifiers for those literary texts that manifest an obvious tendency towards opacity or transparency, respectively. A sharp distinction between the two types of texts is less probable when authors employed a mixture of devices to serve their particular purposes.

1.4. THE VERBAL S EQUENCE VARIATION IN HEB REW POETRY Unlike narratives, poems allow a more flexible use of tense in verbs, which can often frustrate the search for its intended meaning. The translators of HPy will find this situation especially difficult to handle. For instance, Psalm 6.10 says ‘The Lord √hear (q*f~l) my plea // the Lord √answer (y]qf{l) my prayer’.2 Most English translations consulted prefer to render the first lexical verb as a past perfect, with the exception of TNK, which prefers the simple present instead. Most translations continue by rendering the second verb as the habitual simple present (RSV, NRSV, NAB, NAS, NIB, NIV, ESV), although some opted for the future tense (KJV, NLT, NKJV, NJB, GNV), and a few others preferred the simple past (JPS, TNK), or even the past perfect instead (BBE). It is obvious from the above selection that there is a clear lack of consensus in relation to how it should be translated. Whereas the large 1 The choice of illustrating such a method with Hebrew proverbs, we suspect, is not the best choice since the canonical book of Proverbs, especially the last twothirds from which it is quoted, except Prov 6.8 (p. 86), is by its own nature composed of binary-structures with compositional unity and self-sufficient content. Niccacci is actually recognizing the limits of this application rather indirectly (1997, 89 n.52), but seems not to be aware of the lack of representativeness of his selection for the method he is proposing. 2 We prefer the more neutral q*f~l and y]qf{l for what traditional grammar know as perfect and imperfect verbal forms, respectively, and will be used as such henceforth.

SETTING OUT THE PROBLEM majority of translations agree on translating the q*f~l verb form as a past perfect, the consensus is lost when translating the verb of the second verseline. However, it is the case that y]qf{l verbal forms are more open to free translation, since they have the ability to function in various ways. Despite his long experience as a scholar within the field of HPy and translation work, Wendland refrains from being dogmatic about translating verbal sequences. Variations in verb usage, both in tense/aspect and voice, for Wendland (2002, 167-8) reflect a situation common to HPy, whereby a ‘logic’ different from that of the PROSE texts inspired the original poets. As a general rule of thumb, he proposes that verbs are to be translated in relation to their near context, that is the strophe, and to their larger context, that is the poem. The presence of temporal adverbs or adverbial time phrases helps by suggesting a past time reference, whereas a vow to perform some religious activities indicates a future time reference. This treatment of HPy is commmonplace among contemporary Bible translators. Grammars also lack objective criteria to decide what translation to prefer, especially in terms of the tense of verbs in HPy. Non-perfective verbal forms fit well with gnomic perfect (IBHS, 506) and would make a good parallel as progressive and habitual non-perfective (pp. 504-6). Nonetheless, IBHS values more the ability of y]qf{l to represent a situation that arises as a consequence of another, particularly as a situation in future, and exemplifies copiously with samples from Psalms (pp. 512-3). The ‘enigma’ of the Hebrew verbal system is far from being elucidated, though.1 The two main options, tense and aspect, are still being disputed. Among the more recent and original contributors in relation to the verbal system as used in HPy, one can mention the divergent opinions of Diethelm Michel2 and Jan Joosten that epitomize the ongoing debate over the function of verbal forms in Hebrew. Michel proposes the aspect theory of the Hebrew verbal system based on a thorough investigation of q*f^l and y]qf)l in the Psalms. Methodologically, he assumed that literary contexts (poetic genres) have 1 McFall’s monograph The Enigma of the Hebrew Verbal System: Solutions from Ewald to the Present Day (1982), doctoral research presented to the University of Cambridge in 1981, provides a good historical overview of the main solutions given not only to the thorny issue of w~Y]qf)l in particular, but also to the Hebrew verbal system in general. 2 Michel’s work is virtually unknown, for, to our knowledge, there were no reviewers of his work, until his contribution was included in IBHS (470-4) and summarized by Gosling (1992, 5-15).

9

10

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

influence on the syntactic usage of verbal forms. Hence, Michel progressed carefully in his study, considering the variations of q*f~l functions in relation to various poetic genres found in Psalms: lamentations, thanksgiving, reports of salvation.1 According to Michel, the suffix conjugation is not employed in the Psalms to express time, but rather to report an action that stands independently, that is ‘self-important’ (pp. 98-99). Such an action is considered absolute and factitive (‘Fakt’). At the same time, the actions expressed by perfect have an ‘akzidentiellen’ character with regard to the acting person (p. 127). Michel noticed that the prefix conjugation is not expressing time either, but is used when meaning is derived from outside the action the verb expresses, and is therefore relative to it, and expressing an action (‘Handlung’) (Michel 1960, 176). In this case, the actions designated by y]qf)l verbs have a ‘substantiellen’ character with regard to the acting person (p. 127). The only place where the analysis of q*f^l and y]qf)l interact in Michel’s work is his chapter on the syntax of the conditional sentence. Four verbal sequence types are identified, some invariable with y]qf)l or q*f^l throughout, and some displaying verb form shifting, from y]qf)l to q*f^l and vice versa (Michel 1960, 190-2). Michel understands them as facts (q*f^l) and actions (y]qf)l) in interaction (p. 194). Most of the samples extend over many clauses and verse-lines and volitives are included in the category of ‘imperfects’.2 Only two samples stay within the limits of couplets, namely Ps 127.1a, b and 63.7. The possibility that the Hebrew verb has aspect was contested repeatedly despite its venerable history, traced back to the German Arabic grammarians of the nineteenth century.3 Most recently, Joosten (2002) rightly challenged the validity of aspect by attempting to prove that the most prominent functions attached to the imperfective in acknowledged aspect languages, such as real present and attendant circumstances in the past, are not regularly expressed by y]qf{l in Hebrew. Conversely, the 1 Isaksson (1987, 15-17) displays a similar awareness of the relationship between genre and grammatical categories that is so much exploited by textlinguistics, although none of them give credit to it. 2 Cf. Ps 40.6; 141.5 (impf - perf. without a conditional particle); 27.3b; 73.21-22; 120.7; 127.1; 139.8-11 (impf - perf. with a conditional particle); 17.3; 37.21; 69.33 (perf. - impf. without a conditional particle); 7.4-6; 21.12; 41.7; 44.21-22; 63.7 (perf. - impf. with a conditional particle). 3 G.H.A. Ewald, Grammatica critica linguae arabicae (2 vols., Leipzig: Hahn, 1831).

SETTING OUT THE PROBLEM

11

predicative participle takes over such functions in Hebrew, whereas y]qf{l expresses modality. Verb usage is particularly important in the study of a verb-initial language, as Classical Hebrew has proved to be. Its importance is increased by the role of the verb in the analysis of PROSE, as recent research indicates (Andersen 1974; Longacre 1989; Eskhult 1990; Niccacci 1990; Schneider 1994; Hatav 1997; Goldfajn 1998 inter alia). Therefore, a careful look into the use of the verb in POETRY may be illuminating for the understanding of its function both at the syntactic and pragmatic levels. As becomes clear from rendering the Psalms into poetic lines, there are several tendencies to be noticed in the use of verbs between the parallel lines. Arguably, the couplet is the regular poetic unit in HPy,1 with one predicate per line and similar verbal forms in both lines being favoured due to parallelismus membrorum. Consequently, verses with multiple lines,2 couplets with multiple predicates per line,3 the elision of the verb in one of the lines4 and the q*f~l//y]qf{l verbal sequence or its variant y]qf{l//q*f~l would be exceptions. Niccacci intuitively noted that the study of the verbal system as it is applied to POETRY is ‘the most remarkable area of disagreement’ between PROSE and POETRY. The unanalysable alternation of q*f~l, y]qf{l and w~Y]qf{l verbal forms in POETRY allowed him to catalogue it as a free alternation. Although he mentions it only in passing, he admits that the

1 Korpel and de Moor (1988, 1-4) advanced the idea that the basic unit of Ugaritic and HPy is the colon, the unit of the oral rendering of the poem, which developed into a couplet only later. For them, the smallest building block is the foot (i.e. a word containing at least one stressed syllable). 2 Verses with three lines can be produced by expanding a couplet (Loewenstamm 1969), while verses with four lines can be the result of rearranging two adjacent couplets producing an alternative couplet (Gray 1915, 62-4) or splitcouplet (Watson 1997). Larger verses can be produced by means of numerical parallelism (Watson 1984a, 148-9; Watson 1991), enumeration and repetition (Watson 1984a, 187-90), vertical parallelism and Janus parallelism (Watson 1984a, 158-9). 3 Internal parallelism represents one notable exception (Watson 1984c, 1985a, 1988b, 1989b). 4 For the ellipsis of the initial verb in the secondary line, see Watson (1984a, 174-6).

12

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

parallelization of same root verbs proves to be a characteristic of UPy, HPy and epistolary literature in Amarna (Niccacci 1990, 194-95).1 Several scholars have analysed verses with the QYYQ verbal sequence in Psalms in particular (cf. Chapter 2), some even in relation to UPy (cf. Chapter 7), as well as in other poetic canonical books,2 thereby legitimating it as an object of study. Further limitations concerning the extent of our research, as well as terminological and methodological options, are detailed in the following sections.

1.5. THE PSALTER AS A CORPUS OF POETIC TEXTS Our investigation will be limited to couplets in Psalms.3 Although, this particular verbal sequence appears in HPy of various ages, including Archaic Hebrew (cf. Appendix 6), we preferred to limit our study to the Psalms for several reasons. First, because there is a strong agreement among scholars that Psalms is a collection of poems representative for HPy. Second, because Psalms stands as a united corpus of Hebrew literature. However, one cannot maintain that this document allows synchronically valuable observations, since it encompasses poems of variegated origins (cf discussion infra). Third, 1 Moshe Held (1962, 281-2) noticed occurrences of various verbal forms with the same root verb in HPy and explained them stylistically, proposing a similar rendering of both verbs as preterites (past simple). 2 Rata (2004, 12-13) quoted Nahum Sarna’s 1955 Dropsie PhD thesis, Studies in the Language of Job, wherein the Jewish scholar recorded twenty-five cases of perfect verbal forms followed by imperfect ones and twenty-seven more cases of imperfect verbal forms followed by perfect ones. Gosling (1992, 514-21) noticed the existence of this phenomenon too and offered four samples (Job 3.3; 4.3; 15.7; 15.18), of which the last three are genuine. He proposed that the verbal forms in these last three samples should be translated similarly with reference to the past. Rata himself discussed occasionally such cases, but, since his interest resided with the verbal forms, his samples are scattered. Job 38.17, 22 on pp. 128-9 and Job 23.12 on p. 135 illustrate the use of long yqtl’s with past reference. Job 6.25, 15.9, 33.13b and 35.6c illustrate the use of long yqtl’s with present time reference in interrogative clauses (pp. 148-9). Job 21.29, 22.15 on pp. 159-160 illustrate the use of qtl’s with past time reference. Job 9.2 on p. 163, Job 33.13 on p. 167 and Job 37.15 on p. 169 illustrate the use of qtl’s with present time reference. The use of qtl alongside short yqtl, long yqtl and even wayqtl gives evidence to Rata for the classification of qtl as perfective and for the use of verbal forms in Job marked for aspect and not for time (Rata 2004, 227-8, as well as the final conclusions). 3 A discussion on their verbal structure follows (cf. chapter 3).

SETTING OUT THE PROBLEM

13

because the phenomenon of couplets with QYYQ verbal sequence appears to be extensively represented in this document. Fourth, since the main thematic in Psalms is liturgical, there is a greater chance that its POETRY would be less transparent than that of wisdom literature or prophecy, and more obscure, hence more poetic than PROSE. Fifth, because a thorough investigation of the whole corpus of the Psalter was too demanding to permit extending it to other documents from Hebrew literature, despite the need and desire to do so. At this point, some clarifications on the nature of our research corpus are necessary. Approaching the MT in order to localize our primary text, several problems arise. Firstly, the age of the literary pieces that the original text consists of varies within rather large limits, because the Psalter is a collection of poems. Secondly, Hebrew itself is not seen any more as a monolithic language. What is the impact of these matters on our study? In recent years, scholars have viewed Hebrew as a language with a diachronical development in its own right. Early Hebrew (also called Archaic),1 Standard Hebrew (or pre-exilic)2 and Late Hebrew (or postexilic)3 were the idioms apparently known in the Holy Land and employed for written documents (Kutscher 1982, 12).4 The picture is further complicated by the existence of diglossia (the existence of two varieties of the same language, one for literary and formal purposes, another for colloquial and informal purposes) and that of local dialects (such as Judahite/Southern or Israelite/Northern).5 Several scholars have offered detailed studies on what the characteristic of the Ancient HPy may be. Cross and Freedman provided since 1975 – when their joint 1950 PhD thesis was eventually published – orthographic, grammatical, lexical, and stylistic evidence for the ancient 1 The biblical documents that fit in this category include mainly the poems from the Pentateuch and the Historical Books. 2 The texts in Standard Hebrew include regularly Genesis-Kings, a terminus ad quem of about 500 B.C.E. being maintained for them. 3 As main representatives of LH, Kutscher and others also list Chronicles, and Ezrah-Nehemiah. 4 Mishnaic Hebrew, Rabbinic Hebrew, Mediaeval Hebrew and Israeli Hebrew followed them. 5 Rendsburg defends the case of an Israelite origin of several Psalms (the Korah and Asaph collections and Ps 29), referring to linguistic evidence whilst applying Hurvitz’s linguistic approach (1990, 14-15). Goulder (1982, 1996) is particularly fond of the theoretical northern origin of psalms from Korah and Asaph’s collections.

14

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

origin or influence of several poems outside the Psalter.1 In his 1976 Harvard PhD dissertation, Geller provides us with structural evidence inspired by parallelismus membrorum displayed by ancient HPy.2 More recently, Hadas-Lebel (1995, 70-72) traced the characteristics of Archaic HPy in several poems outside the Psalter (Exod 15.1-11; Judg 5; Num 23-24; Deut 32; 33; Gen 49; 1 Sam 2.1-10) and in some poems from Psalms as well (Pss 17; 103; 116; 135). Sáenz-Badillos (1993, 57) also renders Psalm 68 as an archaic poem. Hadas-Lebel does not localize any post-exilic influences in any of the Psalms, although she admits that books like Job, Ecclesiastes and Canticles must have been written during this period (HadasLebel 1995, 103). A similar approach is taken by Sáenz-Badillos (1993, 115).3 Of course, such a perception contradicts the expectations of the historical criticism school, which tended to consider, with very few exceptions, that the poems in the Psalms were composed in the exilic and post-exilic period.4 As becomes apparent from the study of the Psalms by means of the historical criticism method, a precise dating is beyond our 1 Here are included Exod 15, Gen 49, Deut 33, 2 Sam 22, and even Judg 5, and 1 Sam 2.20-26. 2 His corpus includes also the Song of Lamech (Gen 4.23-24), the Song of the Ark (Num 10.35), the Psalm of Habakkuk (Hab 3),2 and Psalms 24, 29, 68, 77 (only v. 17), 89, and 114. 3 For important studies on the lexical indications of post-exilic dating see A. Hurvitz, A Linguistic Study of the Relationship between the Priestly Source and the book of Ezekiel: A New Approach to an Old Problem (Cahiers de la Revue Biblique, 20; Paris: Gabalda, 1982). At the syntactic level, other studies could be useful, such as R. Corwin, The Verb and the Sentence in Chronicles, Ezra and Nehemiah (Borna, 1909), R. Polzin, Late Biblical Hebrew: Towards an Historical Typology of Biblical Hebrew Prose (HSM, 12; Missoula, MO: 1976), A.R. Guenther, A Diachronic Study of Biblical Hebrew Prose (Ph.D. diss., University of Toronto, 1977), M. Eskult, Studies in Verbal Aspect and Narrative Technique in Biblical Hebrew Prose (Uppsala: Almqvist & Wiksell, 1990) as well as “Traces of Linguistic Development in Biblical Hebrew” (Hebrew Studies 46: 353-70) and Jan Joosten (2005). 4 Although there is agreement in principle, historical critics do not agree on the historical date of the Psalms individually. Following closely Ewald’s Commentary on the Psalms (German edition 1866, English edition 1880), Arnold, Kitchener, Potts and Phillpotts (1867) accepted few Psalms as pre-exilic, namely 15, 19, 24, 29, 60. Briggs and Briggs (1906, lix-lxxix) proposed thirty-four psalms as belonging to the monarchic period (cf. pp. xc-xci). Buttenwieser (1938) reduced the number of preexilic psalms to twenty, but only six of them coincide with the ones in Briggs and Briggs list (i.e. 19, 20, 21, 24, 45, and 60).

SETTING OUT THE PROBLEM

15

reach. The criteria most preferred for dating Psalms include the historical evaluation of superscriptions, generally considered as editorial additions, internal references to cultic ceremonies or religious institutions, assessing the general tenor of poems in relation to historical incidents and character development accounted for in the historical books.1 Although historical critics also investigated the preference of terms in Psalms, it was only later that this type of study satisfactorily reached a higher degree of specificity and relevance.2 Tsevat’s 1953 Jerusalem PhD dissertation proposes, on lexicographical grounds, the originality of the Psalms devotional language compared to other biblical documents, conserving ancient terminology specific to ANE lyricism of Canaanite and Ugaritic origin and even of Akkadian hymnic-epic origin (Tsevat 1955, 5560). In terms of dating the Psalter, after comparing its language to that of the poems in Chronicles and noticing that they are very different, Tsevat (1955, 61-72) concludes that the Psalter must have been completed by 400 B.C., when the academic consensus indicates that the writing of Chronicles took place. That leaves us with a corpus of poems produced during the monarchic and exilic periods of Israel’s history, with several poems displaying characteristics specific to Archaic Hebrew or even originating before the monarchy. Although we cannot expect the same consistency as in a corpus of POETRY that originated during a fifty years period, we are not left with a corpus that extends over almost a millenium either. But five hundred years is still a long period and many changes must have been taking place diachronically in terms of style preferences. Nonetheless, as we will see below, the QYYQ verbal sequence is one of those poetic devices that found a place in the literature of all these ages. This association of a given poem and its respective narrative (historical context) is particularly visible in Goulder’s approach to setting psalms from Book Five in the context of Nehemiah, rebuilding the Temple and Ezra, respectively (cf. Goulder 1998). 2 Various authors noticed several features that hold together individual poems into collections such as the Prayers of David (Pss 51-72) (cf. Goulder 1990, 20-30), the psalms of the Sons of Korah (cf. Goulder 1982, 1-23), or the psalms of Asaph (cf. Goulder 1996, 15-36 and 190 ff; and Nasuti 1988), the psalms of Ascent (cf. Crow 1996, 129-58, Goulder 1998, 24-7). Nasuti (1999, 141-62) defends the idea that the typicality of David’s experience in the poems that are ascribed to his name functions as a hermeneutical key allowing the reader to unlock the meaning of individual poems that were built on a similar genre model. 1

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

16

As any respectable linguistic exercise would demand, we rely extensively on a given text. Its quality should be established by its age, accuracy and authority among MSS of similar persuasion. Although there are other Heb MSS available that are older than this, the only complete Heb MS known to date is Codex Leningradensis published as Biblia Hebraica Stuttgartensia. Its main weaknesses derive from its Western Masoretic origin, relatively late, though strong and coherent.

1.6. TERMINOLOGY Where terminology is concerned there have been three main solutions offered so far. The first one used to compare HPy with POETRY from other world literatures more or less cognate with Hebrew and evaluate them from the perspective offered by the POETRY theory at use at a given time. The second solution belongs to the other extreme, since it refutes any foreign influence upon the Hebrew literature, preferring the original Hebrew terminology.1 In between, a more balanced solution offered neutral terms, since the Graeco-Roman terminology was indeed alien to the Hebrew literature, whilst the Hebrew original terminology was completely obscure. This work develops Margalit’s terminology, valued for its coherence, for using neutral terms, and for being well defined and tested both in Hebrew and in Ugaritic literature (Margalit 1980, 219-28).2 Thus, the verse represents the main unit of POETRY (line for others); it corresponds to the sentence in PROSE, meaning that it is at once a grammatical and a prosodic entity.3 The verse is subdivided by principal caesura(e) into verse-lines (verset for others, A and B). The basic verse-type is the binary structure, called couplet (distich for others), that consists of two verse-lines divided by a caesura.4 Monolines (monostichs for others), tercets (tristichs for others), For an expanded debate over similar terminological issues, see Tatu (2005). Launched in 1975 (Studia Ugaritica I: Introduction to Ugaritic Prosody, UgaritForschungen 7: 289- 313), the method was applied extensively to Ugaritic literature in A Matter if Life and Death: A Study of the Baal-Mot Epic (CTA 4-5-6) [AOAT 206; Kevelaer: Butzon and Bercker, Neukirchener-Vluyn: Neukirchener Verlagsgesselschaft, 1980], and in The Ugaritic Poem of Aqht: Text, Translation and Commentary [Berlin, Walter de Gruyter, 1989]. 3 Prosody is a term that assumes metre, whose presence, particularly in Ugaritic literature, has long been a serious bone of contention. 4 Although the idea that the couplet stands as the main prosodical and meaning unit of HPy is still debated, one can certainly promote it the most frequent type of verse. Our investigation indicates that 84 per cent (381) of the verses in the Psalter 1 2

SETTING OUT THE PROBLEM

17

and even quatrains (tetrastichs for others)1 are its structural variations with one, three or four lines, respectively. The verse-line is subdivided by blanks into verse-units, mostly coextensive with the individual vocable or lexeme. There is a tendency towards limiting the number of verse-units per verse-line to three, but there are exceptions. The number of verse-units/verse is called the valence of that verse. The pattern of a verse, according to the valence of its corresponding lines, is represented as the numerical sum of the verse-units in the component lines (e.g., 4 + 3; 3 + 3; 3 + 2). Scansion stands for the process of determining the verse valence. The strophe is a conventionally structured sequence of verses in indeterminate quantity.2

1.7. METHOD Our investigation has two main stages, both applied to the primary text: verse-line delimitation and systemic functional analysis. 1.7.1. Verse-line Delimitation Verse-line delimitation is a heuristic process, meaning that a solution is proposed and then its counter-offers assessed. As Kaddari (1970-3, 168) noticed, complex semantic units may appear (compound words, ‘series’, ‘expressions’), and ‘essential semantic links’ (single semantic units whose constituent parts do not have any exact parallels) may be present as well. A practical way to check the results is by identifying the topic of the semantic field as a whole. Units can stand in a topical relationship although particular words may stand alone without explicit counterparts (Kaddari 1970-3, 170).3 Once this analysis is completed, the semantic critic identifies are couplets, followed by a 15 per cent presence of tercets (Appendix 7). Even though van der Lugt’s computation is lower, only 350 (2006, 425), closer to Fokkelman’s 348 total (2000, 348), the variation is too low to count. 1 Both Fokkelman (2000, 17 and 41) and van der Lugt (2006, 523), against Korpel and de Moor (1988, 27-28), do not accept larger verses than the tercet. 2 For the most recent study of strophe as a structural unit in Psalms, see van der Lugt’s 2006 monograph. 3 Nevertheless, Kaddari made use of the grammatical-syntactic structure of the utterance; he examined it in order to validate the parallelism traced down semantically. This procedure is considered ‘a prerequisite for the proper establishment of a parallelism in regard to any word or phrase’ (Kaddari, 1970-3, 172). By doing that, he actually discarded everything that did not fit the synonymous and antithetic categories, those previously named synthetic parallels.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

18

only parallels sharing a common semantic field. Further finer description will reveal the relationship between semantic units belonging to a common field.1 The result is checked against the Masoretic accents. When accents happen to contradict the fluency of syntax, they may also hinder the ease of interpretation (Kugel 1981, 109-16). Given their failure to provide strictly objective criteria for verse division, the Masoretic assessment of Hebrew verse is supplemented by a careful evaluation of the rhythmical balance of each line produced. It appears that in terms of predication, the basic rule points to the fact that each poetic line has to contain not more than one predicate. In those cases where more than one finite verb is present in a particular line, one has to assess their role as predicates. Occasionally even non-finite verbs (such as absolute infinitives or participles) can function predicatively. The nominal clause represents the most notable exception to the previous observations, because of its implied verb and/or non-verbal predicators. Adapting Margalit’s terminology, we speak of kernel elements: verse, verse-lines, and verse-units, as well as expanded elements: strophes. Whenever more than one verse appears in a verset, verses are marked with small Latin letters (a, b, c). Since we work with the Hebrew text as preserved by the Masoretes, versets are marked according to the Masoretic text, although it does not coincide entirely with all Christian traditions. The psalm headings are not included in our analysis because they lack the poetic character of the verse-lines of the poems. Following the scanning of the poems, we identify the couplets with one verb per line, either q*f~l or y]qf{l, but differing from one verse-line to the other, and not necessarily of the same verb root. The division into According to Kaddari, there are two main situations semantic units can find themselves in. Sometimes they are in a parallel relation although they regularly occur separately, outside the sphere of parallelism. At other times, semantic units appearing regularly together, even in non-parallel contexts, are split into the cola of a parallel verse. Semantic units belonging to the first category can develop three types of relations: synonymous, heteronomous, and partially antynomous. Semantic units belonging to the second category create two types of relations: composites of linked phrases are divided into two co-ordinated parallel cola, and a series of coordinated types is split in parallelism (Kaddari 1970-3, 172-4). Despite all these ramifications, the author is realistic enough to accept that there are poetical units that do not display semantic parallelism, and also non-poetical texts, which can be constructed in the style of semantic parallelism as well. 1

SETTING OUT THE PROBLEM

19

verse-lines is compared with the ones suggested by various ancient witnesses and modern interpreters of the Masoretic Text (henceforth, MT).1 Variant relevant readings are given in the footnotes. Then, the couplets displaying a QYYQ verbal sequence are collected in two entirely new databases, one for the q*f~l//y]qf{l verbal sequence and one for the reverse sequence (cf. Appendices 1 and 2). In order to make the analysis easier to understand, original texts appear in transliteration with a rendering in English accompanying them. 1.7.2. Systemic Functional Analysis When the first stage of our research has been completed, we can embark on the more elaborate second stage. Systemic Functional Grammar (henceforth, SFG) is a linguistic tool that integrates meaning with form and intentionality. Each of the couplets that display a QYYQ verbal sequence will be analysed at the three levels SFG holds meaning is realized: textual, interpersonal and ideational (experiential and logical). Individual constituents can have different functions at each of the three levels of meaning. They will be analysed in their relationship with adjacent constituents with focus on the verbs. The results of such analysis are printed in tables accompanying each verse-line, so that the multiplicity of functions is notable at a glimpse. Before interpreting the general results of the couplets with a QYYQ verbal sequence identified, one entire Psalm is analysed at each of the three levels of meaning in order to compare the use of the QYYQ verbal sequence in relation to other verbal sequence types.

1.8. O UTLINE OF THIS STUDY Chapter 2 is a survey of the most representative scholars who engaged with the QYYQ verbal sequence in HPy. Their understanding of the phenomenon is catalogued in two main categories: rhetorical and pragmatic. 1 Commentators such as Talstra (1999, 114ff) give more prominence to the clause in their analysis than to the line so that may find its value when poetical texts are analysed syntactically. Nonetheless, the main deficiency of such an approach is that it disregards the binding force of parallelism in HPy, whereby incomplete clauses or even groups of constituents can echo each other and hence produce artistic lines (though not co-extensive with clauses). Therefore, the syntactic analysis cannot be done in a vacuum, as Talstra suggests, without endangering the understanding of the text. Syntactic variations can indeed have non-syntactic explanations.

20

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

In both cases, a certain discontent with the explanations offered by the traditional grammars can be sensed. Chapter 3 aims to produce a method for verse-line delimitation to be applied to the Masoretic text of the Psalter. For this, the value of the Masoretic accentual tradition is assessed, the contribution of ‘delimitation school’ is estimated and the contribution of several modern scholars who dealt with the Masoretic text of the Psalter is critically surveyed. As a result, two major tools are produced. First, the main database consists of all 150 Hebrew Psalms as preserved by the Masoretes displayed in verse-lines, with references to other ancient, mediaeval and modern authorities (cf. Appendix 3), who also split these poems into poetic units and subunits. Each verse-line is numbered according to the poetic verse it is a part of. The Judaic verset and chapter tradition is preserved. The verbal structure is noted in the margin. Excerpts from this database are included in Appendix 9. Second, upon completing the database, one hundred and two samples of couplets with QYYQ verbal sequence are identified. Appendix 1 includes the sixty samples of couplets with q*f~l//y]qf{l verbal sequence, and Appendix 2 the fourty-two samples of couplets with y]qf{l//q*f~l verbal sequence. Supplementary, as auxiliary aids to this thesis, two more appendices are produced. Whereas Appendix 7 surveys the distribution of couplets in Psalms, Appendix 8 selected the QYYQ verbal sequence as it appears irregularly in couplets or tercets. Chapter 4 investigates the SFG of English as proposed by M.A.K. Halliday inasmuch lexicogrammar is concerned. Thus, it represents the theoretical basis of this study. Examples used to illustrate the principles of SFG are selected from James Joyce’s Dubliners. Chapter 5 applies the principles of SFG to SH PROSE texts from the collection Genesis to Kings.1 Hundreds of examples are quoted in references as footnotes, and 352 examples are quoted in full, with transliteration, literal translation and Systemic Functional analysis. As a less natural type of text, POETRY requires a preliminary understanding of the phenomenon in PROSE. 1 To the best of our knowledge, there has been only one previous attempt, see A. Garnowski, A study in the development of a systemic functional grammar for Biblical Hebrew clauses: a preliminary demonstration of systemic functional grammar's explanatory power in clause level structure and meaning (MPhil Birmingham thesis, 2000).

SETTING OUT THE PROBLEM

21

This preliminary clarification equipped us for analysing individual poems and the collection of couplets with QYYQ verbal sequence, which follows in chapter 6. An interpretation of all results is also included. The Systemic Functional analysis is added to the texts in Appendices 1 and 2, in a tabular format that extends over several lines. Two more appendices are generated at this level. Appendix 4 is created in order to substantiate the connection among verb forms, Subject and Marked Theme. Appendix 6 includes samples of couplets with QYYQ verbal sequence in Ancient HPy. Chapter 7 extends our research into the domain of Ugaritic literature, comparing the results from HPy with those in UPy. Appendix 5 includes the couplets with QYYQ verbal sequence identified in the UPy following the format used for the Psalm couplets. Conclusions are drawn in chapter 8 and further developments of the current research are suggested.

1.9. RELEVANCE OF THIS S TUDY 1.9.1. Relevance to the Literary Criticism of Hebrew Po etry Particularizing the ongoing debate on the difference between PROSE and POETRY to Hebrew literature, critics agree that there is room for further refinement. There is an increasing consensus among scholars, though, with regard to the lack of decisive features that would provide the long expected criteria that distinguish a text as being POETRY. This study argues that the QYYQ verbal sequence is a poetical feature that allows the reader to distinguish with more accuracy between PROSE and POETRY. Since it is not present in PROSE texts, with the exception of some conditional clauses, the QYYQ verbal sequence could be the feature that uniquely marks the poetic language. HPy is described by a multitude of features, of which parallelism (at various levels), rhythm, conciseness and ellipsis, inter alia, are particularly important. More recently, the QYYQ verbal sequence entered the gallery of poetic devices as well. This study defends the idea that the QYYQ verbal sequence is a poetic device in its own right, used successfully by ancient poets in the HPy of the Psaltire and the alphabetic cuneiform tradition of UPy. Evidence of its usage also includes Ancient HPy.

22

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

1.9.2. Relevance to the Grammar of Classical Hebrew Translators of the HB are not at ease with the deviation from the regular usage of verbal forms that is found in POETRY. The lack of theoretical consensus one finds in grammars in relation to the use of Hebrew verbs aggravates the practical dilemma. Based on the careful observation of verbal usage in PROSE, it is noticed that scholars have relied on mere intuition in promoting the QYYQ verbal sequence as a valuable research topic and poetic device specific to HPy. This study argues that q*f~l and y]qf{l verbal forms, when part of the QYYQ verbal sequence in Psalms poetic couplets, can be used primarily for aesthetic reasons, with no individual reference to time or aspect. Arguably, the SFG analysis of lexicogrammar can provide a comprehensive interpretation of form and function, and an integrated approach to phonetics, morphology and syntax. 1.9.3. Relevance to the Exeg esis of the Hebrew Psalms as Preserved by the Masoretes Various practical and theoretical reasons allowed the researcher to limit the object of the research to the Psalms. Admittedly, Psalms is a variegated corpus in terms of authorship, implicit style, origin, topic, inter alia. At the same time, though, Psalms stands as a unified corpus of Classical Hebrew poems, which does not display the features of Late Hebrew texts, and only some Psalms would convey features specific to Ancient HPy. This study proposes a method for verse-line delimitation of HPy that integrates various contributions into a coherent step-by-step approach. It also argues that the Masoretic text of the HB provides the best complete document for the study of the Psalter. Had this thesis not been correct, the Masoretic text can still be successfully employed for the study of the Psalter, since it stands as the instantiation of a civilization whose both cultural values and language are dead. At the same time, it generates a resourceful and fully retrievable database on the Masoretic text of the Psalter, which includes references to ancient, mediaeval and modern authorities. Moreover, it argues that SFG offers the tools that allow the exegete to access information beyond the form, through the various functions a constituent can simultaneously hold at various levels of meaning. The main contribution of this study consists in the SFG interpretation of the QYYQ verbal sequence it offers. Although previously accounted for as a rhetorical device and explained by means of pragmatic devices, this

SETTING OUT THE PROBLEM

23

study argues that the preference for one verbal form or another in a couplet with QYYQ sequence can be explained successfully in relation to thematic, mood, transitivity/ergativity and clause complex structure. 1.9.4. Relevance to Systemic Functional Grammar Since its first theorisation in English, SFG was written and rewritten for various world languages and dialects. One of the basic assumptions of Systemic theory is that the language concerned is privileged for the description of each language. In other words, SFG offers only a general linguistic theory not a theory of universals, and it has to be applied to particular linguistic descriptions. This study attempts to apply the principles of systemic functionalism for the first time to a corpus of Standard Hebrew PROSE texts, producing a tentative Systemic Functional Grammar of Hebrew lexicogrammar. 1.9.5. Relevance to Ugaritic studies The connection between the Ugaritic literary tradition and its Hebrew counterpart has long been admitted based on various poetic features common to both. Assuming that the QYYQ verbal sequence is a genuine poetical feature, its presence in literary texts originated in both traditions proves one more time their kinship. This paper argues that the Hebrew literary tradition moves away from its Ugaritic kin when the use of the QYYQ verbal sequence is concerned.

2 PREVIOUS CONSIDERATIONS OF THE QYYQ VERBAL SEQUENCE At first, the QYYQ verbal sequence was not considered at all to be an intentional creative act on the part of original bards, but simply reflections of morphological options.1 After the discoveries at Ugarit and in the surrounding area took place (during the first four decades of the twentieth century), a heavy scholastic assault took place on the alphabetic-cuneiform literature found there, with the effect of illuminating various areas of BH research including verb usage. It was soon noted that such a verbal sequence does not seem to be an accident, but rather a stylistic device among many others, skilfully exercised by the ancient poets. Simultaneously, linguistic research advanced and many grammatical irregularities found a deeper and more complex explanation by the pragmatic schools of linguistics. Ugaritic and Hebrew literature were caught up in this process and new things started to be revealed.

2.1. THE RHETORICAL S OLUTION Among the scholars surveyed in this section, we included only those we were aware of who studied the QYYQ verbal sequence in HPy and interpreted it as stylistic device. Most of them were not content to remain within the limits of HPy and extended their domain of samples to cognate languages, such as Ugaritic and Akkadian, or to Hebrew literature from other ages than that of the Psalms, such as Archaic HPy or Qumranic hymns. Those authors with an exclusive contribution on UPy will be surveyed in chapter 7. 2.1.1. Moshe Held Held (1962) draws our attention to several examples found both in HPy and UPy, that consistently reflect a similar pattern. He noticed that the 1

E.g. Kautzsch (1910, §107b) and Joüon (1996, §113 o, p). 25

26

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

q*f^l verbal forms could stand in either one of two parallel lines, preceding or following a line with either a y]qf)l or a w^yy]qf)l, and sharing its meaning with the verb from the corresponding verse-line. Consequently, there are four types of possibilities for verbal forms to join in verbal chains, displayed here in abbreviated form: A – q*f^l, B – y]qf)l (Ps. 93.3)1; A – y]qf)l, B – q*f^l (Ps. 38.12; Prov. 11.7); A – q*f^l, B – w^yy]qf)l (Ps. 29.10); A – w^yy]qf)l, B – q*f^l (Is. 60.16; Amos 7.4).

Held (1962, 285-6) interpreted these possibilities as evidence for the presence of the QYYQ verbal sequence as a traditional poetical device, extant for a long time in UPy by the time it was found and valued by Hebrew poets.2 As it happens, the new poets, Hebrew in this case, took the risk – and the pleasure – of being critical with the inherited poetic tradition and of adapting it to their own needs.3 There are two observations that can be made on the structure of Held’s first two pattern types. First, there is a very regular verbal sequence in both lines: line A contains a q*f^l verbal form, and line B contains a y]qf)l, and vice versa. Second, the same verbal root is used for the q*f^l verbal form in line A, as for the y]qf)l verbal form in line B. Held also 1 Loewenstamm considers this verse a clear example of a 1-2-1-2 expanded colon with a third parallel line added (1964, 187). Therefore, it is considered by some scholars a tercet, rather than a regular couplet. Indeed, if lines are divided following the syntactical criteria, this solution makes much more sense than Held’s. 2 His samples from UPy are considered in more detail in the seventh chapter of this dissertation. 3 For example, Psalm 26.4-5 has four lines as follows: 4A – ytbvy-al – q*f^l // 4B – awba al – y]qf)l and 5A – ytanf – q*f^l // 5B – bva – y]qf)l. Similar evidence can be drawn from the issue raised in Ps. 29.5 and 29.10, where the verbal form of line 5A is read by the Masoretes as a participle (q)f}l), whereas if in relation to the following verb is understood, in respect with the poetic tradition aforementioned, a Pi q*f^l is more appropriate. To be noticed that verse ten of the same psalm follows the regular paradigm, i.e., 10A – q*f^l √bvy // 10B – w^yy]qf)l √bvy. Then, the verb catena in verse 5 is similar, 5A – q*f~l √rbv // 5B – w~Y]qf{l √rbv. The w~Y!qf)l plays the role of a regular y]qf{l as previously noted (Held 1962, 281).

PREVIOUS CONSIDERATIONS OF THE QYYQ VERBAL S EQUENCE

27

hinted that a similar case may be offered by an active/passive sequence constructed from identical root, a study that he was to undertake later (Held 1965). His insights into this verbal sequence as a poetic device known in UPy and HPy gave him the chance to emend ‘corrupted’ texts from both literary traditions. 2.1.2. Mitchell Dahood Dahood is a scholar who dedicated himself to the study of Hebrew grammar from a comparative viewpoint in relation to Ugaritic. His essay published in 1965, in which he embarks on a dialogue with Ugaritic scholars, such as Gordon or Ginsberg, has the character of grammatical notes resulting from Dahood’s past decade of comparative work in the MT and Ras Shamra texts, and aims to stimulate scholarly thinking in the field. His later contribution concerning the verbs can be summarized in two main principles of interest to us. First, discoveries of vowel ‘~’ in several Mari texts makes very probable the usage of the vowel in the Ugaritic preterite (p. 19).1 Second, because Ugaritic and Hebrew are cognate languages, preserving the Canaanite q~f~l~ in HPy for metrical and other reasons is very plausible (p. 20). Dahood’s three volume commentary on the Psalms, or rather his ‘prolegomenon’ to a commentary, as the author preferred to call it (Dahood 1965, XVII), develops the idea that Ugaritic literature offers a very good setting for a better understanding of Psalms. In his quest for originality, he found that observing the Psalms in relation to the Ras Shamra tablets from a comparative philological approach could provide new opportunities to approach an old subject with freshness and erudition. Dahood was perfectly aware that not all the results of his detailed study will stand the test of time (further discoveries and ongoing clarifications are meant here), but for the sake of positive results that may arise he chose to engage in doing it (Dahood 1965, xx). Although Dahood’s initial methodological approach aimed at reconstructing MT and dating its books, we note a growing understanding of the stylistic incidents in UPy and HPy and their bearing on aiding the analytical observer to perceive their inner literary mechanisms. Thus, Dahood’s introduction to the first volume includes a short section on ‘Poetic Principles’ but he enlarged it extensively by the time he published As opposed to Gordon, who could not commit himself pro or contra due to the paucity of evidence (Gordon 1955, 55). 1

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

28

the last volume. Whereas initially, Dahood implied stable structural patterns of the verses, word-pairs,1 double-duty interrogatives, vocatives, prepositions, etc., and broken stereotype phrases,2 as poetic devices subordinated to the most general principle of the parallelism of thought (Dahood 1965, xxxiii-xxxv), as expected, later on he is more specific.3 Published at the end of Dahood’s three volume project on the Psalms, the section entitled ‘The Grammar of the Psalter’ requires our attention (Dahood 1970, 361-456). Extremely compact and illustrative, this final section would deserve the title ‘A Compendium of Grammatical Incidents in the Psalms’ instead, because the large diversity of incidents scattered throughout the three volume commentary are neatly organized in sections and subsections. Even though the page number is not provided, its character is closer to a modern index more than anything else, a useful tool for scholars and teachers. The material listed under ‘Syntax and poetic devices’ is particularly relevant for our discussion. The association of the terms ‘syntax’ and ‘poetic device’ is very suggestive, both for the entries listed and for the practical prominence of the concept. In Dahood’s understanding, there is a close relationship between the specific type of syntax used in the Psalms and the style of their authors, a recurrent modern theme in Biblical literary approaches. The QYYQ verbal sequence is illustrated by sixty-two examples arranged according to the time sequence they were understood to refer to (Dahood 1970, 420-3). As becomes apparent from comparing Dahood’s Dahood recognizes the scholarship of Umberto Cassuto, then available only in Hebrew, quoting his 1971 work. Similar articles, originally published in Hebrew, appeared even earlier: Biblical and Canaanite Literature, Tarbiz XIII(1942): 197-212 and XIV(1943): 1-10 [Later trans. into English and published in Cassuto 1975, 1659]), and The Israelite Epic, Keneset VIII (1943), Pt. 3, pp. 121-142 [later translated into English and published in Cassuto 1975, 69-109]). 2 At this level, Dahood is indebted to E.Z. Melamed (1961), whose work is briefly abstracted. 3 Dahood’s introduction to his second volume focuses on the method he uses on textual criticism being applied to Psalms. Suffice to say that Dahood does not allow any of the four types of observations the analyst may apply to the text to determine a hasty emendation of MT but rather to suggest points of dissolution. Therefore, neither consonantal construction of the text, nor sense, nor grammar, nor prosody could change ad libitum changes without thorough verification of proposals (Dahood 1968, xvii-xxii). These observations are reiterated in the introductory note Dahood places in front of his third volume. 1

PREVIOUS CONSIDERATIONS OF THE QYYQ VERBAL S EQUENCE

29

database with our own (cf. appendices 1 and 2), many other examples could have been added to his list. If we limit the data to couplets only, many of Dahood’s samples cannot be be included in our list, because they are tercets. Others, further, are not accepted because they have been reconstructed. However, nineteen examples with q*f~l//y]qf{l verbal sequence,1 and eight examples with y]qf{l//q*f~l verbal sequence are admitted entry.2 Dahood succeeded in advancing the study of QYYQ verbal sequence by multiplying the number of examples in HPy, particularly in Psalms, and even attempting a translation that would fit them in their context. 2.1.3. Stanley Gevirtz Although his interest is in providing evidence of conjugational variation of the parallelization of verbs with identical roots in the Amarna Letters and to support a Canaanite literary tradition, which extended throughout the Levant, Stanley Gevirtz (1973a, 100) provides an extra sample from HPy, namely Deuteronomy 33.12. His evidence from the Amarna Letters includes EA 197: 31-34 (q*f~l//y]qf{l), 123: 22-26, 33-35 (y]qf{l//q*f~l), and 112: 7-9 (active // passive). An even more complicated type of parallelization of same root verbs is supplied by Gevirtz (1973b, 168-9), this time at the level of a tercet: EA 155: 67-69 (q*f~l//y]qf{l//y]qf{l). The similar samples quoted from HPy (Ps 93.3; 57.9) are actually good illustrations of an expanded tercet, distinct from the numerical tercet involved in the Amarna text and its Ugaritic counterpart (KTU 1.12: 49-52) (y]qf{l//q*f~l //q*f~l).3 2.1.4. Frank C. Fensham On the basis provided by the comparative-historical solution, as McFall preferred to categorise it, Fensham continued the study of verbal usage. Noticing an increasing agreement among scholars regarding the process of developing that Hebrew grammar passed through, Fensham identifies two main potential failures this type of academic enterprise may run into, Ps 6.10; 7.14; 26.4; 26.5; 26.12; 50.19; 63.7; 71.17; 73.3; 73.9; 74.14; 78.64; 81.7; 102.15; 111.5; 116.2; 118.10; 139.13; 140.3. 2 Ps 8.7; 9.8; 38.12; 55.5; 73.18; 83.3; 89.44; 132.17. 3 Although the number of tercets with a QYYQ verbal sequence continued by another similar verb could have been extended, for the sake of our argument, we preferred not to include them in our debate. 1

30

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

namely the editorial activity of the HB and the insufficient yet growing knowledge of Hebrew. Regarding the editorial activity characterizing the HB, it is expected that later editors were tempted to change archaic grammatical constructions into more readable and understandable constructions for their time, so concealing valuable pieces of information both for the reconstruction of AH and for the diachronic study of its dialects (Fensham 1978, 9-10). Yet some valuable observations of this type can still be traced in Ancient HPy, as W.F. Albright (1968), and F.M. Cross and D.N. Freedman (1973) have already demonstrated. The possibility of finding the distinction between the suffix and prefix conjugations in early HPy intrigued Fensham too, but his conclusion is that choosing between suffix and prefix verbal forms was simply a matter of preference. He proposes that a closer look should be taken at Psalms belonging to the AH literature, to be precise Psalms 29, 82, and 93. The ambiguity of his method is reflected by the highly conjectural emendations he proposes. The main presupposition is that w~Y]qf{l is a late achievement of Ugaritic literature, and conjugations were misspelled by the Masoretes. Otherwise, the relationship between q*f^l and y]qf)l is derived from the context. Whenever a prefix form has the Lord as subject, it reflects a habitual action. Where Psalm 82 is concerned, the deviation in usage of prefix forms is so irregular that the conclusion drawn focuses on its later origin. A similar type of exercise is applied to Psalm 93, stressing the interchangeability of suffix and prefix verbs. Fensham’s plurality of approaches and indecisiveness is obvious from his conclusions (p.18). Interpreting the relationship between the suffix and the prefix verbs is a complex matter related to the ACTOR, the way the action is performed, and the relation of the verbal forms to one another. 2.1.5. Randall Buth Buth’s main contribution to the field of Biblical studies consists in applying functionalism to Aramaic texts in the Bible (Buth 1987) and to the Hebrew texts, mainly PROSE (e.g. Buth 1994, 1999), but also POETRY (Buth 1992). Despite his late rather weak efforts to explain word order variations in HPy by means of functionalist categories (cf. Lunn 2004, 98-102), earlier in his publishing career he entertained the idea that verbal form variations were a mere reflection of aesthetic values.

PREVIOUS CONSIDERATIONS OF THE QYYQ VERBAL S EQUENCE

31

In one of his early articles, Buth (1986)1 attempted to identify the variations tense-shifting displayed in Psalms and even to offer suggestions in relation to their function and origin. By means of three samples (Ps 2.1-2; 55.5-6 and 24.2), and after discarding the alternative explanations one by one (aspect, metre, syntax, discourse), Buth promoted tense-shifting as means to achieve poetic effect, namely ‘a kind of antithetical grammatical parallelism’ (p. 28).2 He also discovered that in the case of his preliminary samples, structural devices such as chiasmus (a-b-b-a) or parallelism (a-b-a-b) interfere with tense-shifting, producing even new word order. This does not imply, however, although the author is not clear, that such structural devices are responsible for arranging constituents in all the samples he could find in Psalms. Buth took w~Y]qf{l verbal forms as ‘the temporal equivalent of a suffix verb’ (p. 27) without giving any clue to the reason for such a choice, although Held did the same,3 and included samples that have such verbs in their verbal sequence as well. He was able to trace 54 samples of poetical texts with q*f~l//y]qf{l (w~Y]qf{l//y]qf{l) verbal sequence and 37 of the y]qf{l//q*f~l (y]qf{l//w~Y]qf{l) type, of which only fourteen

1 Originally, a paper presented at the eleventh Congress of IOSOT in Salamanca (1983), the study was not published along with the other congress papers in 1985 (VTSuppl, 36). 2 Another independent, though more modest, attempt to explain the QYYQ verbal sequence as antithetic grammatical parallelism belongs to Krašovec. Although welcoming a study of antithetic structures in HPy, Kugel (1986) and Watson (1988a) criticise Krašovec, the former for his lack of clarity in terms of methodology and terminology, the latter for lacking comprehensiveness. Watson (1986) follows up on the suggestions that UPy possesses many samples of antithetic structures. While analysing Psalm 73, the Slovakian scholar touched, in passing, the ‘parallelism between the imperfect (tyvt, wdbay) and the perfect (tlpt, htmch) in vv. 18 and 27’ (p. 46). Among the psalms selected to illustrate the antithetical structure at work (1, 2, 3, 5, 30, 31, 32, 34, 35, 55, 58, 59, 73, 82, 89), one could have identified many other similar examples of verbal parallelism (e.g. 3.7; 5.6; 30.7; 31.6; 55.5; 58.9; 73.3, 6, 9; 89.13, 36, 44). Given the fact that Dahood’s revolutionary commentary on Psalms has long been published, such an ignorance of basic work on the parallelism of verbal forms is not acceptable. Interpreting the q*f~l//y]qf{l verbal sequence as a kind of grammatical parallelism was previously proposed by Berlin (1985, 36). 3 Talmy Givón would not agree with this association, though (cf. 2.2.1.).

32

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

exhibit a chiastic structure (e.g. Ps. 2.1-2; 7.13-14) and ten the parallelistic structure (e.g. Ps. 55.5-6; 26.4-5).1 Buth correctly noticed that this device appears also in Ancient HPy (Exod 15.5) and in Qumran Hodayot (1QH 2.29; 11.32). He could have included other samples from Ancient HPy, though not from Exodus 15 (cf. Appendix 6). With the exception of the aforementioned passages, Buth did not provide a list of the examples he found, thus weakening his argument. Buth allows himself too much freedom because he included in his list of samples even some with volitives. Once he had done this, provided that w~Y]qf{l is the equivalent form of q*f~l and completely ignoring the nominal clauses, the verbal sequence in Psalms is suddenly described only by q*f~ly]qf{l-y]qf{l-q*f~l. By doing this, Buth can accommodate even samples that extend over more verses (e.g. Ps 44.10-17). Although his approach was reductionistic, Buth succeded in offering an alternative interpretation to one of the less familiar verbal sequences in Psalms. Its value would have been greater if he had handled his examples more carefully. 2.1.6. Wilfred G.E. Watson In his 1984 textbook of Classical Hebrew Poetry, Watson included the ‘variation of tense (y]qf{l//q*f~l and q*f~l//y]qf{l)’ as a device used only to avoid repetitive parallelism (p. 279). Among the examples quoted, Watson included Ps 38.12, Amos 7.4 and Deut 33.12 (perhaps)2 as samples for the y]qf{l//q*f~l verbal sequence, as well as Prov 11.7 for the q*f~l//y]qf{l verbal sequence. At that date, Watson did not admit the existence of such samples in Ugaritic (Watson 1984, 280). Nevertheless, he concurred with Gevirtz (1973, 102-4) on the existence of similar cases in Akkadian (Watson 1984a, 280 n. 28). More recently, Watson (1989a) provided an overview of UPy passages where q*f^l is used in connection with other forms. He considers the phenomenon as part of wider grammatical parallelism, realising that studies 1 Browsing the database for couplets with a w~Y]qf{l//y]qf{l (y]qf{l//w~Y]qf{l) verbal sequence produces the following results: 8.6; 18.40; 28.7; 64.9; 65.9; 69.22; 78.36, 44, 58, 72; 81.13, 17; 105.44; 106.12, 18; 107.6, 19, 28; 143.4 (w~Y]qf{l//y]qf{l); 3.15; 18.19; 55.6; 78.15; 106.19; 107.18; 107.29 (y]qf{l//w~Y]qf{l). 2 Most certainly, the sample from Deut 33.12 does not fit the description of such a phenomenon. We identified three samples in Deuteronomy (33.3b, 7c, 9b), though (cf. Appendix 6).

PREVIOUS CONSIDERATIONS OF THE QYYQ VERBAL S EQUENCE

33

of their independent sentence usage are not sufficient. Before presenting the data, Watson surveys the history of research, with the main outcome of highlighting previously agreed conclusions. Except for Held’s essay, seen as unfairly restricting itself to forms of identical verbal root, the rest of the studies surveyed merely ignored the co-occurrence of q*f^l and y]qf)l. Watson also focused on two developments with good potential for the interpretation of the QYYQ verbal sequence. First, q*f^l is preferred in the parallel line if it pairs a verb from a similar root (a stylistic device as seen by Cassuto and Held). Second, q*f^l is preferred in the parallel line if a member of an adjacency pair describes the fulfilment of a command, where the first member is the command itself. Even so, there are many other instances, which are not included by these two general principles. These are not to be confused with the general use of q*f^l verbal forms in Ugaritic, as marked structure, as pluperfect, as a mark for rapid development of the plot, simultaneity of scenes, shortening the narrative (Delekat), or focusing on new material (Fenton). As can be seen, most of the later quoted usages are functional rather than grammatical. A further clarification prefaces the data, namely the distinction between narrative and discourse. In this case, the term ‘discourse’ refers to utterances taking place in a present situation of communication between a speaker and a listener (dialogue, sermon, prayer, etc.). As opposed to discourse, ‘narrative’ refers to utterances concerning persons not present in the situation of communication (past). Since samples are neatly distributed among discourse and narrative texts, the distinction between them is either useless or too artificial. His method can be criticised for assigning passages from Ugaritic literature to one of these types without spelling out the principles to do so other than thematic content, and for the verse-line delimitation of the texts. After surveying the instances, Watson concluded that the sequence y]qf)l//q*f^l is commoner than q*f^l//y]qf)l, which, alongside Held’s conclusions, coalesces into a stronger argument. There is common ground among all the aforementioned scholars. They shared the view that the QYYQ verbal sequence in HPy is neither a mere accident nor a poetic incident, but rather a poetic device well known to the original bards. Originally studied only in relation to same root verbs, the QYYQ verbal sequence included in Dahood the situations with different root verbs. The phenomenon is found also in UPy (Held and Watson, for which see chapter 7) as well as in the Amarna Letters (Gevirtz), Ancient HPy and Qumran poetry (Buth).

34

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

So far, it was quietly admitted that the QYYQ verbal sequence can appear as part of various verbal structure types in verses of various sizes, hence without any restriction to the couplet with one verb per verse-line functioning as predicate of one clause. When only the same root verbs verses are considered, y]qf)l//q*f~l is commoner than q*f~l//y]qf)l. The situation changes in favour of the latter when different root verbs verses are included. The rhetorical approach to the QYYQ verbal sequence in couplets of HPy understands verbal option as an expression of aesthetic reasons to produce a well balanced poetic unit, contributing to the dissimilarity character of verse-lines, as opposed to other poetic devices, such as grammatical parallelism for instance, which would contribute to their similarity. Given its presence in various neighbouring cognate cultures and even at various ages within the same literature, it can be inferred that the ancient poets employed the QYYQ verbal sequence as a pan-Levantian poetic device. Considering the examples from the UPy (chapter 7 and Appendix 5), there is sufficient evidence to argue that, diachronically, the Ugaritic poetical tradition also included the QYYQ verbal sequence. The Hebrew poets imported this along with the other poetic devices and employed it for creating good quality POETRY. The character of the Hebrew originality will be assessed only after a proper analysis of examples from both lyric literatures is completed (cf. 7.4.5.).

2.2. THE PRAGMATIC S OLUTION The uneasiness of solutions offered by the traditional grammars was overcome once the linguists and language philosophers noticed that some grammatical incidents cannot be explained in the traditional way. Even some of the traditional explanations do not fit well within the morphological or rhetorical setting. For example, one of the functions Delekat (1972, 11-26) assigned to the suffix conjugation was that of the overlapping of events, a specific narrative strategy used to portray simultaneous events. A similar intuition belongs to Fenton (1969, 34-38), who noticed that the q*f~l form is used to replace the more common y]qf{l form when quasi new material appears to be the focus of the poet once again. Within his rather long definition of focus, two different pragmatic functions of marked material are concealed, namely focus and foreground.

PREVIOUS CONSIDERATIONS OF THE QYYQ VERBAL S EQUENCE

35

Pragmatics is no longer a recent development, but its significance in explaining difficult grammatical occurences is valued by an increasing number of scholars. The matter of sequence, as noticed by several erudite predecessors, is a good starting point for a pragmatic approach to our topic. Taking into consideration the Subject position, relative to its corresponding verb, it is known that all languages are assigned to three major categories: S(ubject)-V(erb)-O(bject)-syntax languages, VSO-syntax languages, and SOV-syntax languages.1 The word order of HPy has fascinated students of the biblical languages for a long time. Until recently, scholars dealt primarily with Hebrew PROSE. Lately, several doctoral theses, investigating the word order of the HPy, be it lyrical or prophetical, emerged, such as those of Talmy Givón, Michael Rosenbaum, Walter Gross, and Nicholas P. Lunn. 2.2.1. Talmy Givón From his field observations, Givón proceeded to propound his theory, according to which, during the diachronic development of Hebrew, the VSO-syntax stage was – pragmatically speaking – very unstable; it is therefore considered a transitory stage. This situation forced the language to resolve the conflict in either of two ways: (a) to emphasize ‘the topic to the left’ principle and shift from VSO-syntax to SVO-syntax; or (b) to emphasize ‘the topic on the right’ principle and shift from VSO-syntax to VOS-syntax (Givón 1977, 241-3). Such a transition was noticed by Givón when studying quite a large corpus of Hebrew texts originating in two distinct linguistic periods, corresponding to AH and LH. The process can be seen more clearly when several grammatical environments are taken into consideration. These environments are ordered in a number of independent implicational hierarchies: degree of subject’s topicality (existential > indefinite > definite > anaphoric-pronoun); degree of predicate’s topicality (specific > generic); verb type (active > stative) modality of the predication (irrealis > realis and affirmative > negative) 1 For example, Sumerian, Old Babylonian, Japanese and other languages are verb-final languages, English, Modern Hebrew and Romance languages are verbmedial languages; Spanish, Russian, Tongan and Classical Hebrew are verb-initial languages.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

36

All these hierarchies manifest a larger preference for SVO-syntax towards the right hand end of them and VSO-syntax towards the left. Thus, existential construction of the subject, specific predicate, active verbs, irrealis and affirmative modality require a VSO-syntax, whereas anaphoricpronoun subject, generic predication, stative verb type, realis and negative modality will favour a SVO-syntax. In the first case, the subject has a low referentiality, whereas in the latter the subject referentiality is superior to that of the verb. The theory of drifting from VSO to SVO-syntax relies intimately on two premises: (a) the subject is the most topical element in the sentence; (2) older or more important information tends to be presented first (Givón 1977, 182-7). Analysing the first stage of Hebrew, Givón noted a striking correspondence to the most simple language system called Universal Creole aspectual system.1 The central matrix of such an aspectual system is a binary feature that controls the flow of the story: the opposition between sequential and non-sequential narration, or in Givón’s own terminology continuity:anterior. The continuity aspect is the one that relates the narrative in the same order as that in which it occurred in real time, whereas anterior aspect is the one that violates the sequentiality rule. The projection may go backwards or forwards. In AH, the continuity aspect is carried almost entirely by the imperfect (w~Y]qf{l is a y]qf{l form preceded by a conjunction) which is extensively employed to relate the events in the actual order of occurrence and to maintain the same topic. The anterior aspect is carried by the perfect (q*f~l-forms include here the participle), which is used to look back in discourse and to switch topic (Givón 1977, 198-200). The participle carries also the non-punctual aspect and is responsible for the major transition towards the SV-syntax, whereas irrealis modality is split between imperfect and perfect, with a later shift entirely to the first. If these incidents are observed carefully as Givón did, it will be noticed that Hebrew completed the VS > SV shift by the time Canticles was written, but only in main clauses. This ongoing process can be observed in biblical books from SH literary tradition (Givón used Genesis, 2 Kings, and Esther) as opposed to books belonging to the LH literary tradition (Lamentations, Ecclesiastes, Song of Songs). Several changes took place in the process: (a) the imperfect (y]qf{l) took over the irrealis aspect (including subjunctives, conditionals and future tense, but only in main clauses); (b) the perfect This is the language ‘invented’ naturally by the young offspring of non-native English speakers of Pidgin and Creole English. 1

PREVIOUS CONSIDERATIONS OF THE QYYQ VERBAL S EQUENCE

37

(q*f~l) became the main past-narrative tense (covering both continuity and anterior functions); (c) the participle became the sentential nonpunctual/continuous/habitual aspect (Givón 1977, 233). It is noted that Givón’s conclusions are not presented dogmatically. He admits that the texts selected may be irrelevant for the process, since several of them are accepted as poetic (he named only Lamentations and Ecclesiastes, because Canticles includes a larger percentage of narrative sections than the other texts, and therefore is more of a complex book) (Givón 1977, 237). Nevertheless, his conclusions are valuable and his study pertinent. He confirms the distinctiveness of w~Y]qf{l (a variant of y]qf{l) as opposed to q*f~l, but his explanation of the diachronic process which moulded their functions is the exact opposite of Fenton’s account (Fenton 1973). The difference arises from their different approaches: whilst the latter remains in the field of cognate languages, the former crosses over to noncognate world languages, trying to find universal principles that apply to the language phenomenon. It can be argued that for the reconstruction of the verbal system of a dead language the best option would be considering the verbal system of cognate languages, provided they are transparent enough. Only then, one can turn to universals to allow other world languages to inform the researcher. Although Givón was aware that the poetic texts in his database are representative of Late HPy, he did not prove to have paid enough attention to the possibility that a particular word order may have been promoted in the poetic circles, as the traditional paradigm of rendering poems as opposed to narrative. Actually, AH literature contains several poems that are generally accepted as old poems on the same criteria applied by Givón for his EH texts. Therefore, Givón’s article is of importance where scientific language and careful research is concerned, but his method demands finer tuning. 2.2.2. Michael Rosenbaum In his revised 1996 Brandeis PhD thesis, Rosenbaum investigates the corpus of Isaiah 40-55 using a mainly functional approach, inspired by Simon C. Dik’s Functional Grammar theory. It distinguishes three levels of relations between the constituents of a given text: semantic (Agent, Goal, Recipient, Beneficiary, etc.), syntactic (Subject, Object, etc.), and pragmatic (Topic, Focus, etc.). Admitting the limitations of his study, the fact that POETRY stands as the most defamiliarised language type, and Isaiah 40-55

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

38

represents LH, Rosenbaum embarks on this study with the desire to contribute to a better understanding of syntax from the perspective that a word-order study can provide.1 Rosenbaum (1997, 26-61) tried to distinguish between initial special position, and other pre-verb positions where various constituents are found without being charged with any Pragmatic value, such as Setting or Theme. According to his theory, focus explains genuine alterations of the pattern (pp. 62-97). Further exceptions are noted in terms of discourse particles or parenthetical information, including Vocatives (pp. 98-136). Eventually, Rosenbaum admitted (pp. 149-208) that, especially in the case of POETRY, there are also poetic forces that can produce the alteration of word order in Isaiah 40-55, such as inversion,2 insertion,3 parallelism, gapping,4 repetition,5 word-pairs,6 chiasmus,7 swapping,8 pivot,1 Prior to Rosenbaum, Buth (1992, 90) asserts in his study on Psalm 51 that the marked order of the secondary line breaks the sequential development of the discourse (topic). He also promotes the idea that the initial/fronted constituent in the secondary line underlines the most prominent information of the primary line (focus). These are not sufficient explanations, however, because there are too many other instances where line B closely follows line A in terms of constituent structure and, when marked, the variations are too many (even multiple constituents fronting the verb), all unaccounted for by Buth, and the evidence he offers is too scarce. Besides, this theory leaves no room for alternative reasons for the use of markedness such as plain artistry (cf. Lunn 2004, 99-102). 2 Given the a-b-c order of constituents in a verse-line, inversion means altering the order of constituents in the equivalent line (Rosenbaum 1997, 151-3). 3 Insertion means breaking up two constituents that constitute a whole by adding an asymmetrical element in a verse-line (Rosenbaum 1997, 153-6). 4 Given the a-b-c order of constituents in a verse-line, gapping is a variation of parallelism that means their fragmentary presence in the equivalent line. The element that does not have a pair is only implied in the equivalent line (Rosenbaum 1997, 158-70). 5 Repetition is a variation of parallelism that employs the reiteration of a constituent in the equivalent line(s) (Rosenbaum 1997, 171-4). 6 The use of word-pairs implies the informed choice of words so that one would echo the other even though they are placed in different but equivalent verselines (Rosenbaum 1997, 174-5). 7 Chiasmus refers to the ability of one verse-line to offer a mirror image to its equivalent verse-line (Rosenbaum 1997, 179-81). 8 Swapping is a partial chiasmus whereby only some of the elements are reflected in a mirror like fashion in the equivalent verse-line (Rosenbaum 1997, 181-4). 1

PREVIOUS CONSIDERATIONS OF THE QYYQ VERBAL S EQUENCE

39

terrace2 and staircase3, a process that is called Foregrounding. Although, he considers the trajectory from linguistics to poetics as a legitimate procedure to be followed in exegesis, Talstra criticises Rosenbaum’s failure to address the shift from functional categories to literary categories and the free interchange of ‘clause’ and ‘line’ in his paper. To Talstra, this situation illustrates the collision between linguistic and rhetorical methodologies (1999, 112-3). Lunn’s main methodological criticism of Rosenbaum’s work has to do with his ‘inadequate analysis of parallelism and its bearing on word order variation’. The atomization of the biblical text underlies the failure Rosenbaum shows when not engaging the relationship between the verselines of a given verse (p. 257). Several samples are discussed to illustrate the limits of Rosenbaum’s approach (Is 41.16; 44.9a; 48.13; 49.14-15; 51.13; 46.4) (Lunn 2004, 257-69) Rosenbaum interprets the evidence he collected by promoting VSO as the basic functional pattern of Isaiah 40-55, although, as Meier (1999) rightfully noticed, that is contradicted by the data in Appendices C and E that promote a rather even 50-50 split. A similar situation is noticed in the case of active versus passive verbs and transitive versus intransitive verbs, and even in relation to the worder-order structure of the lines. The argument is weakened even further by the misuse of clichés as entries with cumulative value in his statistical evaluation (Meier 1999). The greatest contribution of Rosenbaum’s work is that of promoting a clear distinction between the genuine Focus pragmatic function and other pragmatic, syntactic or semantic functions. Although Foregrounding is a Discourse Analysis category, Rosenbaum imported it into his study in order to distinguish between constituents with a Focus pragmatic function and other constituents that occupy the most salient slot of a poetic line for poetic reasons only. Although the study is not exhaustive and the author fails occasionally to support his claims with a proper list of samples (Meier 1999), Rosenbaum proves that functionalism can provide some answers for the enigma of word order in HPy. Pivot is a variant of gapping whereby the last constituent of line A does not appear in line B, connecting the two verse-lines as a hinge (Rosenbaum 1997, 1846). 2 Terrace means that the equivalent verse-lines have in common some elements that are repeated but the others are different (Rosenbaum 1997, 186-7). 3 Staircase is a variant of terrace with the repetition of the first constituent and the variation of the others that follow (Rosenbaum 1997, 187-95). 1

40

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

2.2.3. Walter Gross Methodologically, Gross (1992, 53-72) adopted the four main distinctions promoted by the Prague School of linguistics, namely Theme-Rheme, Topic-Comment, Focus-Foreground, New Information-Given Information. He proposed that the term ‘focus’ takes over the rather vague term ‘emphasis’, preferred since Muraoka’s pioneering work (1985), in order to describe the unfamiliar pre-verb constituents (Gross 1996, 92). Although temporal constituents and subjects are often fronted without being the focus of the utterance, non-subject and non-temporal constituents function in most cases as focus. Undoubtedly, he advances Muraoka’s formulation of his methodological framework, but one could still sense a loose taxonomy for the definition of the various functions (van der Merwe 1999a, 178).1 In his 2001 monograph, Gross limited the field of his research to clauses with doubly-filled prefield forms2 in BH texts, observing that these structures are more frequent in HPy than in PROSE (698 samples and 135 samples, respectively, cf. Gross 2001, 21-24). His eagerness to explain every phenomenon from a traditional pragmatic perspective is sanctioned by Lunn. We believe that Lunn’s critique of Gross’s failing to see that the variation of word order could be genuinely inspired by poetic freedom, independently of pragmatic concerns (Lunn 2004, 270), is not entirely correct. Gross, indeed, dedicated the fourth chapter of his work to relationships among clauses with reference to poetic couplets. In his opinion, doubly filled prefield forms, so frequent in HPy, favour parallel and chiastic structures, whether complete or incomplete, fundamentally loaded with aesthetic or ornamental functions (Gross 2001, 78). These stylistic and ornamental grounds seem to offer sufficient explanations for the variation in word order of the data considered (p. 102). For all the other samples, Gross attempts pragmatic explanations in the following three chapters (pp. 105 ff.).

1 Van der Merwe himself proposed a methodological and terminological framework for an information structure analysis of Deut 31-32, but his comments are rather brief, his conclusions highly compressed. He is deriving his authority from Lambrecht’s masterpiece. It is worth noting that 33 per cent of the verbal clauses in Deut 32 have fronted constituents, not far from the average found in Deut 31 (24 per cent) (1999a, 183-4). 2 Clauses in which the verbal slot is occupied by two other constituents that are not regularly expected there.

PREVIOUS CONSIDERATIONS OF THE QYYQ VERBAL S EQUENCE

41

Among the samples quoted in this chapter, we could identify three that qualify as couplets with QYYQ verbal sequence: Ps 38.12 displays complete parallelism (p. 79), Ps 78.64 displays incomplete parallelism (p. 87 n. 29) and Ps 81.7 displays chiasmus (p. 90). Although he does not discuss the relationship between the verbal forms in these couplets, we can derive an understanding of Gross, in relation to this, from their respective translations. Whenever available, the translation indicates a preference for similar verb forms in both lines. Where stylistics and pragmatics collide, precisely in the case of complete parallelism and chiasmus, Gross holds the view that the aesthetics from the foreground overrides the pragmatic function that individual constituents may have (p. 122). He also proposes that the whole clause, which displays doubly-filled prefield forms is in focus. This means that both verse-lines/clauses of a parallel couplet are included (e.g. Is 10.7aAB; Joel 3.1bAB; Ps 72.10AB; Lam 5.13AB) and only the verse-line/clause with altered word order in the case of chiasmus (e.g. Is. 30.4B).1 Gross’s work stands as a consistent application of the principles of functionalism to HPy texts. He was able to notice that the defamiliarised variations of word order support two different interpretations, a pragmatic one and a rhetorical one, with the former serving the purposes of the latter. Although his contribution does not pinpoint the relationship between verb form and functional role, his insights into the inner workings of the variations of word order in verbal clauses are very helpful towards providing a comprehensive understanding of the Hebrew verb, especially in HPy. 2.2.4. Nicholas P. Lunn In his Brunel Ph.D. thesis, Lunn employs pragmatics, more specifically the theory of information structure,2 in order to account for the large number

1 These samples display a Subject ^ Object word order at the front of the verbal clause. Further samples with varying order of the fronted constituents are offered at p. 188 (Subject ^ Circumstantials), p. 202 (Circumstantials ^ Subject), p. 237 (Object ^ Not-Subject), p. 268 (Circumstantials ^ Subject). 2 Following Knud Lambrecht’s Information Structure and Sentence Form: Topic, Focus and the mental representations of discourse referents (Cambridge Studies in Linguistics, 71; Cambridge: University, 1994), and Robert D. Van Valin Jr. and Randy J. LaPolla’s Syntax, Structure, Meaning and Function (Cambridge: University, 1997).

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

42

of defamiliarised clauses (non-canonical word order) found in HPy.1 Deriving from the linguistic tradition of Summer Institute of Linguistics (SIL), Lunn embarks on a long-awaited study. He is not aware, though, that on the other side of the Atlantic, in a similar theoretical context, S.L. Fariss approached the matter of word order in HPy producing prima facie opposite results.2 He praised Knud Lambrecht’s theory of information as a more refined one and criticised Rosenbaum and Walter Gross for preferring a more traditional approach. He also argued that none of them was able to differentiate objectively between variation, which is purely poetic and that which is pragmatically marked. Although restricted to a particular group of texts or samples, none of the aforementioned authors attempted a more integrated analysis of the phenomena (p. 255-6). Hence, Lunn prefers an extended database to accommodate many texts from HPy and his interpretation of the data is very detailed and tabular. It is exactly here that the deficiency of Lunn’s method lies, because the texts in his database are too varied in terms of genre and date, a matter that he never addresses. Lunn noticed that HPy displays a higher degree of deviation from the regular VSO word order than BH prose. Lunn’s database included 1190 verbal clauses from Hebrew prose and 1243 verbal clauses from HPy, with a higher percentage of verbal clauses in HPy preferring a marked word order (34 per cent) than in Hebrew prose (only 14.5 per cent) (Lunn 2004, 7). The database of verbal clauses from HPy included, though, more than 7400 samples, drawn from more than 4000 verses of HPy, including Psalms, 1

This section was previously published in Journal for the Study of Religions and Ideologies 17 (2007): 145-148. 2 Working under the supervision of Longacre, under the auspices of the Summer Institute of Linguistics, Fariss (2003) engaged in a study that sprang out of the Herring and Hock survey of word order and syntax in old Tamil poetry and early Indo-European poetry, respectively (Susan C. Herring, Poeticality and word order in Old Tamil (pp. 197-236) and Hans H. Hock, Genre, Discourse, and Syntax in Early Indo-European, with Emphasis on Sanskrit (pp. 163-96), both in the volume Textual Parameters in Older Languages, edited by S.C. Herring, P. van Reenen and L. Schosler (Amsterdam: John Benjamins, 2000)). She was inspired to conduct her research on HPy by the lack of interest towards this particular discourse type. In order to discover word order choice and variation, Fariss selected thirty texts, of five different text types (speaking in textlinguistics terminology), and analysed them from the perspective of poeticality and transitivity. It appears that the probability for word order variation grows proportionately to transitivity.

PREVIOUS CONSIDERATIONS OF THE QYYQ VERBAL S EQUENCE

43

Isaiah 40-66, Job 3-14, Proverbs 1-9, Song of Songs, and Numbers 23-24 (cf. appendix 2 of his thesis). As becomes apparent from the ninth chapter of his thesis (pp. 196230), the equivalent proportions produced after studying a corpus of 112 verbal clauses confirmed the more general results. Here Lunn discusses only those verses with verbal clauses that display non-canonical word order derived from a database that includes Psalm 1 and 103, Job 12, Song of Songs 1, and Num 23.7-10. Seventy-four (66 per cent) of those clauses displayed a canonical (regular) word order, whereas the rest (34 per cent) displayed a non-canonical one. After criticising Buth (1992) for the insufficiency of his explanations and Shimasaki for his ignorance (rather unfairly picking a fight with him), Lunn concludes ‘as a general rule in HPy that the ordering of clause constituents in B-lines of parallel cola is not something governed by linguistic rules relating to pragmatic functions.’ (p. 106) Alternatively, he maintains that the word order variation is purely poetical, the technique thus being readily branded ‘poetic freedom’. This conclusion is based on the indecisive results of marked word order against the regular word order (330 vs 360), on the degree of variation (one or more preverbal constituents) and on the diversity of variations found in line B (various permutations of the constituents) (p. 106). As a result, one should look for poetic devices that influence a certain preference with respect to word order. Among the possible options, Lunn includes chiasm (pp. 108-109) and parallelism (pp. 110-111). Following Lambrecht, Lunn promotes Setting as pragmatically unmarked (pp. 56-60). Particles ~g, @a, qr, and %a are interpreted as focus particles that accompany marked clauses, the first two expanding the focus, the last two restricting it (pp. 64-71). Chapter 4 (pp. 61-95) argues for the usage of marked constituent order in HPy after the model of Hebrew prose. ‘Variation in word order figures prominently within the context of parallelism.’ (p. 96). Lunn advocates a distinction between a defamiliarised line structure (DEF) for poetical reasons and a Marked line structure (MKD) for pragmatical reasons (pp. 112). He also proposes several factors that serve to distinguish defamiliarised word order from that which is pragmatically marked. First, the literary environment (whether line B, text-boundary, or peak); second, the presence/absence of pragmatic connotations which require an explanation of word order with reference to topicality or focality; third, the presence of a focus particle hinting at a pragmatic function of the pre-verb

44

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

structure; and fourth, the extent of the variation, since the more divergent the word order is, the more likely it is to be the product of defamiliarisation (pp. 229-30). Pragmatic markedness is employed when introducing a new topic, contrast, replacing focus, specifying focus and other categories of focus (p. 230). Secondary lines of poetic units are said to display the highest level of freedom in terms of word order. The B-line paralleling a regular A-line is considered the typical place where defamiliarisation occurs (CAN//DEF) and is, therefore, called the ‘common environment’. Its counterpart (DEF//CAN) is rarer, therefore labelled ‘limited environment’ (p. 116). Occasionally, pragmatic marking co-occurs with poetic defamiliarisation. Since most samples come from Isaiah (44.16; 61.11; 64.9; 65.25a; 66.3-4 and Ps 44.9), Lunn explains this particular type of marked parallelism as a matter of style. Constituents that index setting or pronominal subject/object in the primary line regularly, do not appear in the secondary line (pp. 153-7). Lunn defends the priority of line A over line B by underlining the dependence of the secondary line on its antecedent in terms of topical reference, gapping, gender parallelism, even tense-shifting, pragmatic dependence (pp. 116-20). In this context, Lunn endorses even Buth (1986, 26) for describing tense-shifting as a purely poetic device, the verbal form of the secondary line being just a variation of its counterpart in the primary line. The verb in line-A establishes the temporal component of meaning that the verb in line-B follows. ‘This phenomenon is therefore simply another manifestation of defamiliar language characteristic of poetic genre.’ (Lunn 2004, 120). It is obvious that Lunn promotes the priority of poeticality over pragmatic markedness in HPy. Chiasmus (a rhetorical pattern) needs to be distinguished from contrast (a semantic relationship) since chiasm exists in HPy predominantly in synonymous relationships. Even the traditional category of antithetic parallelism, as defined by Lowth, becomes obsolete, because the construction of the secondary line in such a case is motivated pragmatically not semantically (pp. 129-30). The probability that the secondary line will be marked is higher when the primary line displays a pragmatically marked word order too (pp. 13351). In the case of a Marked A-line, defamiliarisation of B-line does not function freely anymore. The constrictive force of markedness constrains the author to draw attention to the equivalent constituent in the secondary line as well (p. 157-8).

PREVIOUS CONSIDERATIONS OF THE QYYQ VERBAL S EQUENCE

45

Although the majority of samples in Lunn’s database, constituting what he calls the ‘universal distribution’, follow the above rules, he noticed some exceptions too, including verse-lines with initial defamiliarised, Marked, or gapped structures. They are interpreted as playing other roles, such as signalling aperture, closure, and peak (pp. 161-94). Surprisingly, although he deals primarily with word order, Lunn delves into the study of clause relationships at the level of poetic verses, noting three main possibilities. Some verses link together two independent clauses (HEAD1 + HEAD2), others connect a subordinate clause by its regent (HEAD + Subordinate),1 or even have the units of the same clause extended over two lines (HEAD [Phrase1 + Phrase2]. When poetic verses extend the couplet, various structures are noted (Lunn 2004, 22-5). The greatest contribution of this work resides in the clear distinction promoted between poetic defamiliarisation and pragmatic marking. Lunn succeeded in applying a modern version of the theory of information structure convincingly to texts from HPy and proving the connection between parallelism and word order variation in HPy. His main thesis is that word order variation appears usually in line B of a verse. Defamiliarisation appears in line B of a verse given an unmarked line A and is by no means uniform. A marked line A prompts a marked line B as well. Gapping appears in line B, given that the correspondent constituent in line A is not marked. The pragmatic solutions surveyed thus far, go beyond the traditional understanding of the Hebrew grammars, at the same time engaging with the rhetorical aspect of language intricately connected with the nature of HPy. Envisaged as a domain of freedom in creativity, POETRY was considered to escape the limitations enforced on language through grammar. Opinions on the relationship between stylistics and pragmatics, a more recent embodiment of the traditional debate between form and content, vary from the dichotomy proposed by Givón to the integration of pragmatic functions into rhetorical structure promoted by Gross. Rosenbaum and Lunn advocate a middle way, according to which, in POETRY, word order serves entirely aesthetic purposes.

It should be stated here that there are situations when the subordinate precedes its regent clause. 1

46

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

2.3. S UMMARY The QYYQ verbal sequence has received different interpretations over the years. When it comes to POETRY, which has the most defamiliarised language, strict traditional grammatical explanations are insufficient to justify a particular chain of verbal forms, although a sense of particular historical development may provide some solutions. Rhetorical approaches indicate that the QYYQ verbal sequence appears as a rhetorical device in HPy, as well as in other cognate literatures, linked to the authorial intention of a given poem. Unfortunately, neither its function nor its inner workings have been explained satisfactorily by these methods. Rhetorical approaches were able to display, though, a particular awareness towards integrating the QYYQ verbal sequence with grammatical parallelism. Although couplets with QYYQ verbal sequence with the same root verbs were initially studied, Dahood extended the phenomenon to verbal sequences that are built on lexical verbs with different roots. Statistically, the q*f~l//y]qf{l verbal sequence is commoner than the y]qf{l//q*f~l one, even though the latter is commoner when only the same root verbal sequence is considered. Arguably, pragmatics has something more to offer by proposing a more in-depth evaluation of grammatical incidents. Being able to accommodate other aspects of language such as word order and pragmatic functions not accounted for in traditional grammars, we suspect pragmatics can offer more answers to the question posed by such a catena of verbal forms. The pragmatic approaches surveyed contributed to our discussion by noting that other linguistic forces at work in texts can explain the QYYQ verbal sequence. ‘Poetic freedom’ would justify the use of rhetorical structures such as chiasmus, pivot, terrace, staircase, parallelistic structures and the like. Pragmatic markedness would provide alternative explanations for cases of word order variations, which do not find a proper justification in aesthetics. Providing we can equip ourselves with a multi-levelled grammar that accommodates various categories and integrates traditional grammar with pragmatics, we anticipate that suitable explanations for the usage of such an exceptional verbal sequence in HPy can be found. Since we have subscribed to the primacy of texts right from the outset of our undertaking, the following chapter will deal in more detail with the corpus of our research.

3 THE STRUCTURE OF POETIC VERSE AS PRESERVED BY THE MASORETES At the outset of this chapter, several fundamental outcomes of the previous chapters are outlined. Firstly, it has been proposed that the text as preserved by the Masoretes will be in the focus of this study. Textual issues that may arise are taken into account, inasmuch as better readings, i.e. those supported by better traditions, are suggested. These informations are included as footnotes with reference to the text segments in discussion (cf. Appendices 1 and 2). Secondly, this study will consider only couplets with the QYYQ verbal sequence in the Psalter. This assumption is grounded in the agreement reached by scholars to consider the QYYQ verbal sequence as integrated in the parallel structure of the verse. Consequently, monolines, couplets with verbal ellipsis, and all fragments that do not exhibit parallelism of thought due to their fragmentary state of preservation (mainly with reference to UPy) are discarded. Although there are verses with a larger number of lines displaying the QYYQ verbal sequence integrated in their poetic structure, they are not considered either. Since attention should be given to the Masoretic accents which are present in the Hebrew manuscript used, an evaluation of this tradition is in place here. Then, the contribution of the ‘delimitation theory’ for the verseline delimitation practice will be investigated. Deriving from the assessment of the Masoretic tradition and its modern application to biblical studies of HPy, we will outline several methodological steps that will allow us to engage with the HB as preserved by the Masoretes. Before we conclude this chapter, several modern works are surveyed and the contribution of their authors to a fresh reading of the Masoretic text is assessed. The natural result of such a survey is a database of the Psalms divided into verse-lines (cf. the excerpts in Appendix 9)1, as well as 1

The database is divided into five sections according to the traditional division 47

48

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

the selection of couplets displaying a QYYQ verbal sequence (Appendices 1 and 2).

3.1. THE MASORETIC T RADITION OF VERSE-LINE DELIMITATION Marking the verse-line of HPy in biblical MSS by various devices has a long history. Due to a very conservative approach to the Biblical transmission tradition, such practices were reluctantly accepted into the Pentateuch and Prophetic corpus. They found a more fertile ground in the tma books instead.1 3.1.1. The relevance of the Masoretic accents Initially, as some of the Qumrânic, early Greek (e.g. Codex Sinaiticus), and Latin (Vulgate) MSS indicate, the use of unwritten areas (blanks, spaces and gaps) was successfully employed to delimit adjacent verse-lines. Such an awareness of poetic rhythm, that must have inspired verse division by means of blanks, is carried forward by other textual traditions with roots in the third century A.D. (Codex Alexandrinus, Peshitta, The Samaritan Pentateuch), which preferred to use a variety of symbols for similar reasons. 2 The masterminds of verse delimitation were the Masoretes, even though pre-Masoretic MSS display an educated awareness of poetic structure too. A large number of MSS from the tenth century onwards (e.g. Codex Aleppo, Codex Leningradensis), possibly reflecting much earlier established textual traditions, converge in their interest to maintain a standard of marking verse-lines in poems, both within and outside the tma books. By now the complicated, rather aesthetic, arrangements (columnar, chessboard and their variants) promoted by the rabbinic tradition found their place within the larger textual tradition of biblical literature, competing against more literary approaches to verse division. Whether the division of verses into verse-lines in one tradition followed another cannot be said without a comparative evaluation of various literary traditions. One would have to consider carefully the division of each poem into its respective verse-lines according to each textual of the Psalter into five books. 1 The Hebrew acronym tma, constructed from the initials of their respective names bwya, ylvm, and ~ylht, stands for ‘truth’. Therefore, they are also known as the Books of Truth. 2 For more details and the supportive relevant data, cf. Tatu (2007a).

THE STRUCTURE OF POETIC VERSE

49

tradition and only then compare the results for more precise conclusions on this matter. For their quality both in terms of material preserved and coherent verse division applied to the text, some of the aforementioned witnesses will be included in the list of authorities we consulted when applying verseline delimitation to the Psalms. Codex Sinaiticus (Lake and Lake 1922) and the Vulgate (Marcotte de Sainte-Marie 1954) are known for their antiquity (produced during the fourth century A.D.) and for the versification technique they used, each new verse-line starting with a new row. Codex Aleppo, earlier by almost a century than Codex Leningradensis (L), is somewhat fragmentary, lacking Pss 15.1-25.2, but its age and quality, as assessed by scholars, is valued (Goshen-Gottstein 1960; Loewinger 1960). Psalms displays here a columnar pattern (Goshen-Gottstein 1976). The mediaeval tradition, as preserved by the Ginsburg Hebrew Bible (1894), is also included in the corpus. Biblia Hebraica Stuttgartensia represents the main core of the text against which all the other sources are compared (cf. Appendix 4). Dotan (2001, xvi) calls our attention to the fact that the original of L is less reliable where the Masoretic use of accents in the Poetical Books is concerned. As Yeivin (1968, 279-356) also noticed, the ancient scribes did not standardize the use of accents in the poetical books. Since there is no unified biblical text, on which there is general agreement, the importance of accents is limited to a consultative role. As Waltke confessed, such a situation should not feed our scepticism but it offers enough ground for ‘cautious confidence’ instead (Waltke 1989, 25). As it is known, Masoretic accents are provided in all biblical books with two different, though similar, sets: one for the three poetical books (Job, Psalms and Proverbs – abbreviated to The Three Books), and another one for the prose books (abbreviated to The Twenty-One Books). Each set presents a group of nine conjunctive accents and a larger group of disjunctive accents (twelve for the Three Books, and eighteen for the Twenty-One Books). It seems that the use of Masoretic accents in the Three-Books is more complex than in the Twenty-One.1 In the case of the Masoretic accents used in the Twenty-One Books, Revell proposes to classify the disjunctive accents into four different categories, according to their disjunctive power. This is a simplified scale derived from Wickes’s five rank hierarchy, which displaced the older pictorial hierarchy of Samuel Bohlius (seventeenth century) and its later variants. Price (1990, 27) changed the name of the ranks from Emperors, Kings, Dukes, Counts and Servants into hierarchical 1

50

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

From a theoretical point of view, Price (1990, 11) identifies four purposes that the Masoretic accents serve at different semantic levels. Firstly, they mark the primary stress of a word or word-cluster (words joined by m*Q}p). It is noted that secondary stress is not as frequent as primary stress, though present and marked by m#t#g or, very infrequently, by similar accents (Price 1990, 2-5).1 This is a phonological function. Secondly, they indicate the degree of separation/connection between adjoining words/phrases/clauses at the syntactical level.2 It is obvious that versets3 end in all circumstances with s]LWq announcing the sop P*sWq.4 In addition, versets are divided into verses and verse-lines by means of various disjunctive accents.5 Conjunctive accents associate words belonging levels I-V with the downgrading of Servants, which included the conjunctive accents. Both Yeivin (1980, 159) and Revell (1992, 595) have a similar four level hierarchy, with the difference that sop P*sWq was assimilated to s]LWq and l+g^rm@H joined the other disjunctive accents from the lowest hierarchical level. Besides sop P*sWq/s]LWq, high frequency is noted for f]pj* (11285), z*q}p q*f~n (6992), A^tn*j (5483), and P~vf* (5429). Most of the other disjunctive accents display a frequency below 1000. 1 This function is accepted by Dotan (1971, 1453) but is contested by Dresher (1994, 5). 2 Actually, Waltke (1989, 26) concludes his stand for the Masoretic accents by affirming that ‘at present it is best to consider the accents as an early and relatively reliable witness to a correct interpretation of the text’. Cf. also Dotan (1971, 1453), Yeivin (1980), Price (1990, 8), among others. 3 In order to avoid terminological ambiguity, we prefer to use ‘versets’ for the canonical basic divisions of the biblical texts. Verses are considered the poetical basic units, regularly identical with couplets, but occasionally coincidental with monolines or tercets. 4 This is particularly clear when the presence of sop P*sWq and s]LWq are compared according to the Numerical Summary of the Disjunctive Accents given by Price in Table 3.1. Their computation in the Pentateuch, run by a computer software, are exactly the same, 5852 to be precise (Price 1990, 23). 5 This technique came to be called ‘dichotomy’. Due to its lack of specificity, when exceptions are exposed, de Hoop favours a variation. Particularly where PROSE texts are considered, since they are the only ones which follow the rules, he identified two variant structures of the regular accentuation paradigm: a ‘dichotomic’ and a ‘trichotomic’ structure. The numeric prefix of the terms refers to the number of textual units into which the verset is divided by use of disjunctive accents. When referring to trichotomic structures he distinguishes two more classes: broken chain (the first two disjunctive accents are inferior in grade to the third) and descending chains (the accent of the highest disjunctive grade is the

THE STRUCTURE OF POETIC VERSE

51

to the same semantic unit, whose margins are marked by two disjunctive accents. Thirdly, they indicate the relative intonation of a word in cantillation, related to the public liturgical reading of the HB in the synagogue.1 Fourthly, they reflect the poetic structure of a text. Price even challenges the position assumed by many Hebrew scholars, in which they disregard the contribution of the Masoretic accents for the economy of the poetic structure, by bluntly expressing his confidence, supported by an increasing number of authorities,2 that in good poetry, grammatical syntax and poetic structure exhibit considerable harmony. Where such harmony fails, it should not be surprising to find the accents being influenced at times by the rhetorical demands of the poetic structure. (Price 1990, 17)

third, the accent of the lowest disjunctive grade being the first). See (de Hoop 2000, 68). 1 Cf. Dotan (1971, 1453). Although Jewish scholars do not accept unanimously this opinion, there are some who tried to prove that the Masoretic accents describe a sequence of tones, which is to be accommodated by the cantor according to the number of syllables is a particular word. Besides, despite the mathematical rigidity that governs the Masoretic accents, they display enough flexibility to fit phonetic and rhythmic requirements. Actually, Herzog admits that their flexibility is determined by various criteria such as the biblical text itself, the liturgical circumstance, the medium of performance, the regional stylistic traditions, the traditional interpretative notes mentioned on the margins (Herzog 1972, 1104). Therefore, there is reason enough to suspect that the same text will not be cantillated in an identical fashion twice, not even by the same cantor. It is known that some Jewish traditions, such as the Ashkenazi community in Rome, preserved even some hand movements encoding the tonality of the accents (Herzog 1972, 1099 quoting I. Adler in J. Porte’s Encyclopédie des musiques sacrées, vol.1, 1968). Pentateuch cantillation is also reproduced by Johann Boeschenstein in J. Reuchlin’s De accentibus et ortographia linguae hebraicae (Hagenau, 1518) and, later on, by A.Z. Idelsohn (Melodien, vol. 2, 1922). J.H. Eaton (1984, 97) appreciates as a brave achievement Vantoura’s reconstruction of original cantillations on the basis of Masoretic accents, who assumed that the Masoretic accent system is a musical notation descended from Old Testament times. While the signs beneath the letters are considered to denote degrees of the scale, the signs above are plain ornaments. Rhythm is assumed to follow the syllables (see Vantoura 1978). 2 Such as Aronoff (1985), Yeivin (1980, §178), Watson (1984, 97-102); Christensen (1985, 1987, and 1989), Dresher (1994).

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

52

This prosodic representation by the Masoretic accents can be exploited with some benefit for verse-line delimitation. Noting that syntax is frequently divergent from the prosodic representation of Hebrew texts, and that phonology follows prosody, Dresher too concluded that Masoretic accents must have had a strong prosodic function (1994, 8). 3.1.2. Rul es of Engageme nt with the Masoretic Accents Following the comprehensive study of Price on the Poetic Accents, we noticed that the most frequent disjunctive Masoretic accents in the tma books are the following (frequency rates in brackets): s]LWq (4465), a^tn`j (4216), r+b'^U (3185), and D+j' (2684).1 The semantic domain marked by inferior disjunctive accents is included (‘embraced’) in the ones marked by superior accents.2 Hierarchy I II

III

IV

Disjunctive s]LWq r+b'^U m%gr*v a^tn`j Uol? w+yor@d D+j' x]Nor r+b'^U G*dol P*z@r l+g^rm@H

Near r+b'^U m%gr*v D+j' D+j' x]Nor l+g^rm@H l+g^rm@H l+g^rm@H l+g^rm@H l+g^rm@H

Remote a^tn`j/ Uol? w+yor@d r+b'^U G*dol r+b'^U G*dol r+b'^U G*dol P*z@r P*z@r P*z@r empty empty

Table 3.1: Hierarchy of Masoretic accents in the Poetical texts (Price 1990, 167)

Price postulates in his ‘law of conjunctives’ that the disjunctive accents are closely associated with conjunctive accents, so that there is only one conjunctive to serve a given disjunctive (Table 3.2). Any given prosodic unit would then be made of not more than two words. If more were present, m*Q}p was employed to connect two adjoining phonetic-units (clitization). Uol? w+yor@d is less frequent (412), of which over 85% of cases are recorded in the Psalms (352). 2 It is to be noticed that sometimes v~lv#l#t G*dol replaces r+b'^U m%gr*v (23 times in the Psalms, that is 74% of the total number of occurrences) and r+b'^U q*f~n replaces x]Nor (23 times in the Psalms, that is almost 79% of the total number of occurrences). Similar conclusions were reached earlier by Dotan (1971, 1456-7). 1

THE STRUCTURE OF POETIC VERSE

53

Price promoted another rule applicable to the disjunctive poetic accents. Some accents, presumably under musical restraints, are subject to transformation, as happens with r+b'^U, which may turn into P~vf*. This is ‘the law of transformation’ (Price 1990, 170-71). A tree-like ramification of prosodical units has been defended since Wickes, even though his approach was rather syntactical or semantical. 1 By what Wickes termed as ‘the law of continuous dichotomy’, he referred to the ability disjunctive accents have to divide a verse into two dichotomous segments, that can be further divided into smaller units, until dichotomous division can no longer take place. 2 Arguably, this principle may be of more value in PROSE texts (Wickes 1887, 44-60). Master disjunctive accent

Slave conjunctive accents

s]LWq r+b'^U m%gr*v v~lv#l#t G*dol A^tn*j Uol? w+yor@d D+j' x]Nor r+b'^U q*f~n r+b'^U G*dol P*z@r a~zl* l+g^rm@H m+h%PP*e

mWn*j/m?r=k*/ u]LWy m?r=k* m?r=k* mWn*j/m?r=k* G~lG~l/m~hP*k mWn*j mWn*j/m?r=k* m?r=k* u]LWy/m~hP*k/x]Nor't-m~hP*k G~lG~l u]LWy/m~hP*k/x]Nor't-m~hP*k none

Number of accents permitted 0-1 0-1 0-1 0-1 0-1 0-1 0-1 0-1 0-1 0-1 0-1 0-1

Table 3.2: Master disjunctive accents and their slave conjunctives (Price 1990, 169)

Dresher (1994, 8-16) noticed the similarity between the modern hierarchy and the Masoretic hierarchy in terms of the prosodic hierarchy, seeing the Phonological word as its basic cell, organized in successive superior levels into Conjunctive phrases, Disjunctive phrases and Biblical For him, the placement of Masoretic accents was decided by logical and syntactical rules, though triggered by musical requirements. Therefore, they convey pedagogical value for their respective readers in terms of musical notation, grammatical marking and logical signal. 2 Janis (1987) prefers an opposite approach, beginning with lower-level groupings of words that are combined into higher-level phrases (cf. Dresher 1994, 29). 1

54

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

verses. He asserts that, unlike modern hierarchy, organized according to the principle of Strict Layer Hypothesis,1 the Masoretic hierarchy is neither rigidly subordinative nor uniform. Instead, it allows nesting of phrases, proof of a recursive prosodic character (Dresher 1994, 22-31), and encourages diversity (pp. 31-7). Wickes also noticed ‘the tendency that any other constituent that will fill the verb slot (initial position), except the verb itself, will be set apart from the rest of the verset by a disjunctive marker’. This is the case with subjects, objects, adverbs and vocatives (1881, 39-43). When more regular constructions are taken into account, such as verbal or nominal clauses, primary dichotomy will preserve the first members of the clause as part of the same member, allowing secondary constructions to refer to them (1881, 43-4). In the POETRY of the Pentateuch, parallelism seems to have functioned as ‘the reigning principle of division’ (hence logical in nature), supporting not only the major dichotomies but minor ones as well. Obviously, the major dichotomy is not relevant as a term for the Poetical Books where it is replaced by caesura (Wickes 1887, 29-31 and 38-40).2 He admits that parallelism is not always as exact as the hierarchical disjunctive accents may demand. When the parallelism is exceptional, caesura will fall after the main idea of the verse (whether it is echoed/repeated or not in the following verse-line), or after the secondary line (in the case of a tercet). If no logical pause exists, the caesura will be placed in accordance with the syntactical pause, i.e. at the border between groups of words related in sense and construction (Wickes 1881, 25-9).3 1 Abbreviated SLH, Strict Layer Hypothesis is a prosodic theory of hierarchy that assumes that any prosodic unit at a given level of hierarchy consists exclusively of units at the next lower level of the hierarchy, as proposed by Elisabeth O. Selkirk (The Syntax of Words, Linguistic Inquiry Monographs, 7; Cambridge, MA: MIT and Phonology and Syntax: The Relation between Sound and Structure, Current studies in linguistics series, 10; Cambridge, MA: MIT). 2 Cf. Wickes (1881, 24-30). 3 Caesura will not overlap with the logical or syntactical pause except in two special cases. First, when under the drive of musical equilibrium, it will move towards a convenient resting-place (e.g., Job 2:2; 4:8; 10:8; Ps. 48:4; 72:20; 74:5; 92:9; 102:8; 116:1; 119:18, 20; 120:1; 122:3; 145:5; Prov. 21:4). Second, when oral formulaic phrases appear (e.g., ‘he said’, ‘he said to himself’), they are not divided from the rest of the verse-line which they front (Job 33:24, 27; 34:9; Ps. 10:11; 12:5; 16:2; 60:8; 83:13; 89:20; 110:4).

THE STRUCTURE OF POETIC VERSE

55

Against Wickes, Price favours the idea that syntactic strings are not binding enough, but instead ‘whenever grammatical syntax and poetical structure fail to harmonize, the accents usually agree with the poetry rather than the syntax’ (1990, 288). Dresher (1994) reaches a similar conclusion at the end of his phonological study, defending the prosodic representation as the best description of the Masoretic accents. Wickes advocated fixed rules for the use of the major disjunctive accents that divide the verset into its respective verse-lines. Thus, given a verset ending with a s]LWq, Uol? w+yor@d can occur only in words placed at least four slots away from s]LWq, but it prefers positions in the sixth word from s]LWq or further. It will never appear, however, within the first word of a verset. The accent a^tn`j can occur on the fourth and fifth word from s]LWq, but also in the other slot closer to the end. There is, however, a tendency for a^tn`j to be replaced by r+b'^U m%gr*v when filling the final slot (Wickes 1881, 30-35). This means that versets longer than four phonetic-units manifest the tendency to be divided both by Uol? w+yor@d and a^tn`j, whereas shorter versets prefer a single major caesura. Although Wickes preferred to differentiate between Uol? w+yor@d and a^tn`j in terms of hierarchical rank, as though the former were superior over the latter, Price favours the opinion that the two disjunctive accents are of equal rank. As proof, he refers to the fact that, in over 65 per cent of all the occurrences of Uol? w+yor@d and a^tn`j together, the two are of equal rank, in most cases dividing the verset into thirds. When a verset consists of more than three verse-lines, Uol? w+yor@d governs more verse-lines than a^tn`j, and r+b'^U m%gr*v may be involved as well. There are, however, examples when a^tn`j is in the majority (1990, 17981). Therefore, they are to be considered as hierarchically equal, as superior disjunctives marking the end of verse-lines. It became apparent that the disjunctives with great value for the verseline delimitation technique are Uol? w+yor@d, a^tn`j, r+b'^U G*dol and P*z@r. Although Masoretic hierarchy can support syntactic and/or semantic representations of verse structure, a prosodic representation seems to be more consistent with the variety of cases and the inner ability of the Masoretic accents itself.

3.2. THE ‘DELIMITATION THEORY’ Under this name, an older interest in the relevance of ancient practices for the transmission of texts is revived, in our case, texts of Canonical

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

56

persuasion. Korpel and de Moor initiated the new ‘delimitation theory’ group, but Revell and Tov are probably the ones who resumed this tradition from the Masoretic seminars. This group derives its roots from scholars such as Maori, Josef Oesch and John Olley and has, as its object of study, the division of biblical texts into sense units, whether larger or smaller.1 Oesch is credited with promoting the technique of delimiting pericopes in the MT under the name of Gliederungskritik (Oesch 1979, 366). However new it may appear to be, the study must have a venerable history, since its object deals with textual elements present in the MT, namely P+fWjoT and s+fWmot.2 Also known as ‘open sections’ and ‘closed sections’ respectively, the Hebrew graphemes that stand for them, p and s, divide the MT into textual chunks, occasionally co-extensive with the liturgical reading units. A z!^j is a variant of s+fWm>, that stands for a blank indentation when preceded by a P+fWj>. Although not present in the Qumrân MSS, a connection between the scribal tradition at Qumrân and the Masoretic tradition in terms of the usage of spaces seems very plausible (Maori 1982). Olley (1983, 29-34) noticed that P+fWjoT and s+fWmot are used in Isaiah (1QIsaa) for marking introductory phrases that fit into five groups: (1) the speech of YHWH, (2) summons to hear, (3) designation of time, (4) exclamations, and (5) oracles against nations. He concluded by cautiously admitting that these markers ‘are not definitive for modern exegesis and reading’ but they are to be understood as ‘features that provide insights as to how the text was understood by early readers’ (Olley 1983, 49). 3.2.1. Marjo C.A. Korpel and Johannes C. de Moor Korpel and de Moor have striven to identify successfully both disjunctive and conjunctive artifices in order to objectivize the process of delimiting the subdivisions of a poem. They noted that a verse-line (called by them a colon) is delimited by the Masoretic dividers s]LWq, a^tn`j, and z*q}p q*f*n (for the prophetic books). Its cohesiveness is provided by internal For a survey of the method, see Korpel (2000). Interest in this matter is by no means new. First noted by L. Blau (Papyri und Talmud in gegenseitiger Beleuchtung, Leipzig, 1913, p. 15), it continued to impress other authors such as H. Bardtke (Die Parascheneinteilung der Jesaiarolle I, pp. 33-75 in Festschrift Franz Dornseiff zum 65. Geburstag, Leipzig, 1953), C. Perrot [Petuhot et setumot: Études sur les alinéas du Pentateuque, Revue Biblique 76 (1969) : 50-91], and J. P. Siegel (The Scribes of Qumran: Studies in the Early History of Jewish Scribal Customs, with Special Reference to the Qumran Biblical Scrolls and to the Tannaitic Traditions of Massekheth Soferim, PhD diss., Brandeis University, 1971). 1 2

THE STRUCTURE OF POETIC VERSE

57

parallelism (when present) (Korpel & de Moor 1998, 10-11). Verses (‘verselines’ in their terminology) are delimited by a chain of Masoretic accents high in rank but of descending order (idem, p. 12). For them, the number of stresses functions as a guiding principle and not a definitive tool to allow amending the text metri causa. As a result, they share an infra-Margalitan position on the rhythmical balance of poetical verses. Ellipsis and parallelism function cohesively at the level of verses. Major division markers, such as P+fWjoT, s+fWmot, z!^j, capital letters, or diamonds, delimit strophes and stanzas (called canticles). Consequently, the distinction between strophes and stanzas becomes blurred. Parallelism of ideas at the level of many verses and thematic unity (words of the same lexical field) function as cohesive principles, even at the levels of strophe and stanza (Korpel & de Moor 1998, 12). Parallelism seems to be, for many other specialists, the main cohesive element of ancient POETRY. A more revolutionary contribution is made by Korpel and de Moor when they consider that the deliberate changing of word order is supposed to mark the boundaries of a strophe (p. 14). The assumption of Korpel and de Moor that the verse-line is the basic structural unit of a poetic text contradicts the one shared by Margalit and Fokkelman. For them, the meaningful form for the latter is the verse, which stands as a poetical unit. None of these scholars identified a plausible mechanism for verse-line delimitation except parallelism, although both schools are agreed on its cohesive function. Korpel continues to prescribe ‘delimitation criticism’ as an intermediary step between textual criticism and any of the other criticisms (literary, rhetorical, or redactional).1 Compared to previous presentations of the Hebrew verse, the terminology used for the poetic units exhibits several changes.2 Korpel, in particular, and the Kampen School, in general, applies this method not only to poetic texts, but also to all biblical texts. Therefore, any text can be divided into cola, which are units consisting of one or more feet that constitute a clause or a group of clauses. They are differentiated by means of major Masoretic disjunctive accents, identified with those listed Actually, the Pericope Project, launched by Korpel and Oesch because of such assumptions, aims to collect data about unit delimitation in ancient MSS of the Bible and make it available to the public by means of the fifth edition of the BHS. 2 This time, more general terms are preferred instead. «Paragraph» replaced the former «canticle» and «macrostructural unit» replaced the former «sub-canto». Moreover, «cantos» can be grouped together into «cantata». 1

58

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

up to number 7 on the BHS table. When a colon consists of only one foot, the Masoretic division has to be doubted and the marker credited with a different meaning. At all times, parallelism has to be considered as the basic rule of thumb for biblical texts (Korpel 2000, 30). 3.2.2. William T. Koopmans De Hoop (2000b, 70-74) assesses the effort of Dutch scholars to achieve the integration of the Masoretic accents into colometric studies as passing through a chronological development. Strategic importance is assigned to Koopmans’s study on the poetry of Joshua 24 (Koopmans 1990), which ‘led to a reappreciation of the Masoretic accents’ (de Hoop 2000b, 75). The necessity for a reappraisal of this method was called for, since serious critique of it revealed several causes of failure: there is no explanation or discussion of the system of accentuation offered; the representation of the disjunctive accents is generally incomplete; and disjunctive accents are not interpreted in a consistent manner (de Hoop 2000a, 54). Koopmans added Masoretic accentuation in support of the Semitic verse. His theory postulates that ‘within a verse-line the order of magnitude of accents is always increasing’, and ‘a distinctive accent of a lower magnitude than the preceding one usually indicates that a new verse-line is starting at that point’ (Koopmans 1990, 178). The representation of the accents was intended to treat the Masoretic accentuation as a guiding principle, not as an overriding rule. 3.2.3. Raymond de Hoop De Hoop himself subscribes to the Kampen School when describing the methodology to be followed in his monograph (1999, 84 ff). Masoretic accents are reliable for the division of the texts because, although of mediaeval origin, they reflect ancient traditions. Then, larger units are marked by means of s+tWm> and P+tWj>. These represent the ‘formal criteria’ that are used to structure the text. At a lower level, the Masoretic text displays two more systems of text division: sop P*sWq and other accents divide the text into verses and its subunits, as well as pausal forms divide the text into semantic units (cf. Revell 1980, 168-9).1

1 For de Hoop and the Kampen School of colometry, the «colon» stands for a «verse-line» in our terminology and the Jewish School through Margalit. Similarly, a Kampen «line» should be equated with our «verse».

THE STRUCTURE OF POETIC VERSE

59

In the acrostics of the poetical books, when a verset is longer, its respective verse-lines (three in number) may end in the following way: r+b'^UG*dol, m+h%PP*e l+g^rm@H or a^Zl* l+g^rm@H, and s]LWq respectively (see Ps 25.5) (de Hoop 2000a, 68). a^tn`j is used to mark the main division of the second half of a longer verse. To summarise, the Masoretic accents valued for their disjunctive role are: so[ P*sWq preceded and anticipated by s]LWq, which together mark the end of each verset, Uol? w+yor@d and a^tn`j. 3.2.4. Ernst J. Revell While investigating the structural role of pausal forms in Biblical poetry, Revell (1981) noticed from a glance at HPy in BHS layout that, in PsalmsJob-Proverbs, most frequently pausal forms split versets into two verse-lines. He then applied the use of pausal forms to poetical division into lines to less common verse patterns, the so-called double couplet and the unbalanced couplet, although some may interpret the latter as a tercet. His theory proposed that a pausal form was used to mark ‘the logical point of division’ between the two parts of a verset. In order to support the Masoretic recommendation of reading some versets as unbalanced couplets, Revell drew attention to the antiquity of this technique (1981, 192-6). It must have been in place by the time the earliest version of the HB including selah appears (i.e. Jerome’s Vulgate), because it was attached after the pausal form (with the exception of Ps 3.9). Besides, Psalms from Qumran written in verse-lines show a consistent tradition with the Masoretic accentuation (4QPsb, 8Q2, 5/6 Hev. Ps, Mass Ps, and mostly 11QPsa). To these, Revell added examples from Akkadian poetic literature. The use of pausal forms indicates the Masoretic perception of verse structure in the tma and reveals the use of four patterns: the simple couplet, the tercet, the double couplet, and the unbalanced couplet (Revell 1981, 197)1. Revell also noticed that pausal forms occur preferably with some of the major disjunctive accents, such as s]LWq, a^tn`j, and Uol? w+yor@d (Revell 1980, 165). The same major accents divide the poetic versets into two halves: if shorter, by a^tn`j, if longer, by Uol? w+yor@d. Whilst the first half of a shorter verset is divided by D+j', the second half is divided by r+b'^Um%gr*v (Revell 1992, 596).

1

For exceptions, though, see the Addenda to Revell’s article (pp. 198-9).

60

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

3.2.5. Jan P. Fokkelman Fokkelman’s treatment of HPy is an original application of Korpel and de Moor. Although he interacted with HPy before 1990,1 Fokkelman’s interest for this literary genre in the HB became increasingly obvious, starting with the publishing of his first informal article on Psalm 68. Although he acknowledges the influence of several specialists in the area (all Dutch except Watson), he is not interacting with them at all. Terminologically, he used technical terms, such as verse, strophe and stanza, charged with different meanings than those in Watson. Fokkelman did not engage in describing his method,2 but subscribed to the Kampen School of literary analysis: The model says, very briefly, that a single line or colon mostly consists of two to five beats, that the full verse mostly has two or three cola, that two to four verses constitute a strophe, and that two to four strophes come together in a stanza. For the huge majority of psalms this is enough, but the longest poems have one more level, which I call sections (p. 73).

Fokkelman’s christening of his method as ‘structural analysis’3 was severely criticised by Vogels (1999 and 2001) but with more precision and intensity by Polzin (1989). Despite this criticism, Fokkelman kept his 1 Fokkelman published previously the first volume of a comprehensive four volume literary study of the Book of Samuel and discussed the problem of poetry when tackled the poems found within its pages. The Song of Hannah (1 Sam 2:110) is analysed only in the fourth volume of the series, pp. 73-111]. Particularly unique among his studies is the rendering of the stresses on lines (see p. 82). 2 He preferred instead to engage with the text, allowing the reader to guess his method from his application of it. The article published in 1995 was not much better in this regard. 3 A good example of a semiotic analysis (also called structural analysis) is offered by J. Holman on Psalm 138, published in the very same collection of essays (pp. 84-100). Fokkelman must have heard it at the conference, but did not prove to get to understand it either the concept or the methodology involved. He persisted in using the same misleading terminology for the rest of the literature he produced during the following years (1995, 1998, 2000, 2002, 2003). The archaeology of the text is defined differently by semioticians: they get involved in searching the structure of the contents, not its expression (Holman 1990, 87). Although Vogels’ criticism is not entirely correct (according to Greimas general outline of generative trajectory, cf. Holman 1990: 88), he was right to stress that Fokkelman’s study does not fit in the field of structuralism at all, not even at the surface level. Structuralism is something completely different to Fokkelman’s enterprise.

THE STRUCTURE OF POETIC VERSE

61

confidence in the ability his method has to allow the user to group correctly the lines into larger units, to understand words (in relation to their word pairs), perceive the unity of a poem (avoiding deletions and emendations), and support the authorship of a single author (as opposed to the suggestions of source criticism). The time for a proper methodological theorization was delayed until 1998 when the first work in a series of three on the Psalms was published. Building on its association with the twelve step scale of organization of the narrative text, POETRY would be organized in a similar fashion on a two step shorter scale, that includes lower units (sounds, syllables, words, phrases, cola and verses) and higher units (strophes, substanzas, stanzas, sections and poem) (Fokkelman 1998, 4).1 Unfortunately, the mixed aspect of the scale, explained as the fusion of language with art in artistic texts display, is neither correct nor convincing. A grammatical analysis has its own identity for specific methodology and terminology. A literary analysis applied to POETRY gets involved with such issues as metre, rhyme, figures of speech, verse, strophe, repetition, etc. The literary discourse does not consist of sounds, syllables, words, phrases or sentences, although some figures of speech can appear at that level too (alliteration, parallelism, repetition, etc.). Here lies the superiority of Margalit’s approach to POETRY. Although Fokkelman derived his method from the Kampen School, he disregarded the proposal of scholars like Korpel and de Moor that the foot is the smallest poetic unit.2 It may be only a terminological matter but it certainly counts, because the nature of the unit is completely different. Moreover, unlike the Kampen School scholars, Fokkelman counts syllables, a technique preferred by the American school (Cross and Freedman).3 The procedure of counting syllables is explained in detail only in the second volume of the series Major Poems of the Hebrew Bible (2000). He had already made clear that ‘counting syllables does not serve as metrical theory, or as a replacement of it. It is descriptive, rather than prescriptive (…)’ 1 Later reduced to nine: sounds > syllables > words > versets > verses > strophes > stanzas > sections > poem (Fokkelman 2001, 30). 2 In a more recent work, the poem subdivisions are considered foot > cola > verse-lines > strophes > canticles > sub-cantos > cantos (Korpel and de Moor 1998, 16). 3 Since 1978, de Moor has promoted the method of counting feet in HPy and UPy, the foot being identified as ‘a word or a cluster of words bearing a main stress’ (de Moor 1978, 139).

62

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(Fokkelman 1998, 20). If so, emendation of the biblical text is by no means permitted as he frequently did, in order to preserve the average unaltered. A meticulous counting of syllables and a sound verse-line delimitation are preliminary before promoting a syllable average as standard for HPy. The rule applied by Fokkelman to foster the belief of regularity in the Hebrew verse is X : Y = Z, where X is the average number of syllables of a poem, Y is the number of its lines, and Z is the average of syllables per verse-line. The average number of syllables/line, normative for the Psalter as a whole, is 8. Variations are accounted for, though not less than 7 and not more than 9 (2000, 46). When counting syllables, Fokkelman assumes the syllables of the preMasoretic text. That means skipping semivowels (shewa, both simple and compound), considering segolates as monosyllabic words, reducing nouns containing a laryngeal to a monosyllabic unit and ignoring the auxiliary vowel (often the hiriq) in yw[ nouns. Besides, verbal forms originally monosyllabic are accepted as such (although they were apocopated in the process). Finally, the so-called secondary dedoubling (virtual gemination) is not accepted, unless in Pi (including here Pi forms of geminate roots). Since Masoretic accents are not considered reliable, whenever the syllable count contradicts them, Fokkelman prefers to ignore the accents (cf. the treatment of Ps. 119.107-108 in Fokkelman 2000, 33-34). As opposed to Collins (1978), O’Connor (1980a) and others, Fokkelman does not accept the line as the basic unit of POETRY for two main reasons. First, the line is not always coextensive with the sentence, and parallelism is considered a constraining force with cohesive power, because of which lines cannot exist individually (except the monoline) but only in structures of two or more lines. Such a critique is justified, the verse having to replace the line as the basic meaning unit of POETRY. Second, since the original bards accepted the verse as the basic literary form responsible for carrying the meaning of acrostic poems (except Psalms 111 and 112), Fokkelman held the view that they must have considered verses as the main poetical units. The division of poems into strophes and stanzas is a process in which internal characteristics of verse-lines are carefully considered: grammatical parallelism, balance of syllables per line, and internal structures (especially chiasmus). Sometimes, the inner cohesive elements are more visible. When they are not, it is necessary that the unit be considered in relation to the adjacent ones (Fokkelman 2000, 37-40). After surveying 85 poems from the Psalter, Fokkelman (2000, 41) concludes that a coherent image takes shape:

THE STRUCTURE OF POETIC VERSE

63

each line has 2-4 beats (words), each verse has 2-3 lines, each strophe has a 2-3 verses, each stanza has 2-3 strophes.1 Fokkelman’s method derives from the rather generous category of functionalist approaches, although he found escape in the broader category of Literaturwissenschaft (Fokkelman 2000, 52). Following Givón’s recommendation (1984, I: 243 ff), we are in the position to suggest a better approach to Fokkelman’s method. The structure of literary texts can be better described in terms of coherence relations between two propositions in a discourse. Discourse usually has a deeper, hierarchic structure: story > chapter > section > paragraph > proposition, which we infer may be equated in POETRY with the following schema: poem > stanza > strophe > verse > proposition. Once a proposition was produced, it entered its corresponding hierarchical structure. Therefore, its connection with preceding and subsequent discourse can be studied. If one continues in this vein, then three parameters mediate the coherence relations with the preceding discourse: the temporal distance between propositions, hierarchic positions of the propositions, and specific thematic relations of propositions. As for the relation of the proposition to its subsequent proposition, that remains open and flexible (Givón 1984, I: 244). The cohesion of poetical units that inspired Fokkelman, but remained equally elusive to him can also be described using the four unities that give concreteness to thematic structure and coherence relations: unity of time, unity of place, unity of action, and unity of participants. ‘The four unities – or continuities – are more likely to be maintained within any particular discourse than across its boundary with another unit’ (Givón 1984, I: 245). Despite the many criticisms Fokkelman’s method had to face, his studies revealed a certain balance that the Hebrew poems display. Since we are dealing with the Masoretic Text, we here assess Korpel’s recommendation, in the name of the Kampen School (1998, 10-11), to consider the colometric value of Masoretic devices as legitimate. Although it represents just one tradition of reading HB, not necessarily stable and definitive, one can still ascribe to the Masoretic tradition a high degree of credibility for the reading of the text. The following section will investigate 1 None of the later published books will bring more light on Fokkelman’s method. In his 2001 introductory guide to Reading Biblical Poetry he will popularize his method previously applied in the aforementioned academic treatises. Of further interest is his 2002 The Psalms in Form, whose only interest is in laying out the Psalms in verse-lines, verses, strophes and stanzas.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

64

the contribution of some modern authorities that dealt with the Hebrew text of the Psalms, taking into consideration the Masoretic tradition that preserved it.

3.3. METHODOLOGICAL S TEPS TO APPROACH THE MASORETIC TEXT OF THE PSALMS As we have already noted and even scholars who practice verse scansion (counting sounds, syllables, or words) admit, the procedure of dividing poems, in our case the Psalms, into poetic units is a heuristic one, meaning that its certainty is built on the basis of the arguments identified. Such an analysis can include several stages, but the superscripts are not included, for having neither poetic structure nor verbal content. Noting the punctuation (Masoretic accents).1 The accent a^tn`j [3] is traced first as the main divisor that splits the verse into more or less regular halves. If the halves are too long, over the average length of a line (i.e. four stress units) or a^tn`j is not present, other major divisors must be identified, Uol? w+yor@d [2] and r+b'^U m%gr*v [5] having priority. Extra-long verses may be subdivided further by D+j' [9], x]Nor [7], and r+b'^U G*dol [4], then P*z@r [10] and a^Zl* l+g^rm@H [12].2 Evaluating the parallel lines. In most cases, they are of close value quantitatively. Therefore, a binary unit 2+5 is improbable, and a 2+4 unit is rather less likely. One stress-unit verse-lines are rare. Unambiguous grammatical signs of coordination or subordination are extremely important to back-up the proposed division. If Revell (1981) is correct, then a doubledistich, as an unbalanced verse, requires one verse-line twice as long as its adjacent verse-line. Noting the cohesion elements and sorting lines into couplets. Grammatical parallelism represents the most basic cohesion factor between poetic lines but it is not the only one. Split couplet, chiastic or envelope structures, or staircase parallelism are not very rare. Although the couplet is by far the most frequent one, care has to be taken with tercets and quatrains whenever they appear (higher structures are very unlikely to appear in Psalms). sm]jWt components belong together in the same line. Since line A is the one carrying the meaning of the poetic structure, having a predicateless line A is not acceptable, especially if it is short (only two stress-units) and does not The number attributed by the Tabula Accentuum in BHS4rev appears in square brackets. 2 For a detailed discussion, see chapter 3. 1

THE STRUCTURE OF POETIC VERSE

65

support the previous poetic unit (cf. 19.4AB). There may be exceptions, though (e.g. 17.4A, 20.8A, 23.4bA). Marking the poetic units. The tradition of marking versets has been established for too long to be neglected. In order to ease reference, verset marking has been preserved, but noting that Arabic numerals reflect the Masoretic tradition of marking versets (where extant, superscripts stand for verset 1). Capital letters (A, B, C, D) mark verse-lines belonging to the same poetic unit. When analysing the verb structure of the poetic units, the limit between lines is marked by //. If internal parallelism is present, the two halves should be delimitated by /. A line coextensive with its poetic unit (monocolon) is marked with M (e.g. 15.5M and 28.5M. Both follow ternary structures). Assessing the clause structure of verse-lines. Since our method applies to verb entities as part of clauses and verse units (a clause is not always coextensive with a verse unit), predicates have to be marked. Although marking clauses may prove both difficult and unnecessary in a tabulary database like the one following, reference to clauses and verbs will be considered during the analysis that follows. For example, the second poetic unit of the third verset of Psalm 17 (17.3bAB) has two predicates for each line (QYYQ verbal sequence). They are to be marked: 17.3bA1, 17.3bA2, 17.3bB1, 17.3bB2. Marking the predicates. On the third column of our database (cf. appendix 9), the type of predicate found in each line is mentioned. A predicate realized by an adjective in predicative position is marked ‘nom’ (nominal). Other verbal forms with predicate function are as follows: infa (infinitive absolute), infc (infinitive construct), ptcp (participle), qtl (perfect), yqtl (imperfect), vol (for any volitive – imperative, cohortative, and jussive), wayqtl (imperfect with waw-consecutive). Any of the previously mentioned forms can be preceded by a conjunction. When no predicate is attested the symbol ø is used instead. Following the application of these methodological steps to the Psalms, the proposed verse-line delimitation enters the laborious process of comparing it against the data available produced by some of the ancient, medieval (cf. 3.1.1. supra) and modern authorities (cf. 3.4. infra). As a result we will be equipped with a powerful tool that will enable us at a glance to visualize the informations relevant for the verse-line delimitation of Psalms as preserved by the Masoretes.

66

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

3.4. THE MASORETIC TEXT AND S OME OF ITS MODERN INTERPRETERS The gallery of modern witnesses, called on to contribute to our overview of the matter of verse division of the Psalter consists of many scholars renowned for their long-standing interest in the HB, who proposed a delimitation of the Psalms in their respective verse-lines, arguably making use of the Masoretic accents (cf. Appendix 3). As the most recent survey of van der Lugt (2006, 3-68) indicates, many other European scholars from the last two centuries manifested interest in rhythmical evaluations of the verselines in Psalms. Our selection includes some of the most representative and accessible ones. Since we do not seek to validate a particular verse-line delimitation, any other selection would have served a similar comparative purpose.1 3.3.1. Julius Wellhausen Although his main contributions are in the area of source criticism as applied to the Pentateuch, Wellhausen produced a critical work on Psalms too (1895). This edition does not depart to any degree from the mediaeval tradition of the incunabula, preserving many of the aesthetic elements that commended them to the aesthetic eye of the public. Nevertheless, what captures attention, in the context of the present work, is the display of the original text in stichographic arrangement. The fact that the text is in two columns per page does not hinder this delimitation, especially since the Hebrew text is not vocalised and in small fonts. Consequently, the objectivity of the method applied by Wellhausen for his delimitation cannot be determined, although he was aware of the balance that the adjacent lines displayed. Red ink is used to mark the main divisions of the Psalms into Books, the subtitles, the selah, the acrostich letters, and the concluding benedictions. Although it is not explicitly stated, one suspects that this material is marked in red to suggest later additions to the text. Critical comments are added as endnotes. 3.3.2. Franz Wutz Unlike many of his German contemporaries, Wutz (1925) works with the text in the original displaying it in poetic lines, after considering carefully the 1 Van der Lugt’s verse-line delimitation could have been useful for a comparative study. Unfortunately, his work was published too late for such a careful study to be conducted.

THE STRUCTURE OF POETIC VERSE

67

Masoretic markers. The only drawback is that Wutz does not link poetic lines into verse-lines. Other German scholars, like Kraus and Loretz, provide a more in-depth engagement with the MT. 3.3.3. Oswald Loretz Loretz assumes, from the outset of his monograph, a defensive position against what he terms ‘die Albright-Dahood Schule’, and promotes the traditional Gunkelian approach of Form criticism. Although he does not deny that the literary tradition of Hebrew song writing started with Ugaritic poems, Loretz sees the Psalms as the result of a long redactional and editorial process that transformed the original texts by means of ‘glosses, additions, and reinterpretations’ (p. 9). Therefore, he acknowledges that the main unit of HPy must be the couplet with lines of 9-12 syllables each, and the tercet is a later development of Semitic POETRY. Consequently, he finds tercets only occasionally. Frequently, he prefers to alter the structure suggested by the Masoretes in order to produce neat couplets instead. Larger verse units are practically absent. One may suggest that this tendency towards reconstructing the Ur-text represents the main assumption that defines Loretz’s technique of organizing lines into verses. Each of the 61 psalms analysed is examined in three sections: Kolometrie, ‘Observations on the Text’, and Textologie. During the initial stage, Loretz displays the original text transliterated in poetic lines, each verse being separated from the following by a line. A literal translation accompanies the original text, the latter being of greater interest to us. Loretz brackets the problematic elements of the text and comments on the contentious issues in the second section of each analysis. Loretz seized the opportunity to discuss his views on the origin of the poem and the impact of the Israelite piety on the formation process during the third section of the analysis. Anthony Ceresko (1981, 280) criticised Loretz’s dealings with the minutiae of the Hebrew text as lacking in the objectivity he claimed for his method, a slacking off academically, which owed much to a very subjective approach to metre. 3.3.4. Hans-Joachim Kraus Kraus produced and published, first in fascicles and in German, a commentary on the Psalms that attracted scholars’ attention from the beginning. Roland E. Murphy monitored and evaluated Kraus’ work positively. When completed, the commentary was published several times in

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

68

German before it reached the English-speaking world by means of translation. Its fifth German edition (1978) was translated by Hilton C. Oswald and published by Augsburg Press in a two-volume edition (vol. 1: Psalms 1-59, 1988; vol. 2: Psalms 60-150, 1989). As Mays (1991) noted, ‘the major strengths of the commentary lie in its form-critical analysis, its interest in the biblical theological dimensions of the text and its engagement with the German tradition of Old Testament scholarship’. Although the author does not pay attention to the rhetorical analysis of the psalms, thus betraying its origin in the German scholarship of the sixties, the work is valuable even for a poetic analysis of the text. Following the standardized format of the form-critical commentaries tradition,1 right after the recommended bibliography, Kraus displays in poetic lines his own translation of the Hebrew text. It should also be noted that Kraus adheres to the classical accentual system of Leys, Sievers, and others, which is seen in his preference for short balanced lines and couplets. His approach to the Masoretic text is also to be commended, since he is not quick to emend the text based on various ancient manuscripts, unless internal information, mostly rhythmical, suggests it. Bright (1961, 194) applauds his caution in handling original texts. Murphy (1961, 76) acknowledges that the great achievement of Kraus is the moderate position he assumes when dealing with form-criticism methodology. Operating between the extremes of Gunkel and Mowinckel, Kraus judges each Psalm individually and refrains from imposing on the text his traditional assumptions. Therefore, contrary to normative formcriticism expectations, Kraus is able to conclude that the main themes of the Psalms are Zion, the Davidic dynasty, tribal traditions and theodicy (Murphy 1961, 77). Zion is Yahweh’s throne city and it is celebrated annually in the feast of the ‘Choice of Jerusalem’ (as opposed to Mowinckel ‘enthronement’ feast), and there the Davidic dynasty is enthroned. Tribal traditions are marginal but still present. As for the theodicy, it is expressed in the national lamentation, so the dilemma of divine justice is present in the individual lamentation.

The so-called BKAT format consists of bibliography, translation, textual notes, setting, commentary, purpose, and thrust (Murphy 1991, 474). 1

THE STRUCTURE OF POETIC VERSE

69

3.3.5. Peter C. Craigie, Marvin E. Tate and Leslie C. Allen The Word Biblical Commentary Series launched a three-volume project on Psalms with the assistance of Craigie (1983), Tate (1990) and Allen (1983). According to the editorial preface, each author of the commentaries included in the series had to provide the volume with a personal translation of the original text. Poems are displayed in poetic lines. The authors apply a similar metrical analysis to the poem, counting the verse-units, not the syllables, nor the consonants (cf. Launderville 1984, 745; Perdue 1984, 378). As the author of the opening volume, Craigie (1983, 38) traces the method in use back to George B. Gray1 and Theodore H. Robinson2. Launderville (1984, 746) assesses Craigie’s response to Dahood’s use of Ugaritic and Northwest Semitic languages as the most distinctive contribution of the commentary. As opposed to Dahood, Craigie pays more attention to the Masoretic text, although occasionally he emends it. By providing a translation that follows the principle of ‘formal correspondence’, Craigie allows the reader to see the original through it. In 1 When it comes to rhythm, Gray (1915, 135) notes that out of the three main theories available on prosodic analysis, there is only one that fits the HPy best. The balance of the poetic lines is due to ‘equality in the number of stressed and accented words or syllables in the two lines’ (Gray 1915, 136). Neither counting syllables nor counting the metrical values of the syllables is taken into account by Gray. He is paying tribute to the system worked out by Sievers and to the theory according to which the Hebrew rhythm rests on an anapaestic basis (Gray 1915, 143-4). The two main principles that underlie Sievers’ theory are: (1) with very few exceptions, each stressed syllable is accompanied by three or fewer unstressed syllables; (2) normally, the stressed syllable comes last, although it can be followed by no more than one unstressed syllable. Some conjectural laws, that establish exceptional situations in which a particular syllable count is emended on phonetic (therefore controversial) grounds, support these principles. As a supplementary principle, Gray (1915, 151-3) proposes that parallel terms should receive the same treatment in respect of stress. The typical rhythm is 3+2 with rarer varieties 4+3 or 4+2. Balanced lines are either 2+2, or 4+4, and 2+2+2. As an alternative, Gray (1915, 184-5) mentions a 3+3 line. 2 According to Robinson (1947, 30), ‘units may be combined to form versemembers and so lines’. The minimum required for a line (Robinson’s ‘versemember’) is two units. A line would rarely contain more than three units. By far, the most common line type is 3+2. Robinson refers also to ‘internal parallelism’ as the balance realized between lines of the same verse, as opposed to ‘external parallelism’, which refers to the balance between lines belonging to two consecutive verses (Robinson 1947, 28).

70

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

terms of his understanding the HPy, Craigie proves to embrace a ‘standard approach to parallelism’ that would have benefited if interaction with Kugel’s reformulation of parallelism had been available (Perdue 1984, 3780). Also to be noted is the author’s awareness of one the most debated problems of HPy, namely, the translating of the y]qf{l and q*f~l verbal forms. For a work with the size and the purposes of such a commentary, an introductory excursus proved to be sufficient (pp. 110-3). After highlighting the problem and surveying the solutions, Craigie doubts whether in the light of current knowledge, one can understand the tense of the verbs in HPy. Therefore, he opted for the conventional approach to aspect, referring to completeness or incompleteness of action, with context support provided. Nevertheless, Craigie’s method is criticised by Emerton (1984) especially for its insufficient interaction with Dahood and the unbalanced use of bibliography. Maybe such expectations could not be met due to the philosophy that principally inspired the production of this series. According to the Editorial Preface, this series was intended to make the technical and scholarly approach to a theological understanding of scripture understandable by—and useful to—the fledgling student, the working minister as well as to colleagues in the guild of professional scholars and teachers.

March (1985a, 314) noted, also, that engagement with critical issues is lacking, sensing a predictable conservative position that the author is reverting to. Despite this criticism, Miller (1985, 83) views Craigie’s commentary positively, labelling it as ‘one of the best treatments of Psalms 1-50 available in English’. After Craigie’s premature death, Tate and Allen continued the commentary on Psalms. Although Tate (1990, x) mentioned that he adopted Craigie’s method of delimitation, he did not produce genuine verse structures, apart from splitting the versets into their respective verse-lines. His dependence on Dahood can be seen at least when it comes to the double standard promoted, for one is the scansion of the Hebrew text and the other is the display of the English rendering (e.g. Ps 59.4; 87.4-5; 94.7; 95.9). Consequently, his scansions for verses with more than three lines are to be taken only as guides to check the division of verses into lines. Creach (1993) and McCann (1993) offer more of a positive review. Kennedy (1984) is very welcoming towards Allen’s commentary, the third part of the Word Series on Psalms. The format and methodology seem to have been preserved, except for conflation of the Notes and Comments.

THE STRUCTURE OF POETIC VERSE

71

Reviewers praised his critical engagement with the bibliography quoted and the intertextuality exercise with reference to New Testament use of the Old. Kennedy is enthusiastic over Allen’s stand for a conservative approach on the one hand, criticising him for holding higher views on the later dating of some Psalms on the other hand. For Sheppard (1985) the ‘“evangelical” emphasis on authorial intent’ is too often misleading. Since the translation produced follows the principle of ‘dynamic equivalence,’ the Hebrew-trained reader would find it difficult to trace the original through it. This preference for a dynamic translation contradicts the intensive usage of the Hebrew text the author employs for his notes and comments, that ‘can be grasped [by the reader] only by having the Hebrew text at hand’ (Launderville 1985, 691). Nonetheless, his awareness of verb usage in HPy and its implications for translation, as well as the careful rhetorical analyses of the poems, are to be commended (Launderville 1985, 692). 3.3.6. Pierre Auffret Auffret’s commentaries (published 1981-2003) are intriguing for several reasons. First, the association of Psalms in each volume seems to be random, except for the volume published in 1999 on Psalms 120-134. Second, the text of the poem is in most cases assumed but not displayed. The volume published in 2003 is a welcome exception, as each chapter starts with a display of the text in translation (French) and in poetic lines. The author only occasionally cited the Hebrew original of the text being commented on, and accompanied it with its transcription, thus allowing the reader to follow his technical comments. Third, the technique used to mark the poetic lines is not clear. Methodologically, Auffret leaves the reader in a deep fog (Lugt 2006, 63). This is particularly the case when the text displayed in poetical lines does not accompany the comments, so that references to segments of the lines are completely misleading, even though they may have relevance for the author himself. For such reasons, except for the 2003 volume, there are not many other of Auffret’s papers that are of real value for an investigation such as ours. However, whenever an insight was gained from his studies, it has been appropriately acknowledged. So far as the method applied is concerned, Anthony Ceresko (1995) is of the opinion that Auffret practises structural analysis on the surface structures, as opposed to the deep structure. Due to Auffret’s high appreciation of Avishur’s Stylistic Studies, the division of the original text into

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

72

poetic lines can be of great value. High appreciation of the minutiae of the text is correctly associated by most reviewers with Canonical Criticism (Christensen 1985, 133; Lundbom 1986, 100). The implications of such rhetorical observations are particularly valuable for such an approach. The bulk of literary connections between Psalms, noted in Auffret’s exquisite Sitz im Text approach is considered, however, to be rather overwhelming by Ceresko (1995, 305). He expresses his doubts by saying: ‘many of the studies are so detailed and complex that one begins to wonder to what degree Auffret is simply observing and noting […] and to what degree he may instead be imposing them on the text.’ Although it would seem to be easy enough to perceive one’s own methodology, Auffret entrusted this to his critics (Culley 1997, Watts 1997). Noting repetitions and patterning is a valuable technique for literary texts in general, and for Semitic POETRY in particular, one legitimately wonders to what extent such literary structures are legitimate or meaningful to Auffret (Lundbom 1986, 100; Culley 1997, 106). A tendency not to distinguish between POETRY and PROSE is criticised by Christensen. Actually, the Kampen School has long promoted such a tendency. Nevertheless such observations are of little value for a work that lacks a bibliography (Lundbom 1986, 101) and, even more, did not interact with the bibliographical material (Watts 1997, 542) except for the author’s own contributions that can be found at the end of his 1995 book (Culley 1997, 106). 3.3.7. Raymond J. Tournay According to Henry Wansbrough who wrote the introduction, The New Jerusalem Bible represents the English version of the second revised French edition, which was published in 1973. The story of bringing the second French edition into existence comes from the most knowledgeable person, Pierre Benoit. He was closely involved in the process of producing this new translation through all the stages of the editorial process.1 Before the first edition was published in one volume, the Steering Committee agreed on publishing the translated biblical books in fascicles. This process started in As a New Testament scholar at Écôle Biblique, Benoit started by translating Matthew and the Prison Letters. Later he became a member of the Steering Committee, the body that coordinated the translators. He was part of the Revision Committee responsible for the preparation of the first translation, then part of the Revision Committee which prepared the second edition for printing, seventeen years after the first edition. 1

THE STRUCTURE OF POETIC VERSE

73

1948 and concluded with the publication of the Bible in full in 1956. A second edition was published in 1973, after intensive revision had taken place. Psalms in particular, after being translated by Tournay, was adapted to liturgical prayer and published in 1961 as The Psalter of the Jerusalem Bible (Benoit 1979, 348). The translation principles that inspired the large group of translators were as follows: (1) translation was based on the original languages (Hebrew, Aramaic, Greek); (2) the text was evaluated critically, taking into account the conclusions of different authorities and manuscripts; (3) a faithful, literary rendering of the text was attempted (Benoit 1979, 341). Exegetical and literary revisions were provided for the first draft (Benoit 1979, 345). Unlike most Bible versions, NJB displays the text in a modern fashion, i.e. one column per page. Therefore, the poetic texts, laid out in lines, are easy to follow for the reader. Subsequent translations of La Bible de Jérusalem into modern languages involved revision activity. Thus, in producing the English version, the editors corrected the faults of the French version and improved it to the best of their abilities (Benoit 1979, 348-9). Although Kilpatrick (1986) asserts that NJB is translated from the original languages, Di Lella doubts this. He concludes that, since the translators preferred the French edition to the MSS, it is rather a secondary translation (Di Lella 1989, 358). Otherwise, its features recommended it as the best annotated Bible in English of its time. Another positive review is offered by Jerome D. Quinn (Worship 1987, pp. 370-2). 3.3.8. Tertulian Langa Contrary to most Eastern translations of the Bible, Langa’s translation follows the Masoretic original and he makes his assumption public by doubting the superiority of the Septuagintal text (Langa 2000, 402). He is raising the two basic problems inherent in the translation process: content and form. In terms of content, Langa advocates a non-literal translation, rendering the meaning of the original into the target language. In terms of form, the debate centres on the layout of the biblical poems (Langa 2000, 402-7). Langa displays his translation of the Psalms from Hebrew into Romanian in one column of poetic lines and strophes. Both at the minutiae level of semantics and the formal level of organizing words into poetic lines, Langa’s translation may be biased, due to its allegiance to the Catholic Church, whose translation admittedly inspired the translator (Langa 2000, 411-2). Nevertheless, the author has a higher

74

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

aim than just offering a modern rendering of the Psalms. Langa is strongly empathetic with the ecumenical movement, which is the raison d’être for his translation using the ecumenical translation of the Bible as his main reference tool (Langa 2000, 412).

3.4. S UMMARY This chapter has looked into the technique of verse-line delimitation with respect to the Psalms. Given that our primary text comes to us as preserved by the Masoretes, the consideration of the Masoretic accents is unavoidable. Even when the role of the Masoretic accents is limited to a consultative one, we have offered an academic justification of such an approach. Next, the contribution of Kampen’s school of delimitation criticism was surveyed as one of the most recent approaches to verse delimitation in Hebrew literature that takes into consideration the Masoretic tradition. As a result, we were persuaded of the importance of the Masoretic accents for verse-line delimitation, thus ascribing them a high degree of credibility, though not definitive. Finally, we investigated the contribution of modern authorities who have dealt with the Hebrew Psalms with consideration for the Masoretic tradition that preserved it. At this stage, we were able to produce and check our database of Psalms with their text divided into lines and verses (cf. the excerpts in Appendix 9). Such a verse-line delimitation exercise was necessary to enable us to select those couplets with a QYYQ verbal sequence and so to produce the main Appendices (1 and 2) which include the relevant data to be subject to a Systemic Functional Grammar analysis (chapter 6). The next two chapters will lay the foundation for this analysis by describing the systemic functional grammar method to be applied to some individual Psalms and to the Psalm couplets with the QYYQ verbal sequence.

4 SYSTEMIC FUNCTIONAL GRAMMAR: A READER Systemic Functional Grammar grew out of the work of J.R. Firth, a British linguist, Professor of General Linguistics at the University of London, but was mainly developed by his student, M.A.K. Halliday.1 Michael Alexander Kirkwood Halliday (b. 1925) continued Firth’s work and, under the influence of European functionalism (Hjelmslev) and American anthropological linguistics (Sapir, Whorf), developed Systemic theory.2 His seminal work Categories of the Theory of Grammar was published in 1961. After a short halt (1965-70) at the same chair as his former mentor, Halliday moved during the seventies to Australia, where he established the department of linguistics at the University of Sydney. He is credited with being the main theoretician of the neo-Firthian group of scholars. Among functional grammars known to date, Systemic Functional Grammar (henceforth SFG) is noted for its sociological orientation and use of a larger register of ungrammatical sentences (Siewierska 1991, 15).3 One of the most important contributions of the Firthian school of linguistics is considered to be ‘polysystemicism’, that is a linguistic analysis based on the view that language patterns cannot be accounted for in terms of a single system of analytic principles and categories, but that different systems may need to be set up at different places within a given level of description (Crystal 1980, 152). 2 Halliday (1985a) gives full credit to his precursors, the influence to build a SFG coming from two main directions: from Firth and his colleagues and from Halliday’s personal contacts with linguists and language teachers. Thus, he includes in the first category Firth’s younger colleagues, namely W.S. Allen, R.H. Robins, Eugenie Henderson and Eileen Whitley. Sapir, Bloomfield, Fries, their successors – Hockett, Harris, Gleason and Pike – and, most importantly, B.L. Whorf, join the group from the other side of the Ocean. Halliday’s personal contacts include Walter Simon, Luo Changpei, Wang Li, David Abercrombie, Angus McIntosh, Jeffrey Ellis, Trevor Hill, Kenneth Albrow, Ian Catford, Sydney Lamb, Basil Bernstein, David McKay, Peter Doughty and many others. 3 Other approaches include Generative Functionalism, as adopted by Prince 75 1

76

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

In relation to grammar, Halliday holds the view that children’s protolanguage (primary semiotics) is, above all, supposed to have meaning (meaning is obvious and natural as given).1 But, mature language (stratified semiotics) allows the user to construct meaning in the environment of an open set of classes (system network) of semantic potential. He is also known for giving more consideration to the view that language form, function, and context are non-arbitrarily related (Thibault 1994, 1286b). The main applications of SFG so far have been to language acquisition (both as first and second language), pathology of speech, computational linguistics (machine translation), sociological interpretation of culture, textual analysis (stylistics and analysis of spoken discourse), and interpretation of non-verbal art forms – music, painting, sculpture, architecture, etc. (Butler 1985, 193-213 ; Halliday 1994, 4507b; Halliday 1996, 20-1). Although there is still a lot of work to be done, it is clearly possible for this technique to be applied to other areas of study also, for example, that of authorial style (see Butt 1990).

4.1. PRINCIPLES OF THE SYSTEMIC THEORY In order to produce a description of language that is concerned with language in its entirety, Halliday, in his Systemic Theory, makes use of five categories describing as many dimensions of language: structure, system, stratification, instantiation and metafunction. To start with, by ‘language’ he means natural, human, adult, verbal language. Thus, mathematics is excluded as non-natural, computer-language as non-human, infant protolanguage as non-adult, and art as non-verbal (including music, dance, etc.) (IFG, 20). and Kuno, Functional Grammar, as inspired by Dik, Role and Reference Grammar, as developed by Van Valin, and Cognitive Grammar as promoted by Langacker. Butler gathers numerous approaches in the USA together into a category labelled West Coast Functionalism (2003, 1: 34, 49-53). 1 Although it implies ‘primary consciousness’, an awareness of the world around, and is common to human and superior animals alike, the semiotic system of children’s crude language is without grammar. It is built of onomatopoeic expressions, interjections (expressions of pain, anger, astonishment, joy, happiness, wonder, etc), and, towards the end of the first year of life, even very simple clauses. On this protolanguage, children build afterwards social values and grammatical rules, leading towards the organization of a system that allows new meanings, alien to other beings and impossible to grasp by infants (Halliday 1975, 1978, 1979 inter alia).

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

77

Structure is the quality of language to have a compositional structure (known in linguistics as ‘constituency’). Viewed as a phonological production, the text consists of phonemes, organized into syllables, organized into feet. This order whereby larger units are made up out of smaller ones is called constituency. Each level of this hierarchy is called a rank and the hierarchy of units is called a rank scale (IFG, 5-6). Thus, the phonological rank scale of English looks as follows: line > foot > syllable > phoneme. Viewed as a written production, modern English displays a graphological rank scale that includes sentence (a fragment of text that begins with a capital letter and ends with a full stop), sub-sentence (unit of sentence bound by colons, semicolons and commas), word (unit of subsentence bounded by spaces) and letters (IFG, 6-7). Most importantly, when a text is viewed as a grammatical production, it is broken down into units of which the clause is the central processing unit. The lexicogrammatical rank scale of English is: clause complex > clause > word > morpheme (IFG, 7-10). Units of every inferior rank can interact logically forming units of superior rank. There is also the possibility that a unit may be downgraded to function in the structure of a unit of its own rank or of an inferior rank. This situation is called rank shift. A unit included within another, splitting the other one into two parts, is called enclosed. System is the quality of language to have a set of alternatives with a set of entry conditions. ‘A system is a set of classes whose members contrast in respect of a single property’ (Halliday 1967, 32). For example, all ‘indicative’ clauses (entry condition) can be either ‘declarative’ or ‘interrogative’ (alternatives).

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

78

Similarly, the ‘polarity’ of a clause can be either ‘positive’ or ‘negative’.

As ‘declarative’ and ‘interrogative’ are contrasting features of the clause in relation to ‘indicative’, so are ‘positive’ and ‘negative’ for ‘polarity’. These options represent an aspect of the meaning potential of the language.1 Systemic theory is a theory of meaning as choice, by which a language, or any other semiotic system, is interpreted as networks of interlocking options: ‘either this, or that, or the other’, ‘either more like the one or more like the other’, and so on. (Halliday 1994b, xiv)

The principle that keeps these options in order is that of delicacy and corresponds to rank in the structure.2 A text is the product of an ongoing, unconscious, selection in a very large network of systems called a system network (whence the name of Systemic theory derives). For example, the system network of MOOD develops as an extensive web of systems (‘status’, ‘mood type’, ‘indicative type’, ‘interrogative type’), and each system contributes to the formation of the structure. In Systemic theory, structure does not stand for a defining characteristic of language, but for forms taken by choices organized in systems. Therefore, Martin (1992, 121a) could say that ‘the grammar of language is treated as meaning-making, rather than meaning-encoding.’ This pre-eminence of meaning over grammar in SFG is clearly spelt out in passages such as the following: A language is interpreted as a system of meanings, accompanied by forms through which the meanings can be realized. The question is rather: “how are these meanings expressed?” This puts the forms of a language in a different perspective: as means to an end, rather than as an end in themselves. (Halliday 1994b, xiv) In reality, the options are not equally refined as the realization of the option of negative polarity indicates (cf. IFG, 23 fig. 1-8). 2 Systemic theory does not see language as a well-defined system, but rather as a semiotic system. Hence, the description of a language is a description of choice. Options coexist in networks and are realized in constituent structures (cf. also Halliday 1985a, 192-3). 1

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

79

Stratification arises from the variety of semiotic levels characteristic of language: phonology, orthography, morphology, syntax, vocabulary. To this traditional understanding of language, Halliday opposes a different model with five strata linked with one another by the process of realization (see fig. 4.2.).

Phonetics stands as the base of stratification with its natural resources for speech and for hearing. Speech sound is then organized into formal structures and systems. These basic layers of language are further developed into two layers that allow the meaning potential of a language to expand, lexicogrammar and semantics, the former dealing with wording, the latter

80

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

with meaning.1 Finally, the context refers to the cultural environment wherein the text is produced and relevant. Therefore, ‘a language is a series of redundancies by which we link our ecosocial environment to non random disturbances in the air (soundwaves)’ (IFG, 26 cf. Halliday 1985a, 196-7). At this dimension of language, the principle that coordinates the relationship among the strata is called realization. From the speaker’s perspective, cultural values are realized in meaning in the layer of semantics, meaning is realized in wording in the layer of lexicogrammar, wording is realized in composing in the layer of phonology and composing is realized in sounding in the layer of phonetics. Instantiation is the quality of language to have representation in any given instance in the form of text. System and instantiation are complementary to each other, because system as potential meaning exists inasmuch as it is incorporated in texts. Given the multitude of contexts that can generate various texts types and the variety of system networks, instantiation can be better explained as a cline with overall potential and particular instance as its two poles. In this case, a text type can be interpreted as a register or, in Halliday’s terms, a functional variety of language standing as a pattern of instantiation of the overall system associated with a given type of context. Therefore, ‘a register can be represented as a particular setting of systemic probabilities’ (IFG, 27). For example, the imperative is more likely to occur in guide books or recipes than in bedtime stories. The text is the basic unit of SFG, valued as constitutive of culture. ‘Text’ refers to ‘any instance of language, in any medium, that makes sense to someone who knows the language’ (IFG, 3). The text-orientedness of SFG is obvious both to critics (Butler 2003, 1: 47, 156) and to its originator: … the approach leans towards the applied rather than the pure, the rhetorical rather than the logical, the actual rather than the ideal, the functional rather than the formal, the text rather than the sentence. The emphasis is on the text analysis as a mode of action, a theory of language as a means of getting things done. (Halliday 1994b, xxvii)

1 Lexicogrammar is an indeterminate, probabilistic and open system. Open – because new words and morphemes appear constantly, probabilistic – because options are not equiprobable, and indeterminate – because two or more possible grammatical interpretations can be given to an item (Halliday 1985a, 193-5).

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

81

Metafunction arises from the capacity of language to convey meaning at three different levels: ideational, interpersonal and textual.1 These are called metafunctions because it is implied that they are intrinsic to language. At these levels, each component of a given discourse has a function that is mutually and probabilistically attributed to it by speakers. Therefore, language is viewed as a resource to produce meaning simultaneously at all three levels in any instance of language use (Thibault 1994, 1286b). First, language construes reality as human experience. This is called the ideational metafunction and it has two components, the experiential (with reference to Processes and participants, cf. 4.4.) and the logical (with reference to abstract logical relations, cf. 4.5.). In this case, the clause represents a Process (e.g., doing, happening, sensing, saying, being, etc.) with various participants taking place in some given circumstances and with connections to adjacent clauses. Second, language also construes reality as an exchange of utterances, enacting the social relationships of the interlocutors. For its interactive and personal qualities, this is called the interpersonal metafunction. In this case, the clause is an utterance, whereby the producer gives information away or aims to find some, gives an order or makes an offer, expresses an attitude towards the interlocutor, etc. Third, language construes reality as a discursive flow. This is the textual metafunction. In this case, the clause is a text characterized by cohesion and continuity. Therefore, three different types of meanings – ideational, interpersonal and textual – are integrated into a single syntagm, as three facets of the same unit, that is the clause. They will be explored in the following sections of this chapter. Before that, though, we need to extend our introductory presentation with a discussion on two more concepts, class and function. Here, a comparison with traditional grammar would help. In general, a class is a set of items that are in some respect alike (IFG, 50). Although a traditional grammar regularly includes in its list of classes mostly classes of words (such as verb, noun, adjective, etc.), the functional grammar of English recognizes classes of morpheme, classes of group and phrase, and classes of clauses too. As for the classes of words mostly used, the following figure display them interconnected. Traditional grammars assign words to classes in a dictionary, but a class label cannot show the part any given item plays in any actual structure. For an early theorization of the three functions of language see Halliday (1970, 1979). 1

82

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

For example, the Subject covers not less than three definitions: (1) a person or thing that is being discussed, described, or dealt with; (2) the element about which the rest of the clause is predicated; (3) the part of a proposition about which a statement is made. They correspond well to the three categories used for Subject since the nineteenth century, ‘psychological Subject’, ‘grammatical Subject’, and ‘logical Subject’, respectively (IFG, 56).

noun nominal

common proper pronoun

adjective numeral determiner word

verb verbal

lexical auxiliary finite

preposition adverb adverbial conjunction

linker binder continuative

Figure 4.3: Word classes recognized in a functional grammar of English (IFG, 52 fig. 2-8)

In the clause, the duke gave my aunt this teapot, ‘the duke’ can easily cover all these three roles. Not so, in the clause, this teapot my aunt was given by the duke. Here, ‘the duke’ plays the logical Subject role only, ‘this teapot’ being the psychological Subject, and ‘my aunt’ the grammatical Subject (IFG, 56). This problem is removed in SFG by introducing the concept of function. Recalling the above example, Systemic Theory preserves the notion of ‘Subject’ but only as function within the limits of the interpersonal metafunction, that coincides with the grammatical Subject. For the other two roles, two different functions are proposed: Theme for the psychological Subject (cf. 4.2.2.) and Actor for the logical Subject (cf. 4.3.1.). Hence, the elements in the clause this teapot [Theme] my aunt [Subject] was given by the duke [Actor] can have different functions, as seen in square brackets.

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

83

Whereas class labels indicate the potential an item has in the grammar of a language, function labels indicate the part that the item is playing in the particular structure under consideration (Halliday 1985b, 27). Therefore, an individual item has no function by itself except as a part of a structure. Consequently, ‘the purpose of functional labelling is to provide a means of interpreting grammatical structure, in such a way as to relate any given instance to the system of the language as a whole’ (Halliday 1985b, 29; author’s emphasis). When the same element can function as more than one of these functions in a clause (cf. ‘the duke’ in the duke gave my aunt this teapot functions as Theme, Subject and Actor), it is said that those functions are conflated, or ‘mapped’ on to one another (IFG, 56). In order to convey the conflation of functions, the sign / is used. There are various forms of conflating functions, different not only in form but in meaning too. They are known as paradigms; four such paradigms in English are given below (cf. IFG, 57 fig. 2-13). I Theme/Subject/Actor

caught

the first ball

I Theme/Subject

was beaten by

the second Actor

the third Theme

I Subject/Actor

stopped

I Subject

was knocked out

by

the fourth Theme/Actor

Each of the three functions in the clause discussed above is part of a different functional configuration, corresponding to as many strands of meaning embodied in the structure of a clause. (1) As part of the textual metafunction, Theme is a function that identifies the point of departure for the message. Here the clause is a quantum of information. (2) As part of the interpersonal metafunction, Subject is a function that identifies the warranty of the exchange. In other words, the Subject ‘is the element the speaker makes responsible for the validity of what he is saying’ (IFG, 59). Here the clause is a verbal transaction between speaker and listener. (3) As part of the experiential metafunction, Actor is a function that identifies the active participant in a given Process. Here the clause is a

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

84

representation of some Process in ongoing human experience (IFG, 59). All these metafunctions will be detailed in the next pages. As noted by Butler (2003, 1: 169), SFG does not assume that every item in a clause has a function within every metafunction. Besides, each component function has to be interpreted in a meaningful way only in relation to the other constituents, within the limits of the metafunction itself. Within each field of metafunctional meaning, there is a network of choices that permits the description of constituent functions. To summarise, beyond the obvious applicability of the SFG theoretical construct to the analysis of any text, i.e. its universal relevance to texts, this section has attempted to introduce the reader to the main principles of SFG and suggest its relevance to textual studies in general (including Biblical studies). As an alternative grammar, it may provide with new explanations to grammatical forms and their meaning in the context. The Systemic Theory avoids claiming universality for too specific categories, promoting the understanding of differences among languages instead (Halliday 1985a, 197). Consequently the description of English is anglocentric, that of Chinese – sinocentric, that of Hebrew – hebreocentric. Thus, Systemic Theory equips the researcher with linguistic principles necessary for the study of any language and encourages the production of an original grammar of the given language. It privileges the language concerned and constructs its grammar from the texts available not from some examples theoretically reconstructed (Halliday 1996, 33) To illustrate the SFG of English, we will throughout this chapter consider samples from James Joyce’s collection of short stories, Dubliners, and only occasionally from his novel Ulysses.1 In the rest of this chapter, we will use the following symbols:

The reason that motivated our decision to make use of this literary work is mainly practical. The text of Dubliners and Ulysses are both available in electronic copyright-free format (Project Gutenberg). The examples used in this chapter were selected independently of Nina Nørgaard’s extensive SFG analysis of James Joyce’s literary work recently published (Systemic Functional Linguistics and Literary Analysis: A Hallidayan Approach to Joyce. A Joycean Approach to Halliday, University of Southern Denmark Studies in Linguistics, 16; Syddansk Universitetsforlag, 2004). Chris Kennedy also employed Halliday’s ideational meaning of clauses to James Joyce’s ‘Two Gallants’, one of the many stories from the collection Dubliners (pp. 83-100 in Language and Literature: An Introductory Reader in Stylistics, ed. by Ronald Carter, London: George Allen & Unwin, 1982). 1

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

85

/ conflated functions ║ the end of a clause ║│ the end of a sentence [] embedded particle ÆØ embedded clause > enclosed clause

4.2. THEMATIC STRUCTURE Below the clause complex, the grammar manages the discourse flow by means of two systems of networks: THEME and INFORMATION. 4.2.1. The System of INFORMATION Information is a system of the information unit, of similar rank with the clause. The information unit is defined as ‘the tension between what is already known or predictable and what is new and unpredictable’ (IFG, 89). Ideally, each information unit consists of a Given element and a New element, but only the latter is obligatory. The information that is presented by the speaker as recoverable is Given; if it is presented as not recoverable, it is New. Something is recoverable when it has been mentioned before, is part of the situation (I and you), is in the air (so to speak), or the speaker wanted to present it as Given for rhetorical purposes. Non-recoverable may be something that has not been mentioned, or something unexpected (whether previously mentioned or not) (IFG, 91). Phonology is closely related to the presentation of information, but the correspondence between phonological and informational units is complex. Following Abercrombie and Pike, Halliday (1967, 18) suggests that any utterance in English is produced, from an intonational perspective, as a result of expressing choices from three phonological systems, independent of each other: tonality, tonicity and tone. Tonality refers to the distribution of an utterance into tone groups. Thus, a tone group is the phonological unit that functions grammatically as realization of a quantum of information in the discourse (IFG, 88). Tonicity describes the placing of the tonic syllable. Tone relates to the choice of primary and secondary tone. These systems are not, however, independent of the system of rhythm (distribution of the utterance into feet). Although there is a tendency for the tone group to correspond with the clause, the division of an utterance into tone groups is not congruent with its division into grammatical units. The situation where a clause is coextensive with a tone group is referred to as neutral in tonality (Halliday

86

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

1967, 19).1 When tonic occurs on the final lexical item within the tone group, Halliday (1967, 22) speaks of neutral tonicity, whether the tone group corresponds to a clause or not. In terms of tone, the discussion of neutrality takes into consideration the clause system of mood. Thus, only the polar interrogative clauses (Yes/No) are not considered as neutral (Halliday 1967, 24-5).2 A ‘natural’realization of the information structure implies that the New element is marked by prominence and the Given element precedes the New. Each information unit is realized as a fluctuant tone − raising, falling or mixing − with one foot carrying the main pitch (tonic prominence). The element that has tonic prominence carries the information focus. Usually, the tonic foot stands as the culmination of what is New, marking the end of the New element (IFG, 89). If an instance is taken out of its context, a situation not common to real life situations, one cannot tell where the Given element starts or where the boundary is between the Given and the New. One can only identify the New element. The verb of the following sample is elliptical, implied from the given context (‘died’), since the question is raised in the very vicinity of a corpse. Did

he... peacefully? New

The unmarked position for the New is at the end of the information unit. Occasionally, the Given element can follow the New. In such a case, the New element functions as marked information focus. It was something New (focus)

to have a friend like that. Given

Two unusual situations must be considered, though. First, since every discourse has to start somewhere, the discourse-initiating units consist of a New element only; see the first clause below, opening a Joycean literary 1 It is possible for a tone group to include more than a clause, that is a sentence, as found in the following two cases: a reporting clause followed by a reported clause, and a conditioned clause followed by a conditioning clause (Halliday 1867, 20). It is also possible that the tone group is less than a clause, mainly when the break into two tone groups comes after the first element of clause structure that contains a lexical item (Halliday 1967, 21). 2 Tonality, tonicity and tone can produce marked realizations (Halliday 1967, 33-47).

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

87

piece (A Little Cloud). Second, since the Given has a phoric nature (it refers to something already present in the context), the Given element can be elided. The Given element is missing in the information structure of clause (3), where the whole content is new to the reader, even though the reader has the knowledge of the first two clauses. Eight years before he had seen his friend off at the North Wall 1║ and wished him godspeed.2│║ New Given New

Gallaher had got on. 3║ New Newness can also be indicated by contrast. London is New in the following information unit, set in contrast with his own sober inartistic life. Every step brought him nearer to London, farther from his own sober inartistic life. ║ Given New Anaphoric and deictic elements are inherently Given because they do not carry information focus (IFG, p. 91). ‘They’ in they seemed to him a band of tramps is an anaphoric element, referring to poor stunted houses that ended the previous clause. Similarly, ‘for the first time in his life’ stand as a deictic element as Given in the information unit for the first time in his life he felt superior to the people he passed. 4.2.2. The System of THEME Theme is a system of the clause and construes the clause as message, made up of Theme + Rheme. The point of departure of a clause as message, i.e. the point at which it ‘takes off’ from the unfolding text, is called Theme. Everything left over from the message, that moves away from the point of departure, is called Rheme. The Theme provides the context wherein the Rheme has to be processed. Whereas Theme expresses relatively little (or no) extra meaning, in addition to what has already been communicated, Rheme carries the highest degree of ‘communicative dynamism’ (Crystal 1980, 388).1 In her meaning-based text analysis, Callow (1998, 230) understands by Theme the prominent core of the developing message. In this case, Theme is a semantic category that refers to the main thrust of a message; it varies according to the text type mainly in relation to the configuration of the text, cognate words, reiteration of utterance (Callow 1998, 232-3). Topic is a semantic category of a higher rank that relates to the conceptual material in a message, of central importance 1

88

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Functionalism, where Systemic Theory is grafted, sees Theme in various ways. For Dik (1978, 130-140), Theme is a left-dislocated constituent, an optional addition to the Predication, thus operating outside the Predication, specifying the frame of reference of its comment or the range of things with respect to which it makes sense to assert that comment. Unlike Theme, Topic operates inside the Predication, marking the constituent about which it is predicated something (Dik 1978, 141). It appears, that Dik’s ‘Theme’ corresponds to Halliday’s ‘Marked Theme’, whereas Dik’s ‘Topic’ is echoed by the regular ‘Theme’ in Halliday. A much closer approach to SFG is the Information Structure Theory. A valuable contribution to the study of texts is the distinction between Topic and Focus. Whereas Topic is the referent about which the proposition is construed as being about, Focus is the unpredictable element of an assertion. Apparently, Topic + Focus is similar to Theme + Rheme. The set of Systemic functions that come nearest to Topic + Focus, though, are New + Given, as seen above (cf. 4.2.1.). Butler (1985, 176-7) argues that Halliday’s major contribution in this area was to refine the Prague School notion of Theme and Rheme into two independent dimensions of patterning: information structuring and thematization.1 In English (as in other languages)2 the organization of the message into Theme + Rheme is realized positionally, Theme is realized by the first position in the clause, and Rheme by what follows. Old Cotter looked at me for a while. Theme Rheme

An elementary thematic pattern such as the one previously displayed has many variants, of which a complex Theme is just one. It can be realized by a multiple nominal group coordinated (a), nominal groups in apposition (b), a phrase complex (c), etc. (a) Her brother and two other boys [Theme] were fighting for their caps [Rheme] throughout a text (Callow 1998, 217 ff.). In Callow, Theme is the substantiation of the message and its three purposes to be achieved by human communication: informative, expressive and conative (cf. Callow and Callow 1992). For an overview of various functionalist and generative approaches to Theme, see Floor (2004, 2047). 1 Contrary to Information Structure Theory, Halliday related information structuring to a phonological unit, the tone group. 2 Languages which have been the subject of systemic studies include Chinese, French, German, Pitjantjatjara, Akan, Tagalog, Japanese (WFG, 1).

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

89

(b) Polly Mooney, the Madam's daughter, [Theme] would also sing. [Rheme] (c) From her hands and wrists too [Theme] (...) a faint perfume arose [Rheme]

The situation is complicated by the existence of another choice, that of mood. Major clauses will be primarily discussed, since minor clauses do not have thematic structure (they have no independent speech function). The mood of a clause can be either indicative or imperative, with the further distinction between declarative and interrogative in the case of an indicative clause. If interrogative, a clause can be either a polar interrogative (‘yes/no’) or a WH- interrogative. indicative: declarative Every boy [Theme] has a little sweetheart.[Rheme] indicative: interrogative: polar Have you [Theme] seen Paris?[Rheme] indicative: interrogative: whWhat will you [Theme] have? [Rheme] imperative Bring [Theme] me a plate of peas [Rheme] ...

The examples above illustrate the various Mood options in their typical or unmarked form, as listed below: mood type declarative declarative: exclamative interrogative: polar interrogative: WHimperative: affirmative imperative: negative

unmarked Theme Subject WH- element Finite^Subject WH- element Predicator Finite^Subject

Table 4.1: The unmarked choice of mood in English

Sometimes, other mood constituents function as Theme, e.g., Adjunct, Circumstance and Complement (for a definition of these categories, see 4.3.1. and 4.3.2.). These constituents will occupy the slot of maximum salience, and constitute marked Theme (cf. 4.3.2.). Although both systems are speaker-related, Halliday rightly notices that Given + New is listener-oriented (the listener already knows about the

90

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Given), Theme + Rheme is reader oriented (the speaker decides what comes first in the clause). This implies that the speaker can exploit the potential of both systems in order to produce rhetorical effects (IFG, 93). 4.2.3. Theme in declarative clauses Typically, there is a mapping of Theme onto Subject (cf. 4.3.1.). In such cases, the Theme is considered unmarked. The meal [Theme] was prolonged beyond an hour [Rheme] Mrs Mercer [Theme] stood up to go [Rheme]

Whenever Theme is something other than a Subject, most frequently a Circumstantial or Complement, SFG calls it a marked Theme: On Saturday morning [Theme] I reminded my uncle [Rheme] At Westland Row Station [Theme] a crowd of people pressed to the carriage doors [Rheme]

Exclamatory clauses, a sub-category of declarative clause, have a WHelement as Theme. What good company [Theme] they were! [Rheme] How pleasant [Theme] it would be to walk out alone [Rheme]

4.2.4. Theme in interrogative clauses Even though questions can be used for various reasons, their basic function is to request information. Some request information about the identity of a particular element in the content (WH- interrogative clauses), others request confirmation or denial of the informational content (Yes/No interrogative clauses). In both cases, the clause constituent specific to the respective clause type will take the first place in the clause. WH- interrogative clauses will put the WH- element first, but in a Yes/No interrogative the Finite verbal operator will hold that position. These represent the unmarked Theme. Where [Theme] will we meet? [Rheme] Has he [Theme] paid you yet? [Rheme]

Any other material that appears in front of the normal unmarked indicator stands as marked Theme. now about The Love Song of Alfred J. Prufrock [Theme], what is that poem about? [Rheme] (IFG, 77)

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

91

4.2.5. Theme in volitive clauses Such clauses express commands, desires or wishes. In imperative clauses, the unmarked Theme slot is occupied by the Predicator (cf. 4.3.2.). When negated, the Finite verbal operator appears in front of the imperative and is part of the unmarked Theme. Let’s [Theme] have a look at her, Corley [Rheme] Bring [Theme] me a plate of peas [Rheme] Don’t be afraid [Theme] of it! [Rheme]

Any other material, including a Subject preceding the Predicator should be considered unfamiliar material, therefore marked Theme. From this crossroads town [Marked Theme] follow the main road south [Rheme] (IFG, 78)

4.2.6. Textual, Interperso nal and Topical Themes Although Rheme is construed as a configuration of three types of components, the Process itself, participants, and circumstantial factors, one and only one of these components may occur in the Theme. In order to distinguish it from other Theme types, this is called the topical Theme. As we discussed above, topical Theme can be either unmarked or marked. Nevertheless, the topical Theme can be preceded by other elements that do not play any part in the experiential meaning of the clause. SFG identifies two classes of typical non-topical Thematic elements: textual and interpersonal. textual:

interpersonal:

continuative conjunction conjunctive Adjunct modal or comment Adjunct vocative finite verbal operator

Table 4.2: Textual and interpersonal Themes (IFG, 79 Table 3(3))

Textual and interpersonal thematic elements do not carry any special function for the f-structure of the clause if preceding the topical Theme, because they occupy their expected place. They are even less important when occurring in a non-normal position (such as a final volitive, for example).

92

well

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

But

cont. structural THEME

then

surely

Jean

wouldn’t

the best idea

conjunctive

modal

vocative

finite

topical

be to join in RHEME

A continuative [textual Theme] is a word that signals a move in the discourse such as a response, a dialogue, or a new move (e.g. yes, no, well, oh, now). A conjunctive [textual Theme] is a word or group that either links (paratactic) or binds (hypotactic) the clause in which it occurs to another one in proximity. A conjunctive Adjunct [textual Theme] is an adverbial group or prepositional phrase, which relate the clause in which it occurs to the preceding text. Vocatives [interpersonal Themes] are nouns being used to address. Their presence is easily recognised by the mere occurrence of a noun that does not accord with the predicate. A modal comment Adjunct [interpersonal Theme] is a word or a group that expresses the speaker/writer’s judgement on an attitude or idea. These in turn give rise to a system giving a choice between two main types. Propositional comment Adjuncts occur only in declarative clauses; speech functional comment Adjuncts occur in both declarative and interrogative clauses (IFG, 129-32). Propositional comment Adjuncts, such as naturally, obviously, surprisingly, evidently, wisely, etc., allow the speaker to comment either on the proposition as a whole or on the part played by the Subject. Speech functional comment Adjuncts refer to the speaker’s viewpoint (e.g., truly, actually, generally, frankly). It is not always easy to identify the metafunction of initial Adjuncts. If they have a modal function (e.g. evidently, frequently, suddenly), their role is considered interpersonal (cf. 4.3.1.). If they have a strictly linking function, their role is considered textual (e.g. anyway, at least, in conclusion). Furthermore, Adjuncts referring to time, with interpersonal function, are not to be confused with experiential Adjuncts, which have topical function. If the Adjunct in question can successfully function as part of a predicated Theme, then it is a topical element. In addition, temporal items with the role of a textual linker function as textual Themes (e.g. first ... then ... and then ...; previously … afterwards … finally … later), whereas temporal expressions specifying a time setting for the clause function topically (e.g. in 2004, at 5 o’clock, by mid November). Yes/No answers can also be interpreted in two different ways, i.e. textually and interpersonally. If they answer a polar interrogative clause,

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

93

they function as interpersonal Theme, as in ‘ “That’s agreed now, isn’t it?” “Yes, that’s agreed!” ’ If they mark a response move in dialogue or a fresh initiation in monologue, without changing the current polarity, they function as continuative Theme, as in ‘ “Ah, yes,” he said, continuing, “it's hard to know what way to bring up children.” ’ WH- interrogative items have a double function too, as interpersonal and topical Themes simultaneously. This situation is due to their double role, as WH- function and participant or circumstance, such as in to what extent is the Snow Leopard a shaped creation? Similarly, relative items can function both as topical and textual Themes (WFG, 29-30), as in [one evening I went into the back drawing-room] in which the priest had died. Finite verbal operators (‘to be’, ‘to have’, ‘to do’) functioning as interpersonal Theme are finite auxiliary verbs construing primary tense or modality (cf. 4.3.1.). In determining the border between Theme and Rheme, it is helpful to note that ‘Theme extends from the beginning of the clause to include any textual and/or interpersonal elements that may be present and also the first experiential element, which is a circumstance, process or participant.’ (WFG, 26) One should, however, be aware that there are variations from the regular type of clause, as in minor clauses and elliptical clauses, as well as in equative constructions and Predicated Theme, to be discussed in the next section. 4.2.7. Equative Constructions and Predicated The m es When the relation between the Theme and the Rheme takes the form of an equation, systemicists speak of a thematic equative, similar to an identification clause, which has a thematic nominalization in it. Traditional grammar calls this structural feature a pseudo-cleft sentence. the one who gave my aunt that teapot the one the duke gave that teapot to what the duke did with that teapot Theme

was the duke was my aunt was give it to my aunt Rheme (IFG, 69)

The thematic equative specifies what the Theme is and identifies it with the Rheme, hence the Theme is an identifying Theme. Whenever the typical pattern is altered, preference being given to a less expected unit, one speaks of a marked alternative.

94

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

this teapot was what the duke gave to my aunt (IFG, 70) that was what brought them every two minutes to the banisters Marked Theme Rheme

Predicated Theme is not to be confused with those clauses in which one of the constituents is delayed to the end of the clause (cf. 4.3.7.). In order to have a Predicated Theme, the matrix clause has to include the verb to be plus a nominal, and the postposed fact clause has to be introduced by that. The identifying Theme is similar to the Predicated Theme in that that they share the equative feature. Nevertheless, ‘they differ in the choice of Theme, and in the mapping of Theme + Rheme onto Given + New’ (IFG, 95). Thus, the conflation of Theme with New is a regular feature of the Predicated Theme. It is frequently associated with an explicit formulation of contrast such as it was . . . who . . . . it clause f-structure Sentence f-structure information structure

was always I

Theme Rheme Predicated Theme New

who Theme Rheme Given

emptied the packet into his black snuff-box Rheme

4.2.8. Theme in bound, minor and elliptical clauses Dependent bound clauses are clauses that construe meaning in a clause subordinate to an adjacent, dominant clause. If finite, they typically have a conjunction as structural Theme. Where present dominant clauses are written in italics and in square brackets. [I knew] that I [Theme] was under observation [Rheme]

A WH- element at the beginning of a dependent bound clause functions as topical Theme. [Mahony asked] why [Theme] couldn't boys read them [Rheme]

If non-finite, the conjunction or the preposition functioning as structural Theme can be followed by a Subject as topical Theme (underlined). while [Theme] attempting to cross the line [Rheme] while my aunt [Theme] was ladling out my stirabout [Rheme] [he said]

Sometimes in non-finite bound clauses both structural and topical Themes are missing, the clause being left with Rheme only.

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

95

[The little boy looked about him wildly but,] seeing no way of escape, [Rheme] [fell upon his knees]

Minor clauses are those clauses that function as calls, greetings, exclamations, or alarms, that do not have either mood or transitivity structure. Since they have no independent speech function, minor clauses do not have a thematic structure. “Polly! Polly!” “Sh, sh!” ... “Pony up, boys.”

Elliptical clauses are those clauses in which one or more of the expected units are missing, frequently met in greetings, salutations, and the like. The unit left in the clause may be a nominal group, a verbal group, an adverbial group or a prepositional phrase. “What age is he,” said Mr. O'Connor. “Nineteen,” [Rheme] said the old man. (elliptical for “He is [Theme] nineteen [Rheme]”) “Which is my bottle?” he asked. “This, lad,” [Rheme] said Mr. Henchy. (elliptical for “This is [Theme] your bottle [Rheme]”) “Another time,” [Rheme] said the young man. (elliptical for “We will [Theme] have a drink together another time [Rheme]”)

4.3. MOOD STRUCTURE ‘Simultaneously with its organization as a message, the clause is also organized as an interactive event involving speaker, or writer, and audience.’ (IFG, 106) Social interaction assumes an interactive Process whereby interactants, whether verbally or otherwise, interchange commodities. Each contribution (move) is part of an exchange (dialogue) and is defined in relation to three systems of variables: move, initiating role, and commodity. initiation give demand

goods-andservices

offer shall I give you this teapot? command give me that teapot!

response: expected acceptance yes, please, do!

response: discretionary rejection no, thanks

undertaking here you are

refusal I won’t

96

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

initiation give demand

information

statement he’s giving her the teapot question what is he giving her?

response: expected acknowledgement is he?

response: discretionary contradiction no, he isn’t

answer a teapot

disclaimer I don’t know

Table 4.3: Speech functions and responses (IFG, 108, table 4(1))

‘Move’ implies either initiation or response, ‘initiating role’ relates to giving or demanding; ‘commodity’ relates to either goods-and-services or information. Although language (spoken or written) is obligatory for the exchange of information, its presence is optional when interchanging services/goods. Those four semantic categories that realize the Mood at the level of lexicogrammar produce different types of initiating and responding clauses. In conversational English, responding clauses are most frequently elliptical. Since the Mood element carries forward the dialogue (cf. 4.4.1.), its presence is normally unavoidable. For its ability to describe discourse as an interchange between interlocutors, Mood is best described as the ‘grammaticalization of the semantic system of speech function associated with the dialogic move’ (WFG, 95). SFG labels the clauses that exchange information as propositions, distinguishing them from proposals, which is used to label exchanges of goods/information (IFG, 110-111). 4.3.1. The Mood Element The grammatical system of Mood represents a network of choices wherein the basic distinction is between imperative and indicative mood types. The presence of the Mood element (Finite + Subject) realizes the feature ‘indicative’ (IFG, 114). Indicative clauses can generate further options, those of declarative and interrogative types. As noted previously (cf. 4.2.2.), interrogative clauses may be of two types: Yes/No type or WH- type. Declarative clauses themselves can be either exclamative or nonexclamative (cf. Figure 4.2). SFG establishes two main units for the f-structure of the clause as exchange: Mood and Residue. Of the two, the most important is, obviously, the Mood element; in English this can be supplemented by a Mood tag (e.g., you didn’t expect that, did you). The Mood function realizes various mood

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

97

categories (cf. 4.3.1.). Residue will be discussed later (cf. 4.3.2.). The Mood element is that part of the f-structure of the clause that makes the clause ‘negotiable’. In English it consists of Finite, Subject and, occasionally, modal Adjuncts. The Mood element has the semantic function of carrying the burden of the clause as an interactive event. Realized by verbal operators, the Finite makes the clause ‘negotiable’ by encoding polarity (positive/negative), by reference to the time of speaking (primary tense: it is/was/will; it isn’t/wasn’t/won’t) and by reference to the judgement of the speaker (modality: it may/will/must; it may not/won’t/mustn’t). Mr. Alleyne I he he Mood Subject

has been won’t wasn’t would

calling for you keep you much of a theologian ask of his hearers Residue

Finite

hasn’t he? will I? was he? wouldn’t he? reconstructed Mood tag Finite Subject

Since neither the simple past active positive nor the present tense need auxiliary verbs for the realization of tense, the Finite element is represented by the tense of the lexical verb. In such cases, the Finite element and the lexical verb are conflated. Therefore, in a paradigm, the Finite and the lexical verb can still be rendered as reconstructed, the lexical verb being made evident by means of single inverted commas. Little Chandler gave them no thought, Subject Finite ‘(past)’ ‘give’ Mood Residue

did Finite

he? Subject

reconstructed Mood tag

The Subject, typically realized by a nominal group, is the element picked up by the pronoun in the question tag of a declarative clause. But Hogan we Subject Mood

has had Finite

a good sit, quite enough of the last year, Residue

hasn’t he? hadn’t we? Finite Subject Mood tag

Frequently in familiar English, an interjection can replace the question tag, as ‘eh’ instead of ‘won’t you’ in you’ll be able to pull it off all right, eh? Since it supplies the information by reference to which the proposition can be affirmed or denied, it follows that Subject implies more than just the grammatical Subject. It indicates the entity ‘in respect of which the assertion

98

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

is claimed to have validity’ (IFG, 117). At the same time, Subject is that which combines with the Finite to form the Mood element in a clause. Thus, Subject is a comprehensive category, not a mere syntactical one. Significant information about the indicative is signalled by the order of Subject in relation to the Finite in English. Any change of Subject or Finite alters the meaning of the propositional structure too. Whereas the order Subject^Finite realizes ‘declarative’, Finite^Subject realizes ‘Yes/No interrogative’. In a WH- interrogative, the order Subject^Finite is preferred if the WH- element is the Subject (IFG, 115; WFG, 63), otherwise Finite^Subject occurs instead (cf. 4.3.4.). Old Cotter looked at me for a while. [declarative clause] Subject Finite Is he dead? [Yes/No interrogative clause] Finite Subject How do you mean, Mr. Cotter? [WH- interrogative clause] Subj Finite

Parts of the Mood element of the clause are also the mood Adjunct and the comment Adjunct. The mood Adjunct is typically realized by an adverbial group expressing (a) temporality, (b) modality, or (c) intensity, of which the first two have a strong tendency to function as Theme. (a) Adjuncts of temporality relate to time deixis, to the time itself or to an expectation with regard to the time at issue (e.g., eventually, once, soon, just, still, already, yet, no longer). and Adj: conj

still they couldn't see a sight of him anywhere Adj: Subj Finite temp Residue Mood

(b) Adjuncts of modality cover the space between ‘yes’ and ‘no’ with discretionary options providing intermediate degrees of probability/usuality (for propositions) or obligation/inclination (for proposals). A proposition can affirm that something ‘is so’ or deny it. Thus, propositions lacking certainty are represented by intermediate possibilities such as degrees of probability (possibly / probably / certainly) or degrees of usuality (sometimes / usually / always). Death, in his opinion, had been probably due to shock and sudden failure of the heart's action. Subj Adj: conj Finite Adj: mod Residue Mood

SYSTEMIC FUNCTIONAL GRAMMAR: A READER usually he Adj: Subj mod Mood

came Finite ‘(past)’

99

home in the small hours

‘come’ Residue

Both probability and usuality can be expressed even without the use of Mood Adjuncts, by a finite modal operator (that will be John). Proposals either prescribe (‘do it’) or proscribe (‘don’t do it’). There are various intermediate possibilities: degrees of obligation referring to a command (allowed/supposed/required), or degrees of inclination in relation to an offer (willing/anxious/determined). A finite modal operator can realize both obligation (you should know that) and inclination (I’ll help them). A passive verb can contribute to the same goal, expanding the Predicator and realizing obligation (you’re supposed to know that) or inclination (I am expected to help). An adjective can serve the same purpose and realize obligation (I’m responsible to know that) or inclination (I’m anxious to help them). The most common proposal is realized by a second person volitive, requesting the listener to do something (go home!). Nevertheless the speaker can also offer to do something (shall I go home?), or suggest to the hearer that they both do something (let’s go home), in which cases first person volitive is preferred. Polarity (positive/negative) is yet another type of modality Adjunct. It may be realized either as part of the Finite (hasn’t) or separately, as a salient variant (not), with the adverb ‘not’ placed in front of the verb. But I could not Adj: Subj Finite cont Mood

remember the end of the dream Residue

When it appears in English, the mood tag reverses the polarity of the main clause. But you will Adj: Subj Finite cont Mood

come, won't you? Residue

When ‘yes’ and ‘no’ function as statements, as responses to an initiation done by somebody else, they are mood Adjuncts. “You don't feel ill or weak?” “No, tired: that's all.” “[...] Do you hear me now?” “Yes, sir.”

100

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

However, both ‘yes’ and ‘no’ can function as part of the Textual Theme; in this case they are continuatives, outside the Mood f-structure of a clause (cf. 4.2.6.). “I was just telling my mother,” he said, “I never heard you sing so well, never. No, I never heard your voice so good as it is tonight.” “It keeps coming like from down in my throat; sickening.” “Yes, yes,” said Mr. M'Coy, “that's the thorax.”

(c) Adjuncts of intensity include two classes: degree and counterexpectancy. Degree exhibits three types ‘total’, ‘high’, ‘low’ (e.g., totally, entirely, almost, nearly, scarcely, hardly), and counter-expectancy two types ‘exceeding’, ‘limiting’ (e.g., actually, really, simply, only). The reunion Subject

had been Finite

almost Adj: intensity

broken up on account of Jack's violence. Residue

Mood

The comment Adjunct is also part of the Mood element. Although comment Adjuncts may be confused with mood Adjuncts, the former are restricted to ‘indicative’ clauses and ‘express the speaker’s attitude either to the proposition as a whole or to the particular speech function’ (IFG, 129 cf. Fig. 4-13 and Table 4(10)). Examples include obviously, clearly, surprisingly, unfortunately, honestly, roughly, truly, strictly, etc. Conjunctive Adjuncts have a textual role by which they indicate connections with other parts of the text, typically the preceding text. Hence, they are not part of the Mood structure, although their role resembles that of modal Adjuncts. Whereas the latter are interpersonal in function, the former are textual. They occur only as an afterthought and construct a context for the clause (IFG, 132-3). Their role is rather that of cohesion (e.g., however, on the one hand, then, etc.). Such men, Subject Mood

however, Adj:conj

seldom Adj: mood

make Finite ‘(present)’

good husbands. Residue

4.3.2. The Residue Eleme nt The second functional element of a Mood structure is the Residue. It consists of three functional elements: Predicator (only one), Complement (one or two), and circumstantial Adjunct (indefinite number).

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

101

The verbal group, traditionally accounted as Predicate, is divided by SFG into Finite and Predicator. Therefore, the Predicator is the non-finite item of the verbal group, i.e. the verbal group minus temporal or modal operators. If not displaced through ellipsis, the Predicator is present in all major clauses. Its function is fourfold: (1) to specify time reference (also called secondary tense), which is other than the reference to the time of the speech event; (2) to specify various other aspects and phases (e.g. seeming, trying, hoping); (3) to specify the voice – active or passive; (4) to specify the Process type – action, event, mental Process, relation (IFG, 122).1 In the following examples, the Predicator is underlined. a little crowd of people would assemble after mass at the corner of Cathedral Street. In the dressing-room behind the stage she was introduced to the secretary of the Society, Mr. Fitzpatrick. Mr. Alleyne has been calling for you All the people in the affected areas are panicking. I have been at him all the evening So we’re going to wash the pot together ...

The auxiliary verbs ‘have’ and ‘be’ were left outside the Predicator unit in the examples above as carrying the Finite element of the verbal group. They can be considered part of the Residue only when they function as fully-fledged verbs in the present tense and/or past tense (see examples below). Thus, the Predicator is non-finite. I have my own theory about it He hasn't got those little pigs' eyes for nothing. If it was not Skerries it was Howth or Greystones.

The Complement is an element within the Residue, realized by a nominal group that has the potential of being Subject, but without functioning as such. Her eyes Subject Mood

Finite/ ‘(past)’

gave him no sign of love or farewell or recognition Predicator Compl. Complement ‘give’ Residue

The Adjunct is an element within the Residue that lacks the potential to beome Subject, but rather represents circumstances in which interactive events take place. It is typically realized by an adverbial group or a All these come regularly under the rubric of ‘verb group’, but it is not considered here since it is irrelevant to our current discussion. 1

102

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

prepositional phrase, which may contain a Complement itself. This type is considered a circumstantial Adjunct (abbreviated to Adj: circ) to distinguish it from textual and interpersonal Adjuncts (cf. 4.3.1.). [he] Subject Mood

read the paragraph again Finite/ Predicator Compl Adj: ‘(past)’ ‘read’ circ. Residue

by the failing light of the window Adjunct: circumstantial

In this example, the first circumstantial Adjunct is realized by an adverb and the second by a prepositional phrase. ‘The failing light of the window’ functions as a complement in relation to the preposition, but since it cannot become Subject without losing its preposition, it is not to be analysed as a Complement within the Residue. In English, the typical order of elements in the Residue is Predicator^ Complement(s)^Adjunct(s). He had set up house Subj Finite Predicator Complement Mood Residue

for her six times Adjunct: circ

Whenever one of these constituents is located apart from the rest of the Residue, as when a Residue element precedes the Mood elements, the Residue is divided around the Subject and is said to be a discontinuous Residue. After three weeks she had found Adjunct: circ Subject Finite Predicator Mood Residue

a wife's life irksome Compl Adj: circ

In this case, the Adjunct receives more attention, occupying the most salient position as Marked Theme. This is not to be confused with a conjunctive Adjunct functioning as textual Theme, as in then he drank a glass of water. Outside the borders of Mood f-structure lie the Vocatives and Expletives. Since both Vocatives and Expletives are features of casual conversation, they occur frequently in dialogues (IFG, 133-4). Vocatives serve more frequently to identify the Addressee in ‘demanding’ clauses (interrogatives and imperatives).

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

I Subj

Fin/ ‘(present)’

Mood

leave it Pred Compl ‘leave’ Residue

all Adj

103

in your hands, Mr. Cunningham. Adjunct Vocative

By means of Expletives, the speaker enacts a particular attitude or state of mind. God forbid, said Mr. Cunningham. Righto! said Mr. Cunningham promptly. By God, I'll teach you to do that again!

4.3.3. Mood in Declarative Clauses An English declarative clause displays a Subject^Finite order of elements. She Subject Mood

seemed to be Finite/ Predicator ‘(past)’ ‘seem to be’ Residue

near him Adjunct: circ

in the darkness. Adjunct: circ

Sometimes, the order of constituents is altered, splitting the Mood (abbreviated to Mo- -od). Mr. H. B. Patterson Finlay, Subject

on behalf of the railway company, Adjunct: circ

Mo-

expressed Finite/ ‘(past)’ -od

his deep regret at the accident.

Predicator ‘express’

Compl

Adjunct: circ

Residue

Although a sub-type of the declarative clause, the exclamative clause makes use of the typical interrogative word order, using the interrogative particle with an exclamatory role. What Wh-/ Compl

innumerable follies Subject Mood

Residue

laid waste Fin/ ‘(past)’

Predicator ‘lay waste’

my waking and sleeping thoughts Complement

after that evening! Adjunct

104

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

4.3.4. Mood in Interrogative Cla uses Interrogative clauses can be of two types: polar interrogative clauses and WH- interrogative clauses. The WH- element of an interrogative clause takes the role of the constituent it replaces in a declarative equivalent clause. For this reason, the elements of the Mood f-structure tend to be discontinuous. How much Compl ResAnd where Adj Adj Res-

is Finite/ ‘(present)’ Mo-

Pred ‘be’ ...idue

did you Finite Subj Mood

a plate of peas? Subject ...od pick her[up], Corley? Pred Compl Voc ...idue

Can't you tell us? Finite Subject Predicator Complement Mood Residue

The WH- element (interrogative pronoun) is the distinguishing factor in the realization of a WH- interrogation. It can be conflated with any one of the three functions of Mood structure: Subject, Complement or Adjunct. Its position in the Mood structure, therefore corresponds to the position of the respective function with which it is conflated. Whenever conflated with the Subject it is part of the Mood constituent, but when conflated with a Complement or Adjunct it is part of the Residue. Where Wh-/ Adj

have all the flowers gone? Finite Subject Predicator Mood

Residue

‘Where’ in this example is conflated with a circumstantial Adjunct because it requests the information necessary to explain the location circumstance demanded by the Predicator (all the flowers have gone to ... ). Occasionally, the WH- element requests information about what happened or what someone had done. The most common WH- element that determines such a conflation is ‘what’ (IFG, 136, fig. 4-18).

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

what have the elephants done WH- / Finite Subject Predicator Compl Mood Residue

105

to the pier? Adjunct

In both Mood and Residue, constituent order follows that of the respective declarative clause that is used for the realization of interrogation. 4.3.5. Mood in Volitive Clauses The imperative is the mood for exchanging goods-and-services. Therefore, its Subject is in the second person (‘you’). Its first person realization expresses command and offer (me’, or ‘you and me’) (IFG, 138-9). Usually, a typical positive volitive clause will not have a Mood constituent. Look Pred Res-

lively, Adjunct

Miss Hill, please! Vocative Adj -idue

A volitive clause can occasionally be accompanied by a Mood tag (IFG, 140 fig. 4-20a). Open that, Jack, will you? Pred Compl Vocative Fin Subj Residue Mood tag

The Mood element appears only in the case of first person and negative volitive clauses. Let us toast them all three together. Subj Pred Compl Adjunct Adjunct Mood Residue Don't let me disturb Finite Compl Pred Mood Residue

you Compl

now! Adjunct

In order to mark the person or the polarity of a volitive clause, English uses intonation when marking for person, and change of word order, when marking for polarity. The capitalized syllables in the following examples represent those that are rhythmically prominent.

106

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Marked for person

Marked for polarity

positive YOU look LET’s look LET me help DO look DO let’s look DO let me help

negative DON’T YOU look DON’T LET’s look DON’T LET ME object DO NOT look LET’S NOT look LET ME NOT object

Table 4.4: Marking person and polarity by intonation in imperative clauses (IFG, 138-9)

4.3.6. Mood in Minor and Elliptical Cla uses Elliptical and minor clauses have in common the absence of certain components, either one of the constituents (Subject, Finite, Predicator), or one of the main Mood elements, most likely the Residue. In such cases, the intonation may make the missing information more explicit. Elliptical clauses are those clauses in which only one element is omitted, its function in the clause being presupposed from the context, or all elements are omitted but one. When the whole Residue is missing in a clause, it is assumed from the previous context (#a) and/or replaced by the auxiliary ‘to do’ (#b). Notice that the reply to the first question consists of two clauses (a) Must I have a candle? Finite Subj Pred Compl Mood Residue O yes, said Adj: Adj: Finite Predicator comm mod ‘(past)’ ‘say’ Residue Mood Mood (b) Did you call on Grimes? Finite Subj Pred Compl Mood Residue I did. Subj Fin Mood

Mr. Cunningham. Subject

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

107

Since Mood carries forward the dialogue, it is normally present. However, in a reply to a WH- question, all Mood constituents may be omitted except the equivalent of the WH- element (IFG, 151 fig. 4-26b). What age are you? Wh-/ Fin/ Subj Compl Residue Mood Seventeen Complement Residue

Due to the brevity of direct speech, the Subject is frequently elided, especially in informal conversations. The elided Subject is readily supplied by the receiver in his response as in the following example (IFG, 152 (c)): – Seen Fred? (‘have Pred Compl Finite Residue Mood – No, I haven’t. Adj Subj Finite Mood

you ...’) Subj

Minor clauses are those verbal constructions that do not display a Mood+Residue structure. In English, minor clauses can be of different types, such as exclamations, calls, greetings, alarms, and continuatives (IFG, 153-4). Exclamations are verbal gestures (expressions of emotion or attitude) without a specific addressee. Some are not language, but rather protolanguage, sometimes even onomatopoeic (Wow! Yuck! Aha! Ouch!). Others are reduced forms of language with recognizable words (Terrific! Righto!), in which case they can be analysed as nominal groups. Calls have the structural function of a Vocative (Charlie! You there! Madam President!), calling for the attention of another being, or even of an inanimate object. Greetings include salutations, both initiating and responding (Good morning! Welcome!), valedictions (Goodbye! See you!), and wellwishings (Cheers! Congratulations!). Alarms are verbal directions to another party, intended to warn (Look out! Keep off!), or to appeal (Help! Fire! Mercy!) (cf. IFG, 153). A continuative is a textual element which indicates how the current clause relates to the previous contribution. Contributions such as ‘Yes/No’, ‘mmh!’, ‘sure!’, ‘Aah!’ do not constitute an authentic turn in a dialogic exchange, but rather ‘serve to ensure the continuity of the interaction by

108

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

supporting the current speaker’s turn’ (IFG, 154). In conversation, paralinguistic indicators such as nodding, gestures, postures, facial expressions, etc. can replace such continuatives. 4.3.7. Extensions of Mood Analysis The Mood structure is further complicated by the existence of embedded material, that is, material that falls within one of the elements of the clause (mostly Subject and Complement). Embedded material (marked with [ ... ] ) can function as Postmodifier in a noun group complex, such as the prepositional phrase of the bridge in the first example below, or of the tall trees in the second example. Moreover, which lined the mall is an embedded clause (marked Æ...Ø ) that also belongs to Subject. Subjects are underlined. The granite stone [of the bridge] was beginning to be warm All the branches [of the tall trees] Æwhich lined the mallØ were gay with little light green leaves

Likewise embedded material can function as Head in a nominal group, as ‘the reading’ does in the first example below, or ‘to please’ in the second. [The reading] of the card persuaded me [To please] me, I mean.

The Complement also can be realized as embedded material as in ‘of the grey web of twilight’ in the first clause below. ‘Through which he passed’ is also part of the Complement as embedded clause. Complements are underlined. He watched earnestly the passing [of the grey web [of twilight]] across its face. Though his eyes took note of many elements [of the crowd Æthrough which he passed Ø]

At times, the Subject can be split, having an embedded clause towards the end of the clause and an anticipatory it occurring in the normal Subject position. it SubMood

was reported that he had a vocation for the priesthood Finite Predicator -ject Residue

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

109

The Complement can be realized by a similar split construction. I Subject Mood

found Finite ‘(past)’

Predicator ‘find’ Residue

it

strange

Com-

Adjunct

that neither I nor the day seemed in a mourning mood -plement

Clausal Subjects of the type above are to be distinguished from those occurring in Theme predication (cf. 4.2.7.).1 The embedded clause as postposed Subject is related to the it by apposition, which SFG calls ‘paratactic elaboration’ (cf. 4.5.2.1.), and the Predicator is not restricted to the verb be. It could well be realized by a mental Process that has the ability to project (cf. 4.4.2.).

4.4. TRANSITIVITY STRUCTURE In the preceding sections, we have introduced the first two metafunctional lines of meaning SFG attributes to a given clause. The textual line interprets the clause as message and renders its meaning as Theme + Rheme. The interpersonal line interprets the clause as exchange and renders its meaning as Mood + Residue. But clauses can also be investigated as expressions of human experience. The Systemic Theory calls this function of language the ideational metafunction, distinguishing it into two components, the experiential and the logical. The experiential component of the ideational metafunction is discussed in this section, while the logical will be approached in the next. From the experiential perspective, the text is a flow of events as consisting of chunks or quanta of change, each quantum being labelled as ‘figure’, corresponding to a clause. Figures are made up of two main units, Process and participant, and a facultative unit, that of circumstance. These three elements each represent a network of choices for the realization of TRANSITIVITY, the grammatical system responsible for their association into coherent blocks. SFG identifies several Process types, three major ones – material, mental, and relational – and three secondary ones – verbal, behavioural, and existential.2 Each of these Processes refers to a particular type of 1 In Theme predication, the final clause is a relative clause functioning as Postmodifier to the it, and the Predicator is restricted to the verb be. 2 More recently, WFG (1997, 114, fig. 4.4) proposed the adoption of verbal

110

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

experience. Major Process types refer respectively to doing-and-happenings, emotions, being and having; secondary Process types refer to sayings, psychological states and exists/happens. Since they will be detailed later in this chapter, few examples will suffice here. material verbal mental behavioural relational existential

produce, develop, grow, burn, fry, distort speak, talk, praise, blame, report, ask, inquire love, hate, want, understand, scare, encourage watch, worry, chatter, sneeze, yawn, dance be, become, remain, comprise, elect, play arise, occur, follow, happen, prevail, erupt

Schematically, the relationship between major and minor Process types can be illustrated by using a sliced circle wherein verbal, behavioural and existential Processes occupy the border areas between the main Process types, that is, material, mental and relational. Thus, verbal Processes are placed between relational and mental, behavioural Processes are between material and mental, and existential Processes between relational and material. Although it may appear that SFG allocates a specific category to each Process without difficulty, this is not in fact the case. SFG rather recognizes that our experience possesses a high degree of indeterminacy and proposes the principle of systemic indeterminacy, which allows alternative models for construing experience. For example, emotion can be construed both as a Process in a ‘mental’ clause (this pleased God) and as a participant in a ‘relational’ clause (this made God sad) (IFG, 173).

Processes among the major ones. Consequently, a new diagram is proposed, like a square shape, in whose corners are placed material, verbal, mental and relational Processes, in a clockwise order. Behavioural Processes are illustrated as the diagonal connecting material with mental Processes. Similarly, existential Processes are illustrated as the diagonal joining relational and verbal Processes.

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

111

Figure 4.4: Process types in English (IFG, 172 fig. 5-2)

Secondary to the Process is the participant. Since Process and participants are the main elements of a clause in a transitivity structure, they constitute the experiential centre of the clause (IFG, 176). With the exception of certain meteorological Processes (e.g., it’s raining), every experiential Process type is enacted by at least one participant. Since they will all be detailed later, for now, more examples will suffice. the lion [Actor] caught [Process: material] the man [Goal] they [Senser] saw [Process: mental] the lion [Phenomenon] the lion [Carrier] was [Process: relational] strong [Attribute] they [Sayer] talked [Process: verbal] to him [Receiver] the man [Behaver] screamed [Process: behavioural] the volcano [Existent] erupted [Process: existential]

Both Process and participants contribute towards a change in the flow of events, but they differ significantly in that participants are relatively stable

112

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

through time, whereas events are ephemeral. Because participants are persistent through time, they can participate in many Processes in a given text.1 As opposed to participants, which are inherent in the Process, circumstantial elements are peripheral to the experiential centre of a clause, therefore their presence is optional. English provides a variety of circumstances (WFG, 104, table 4.4). They are usually realized by prepositional phrases and can be organized into nine types: extent (responding to the questions for how long? how far?), location (at what point?), manner (how?), cause (why?), contingency (in what circumstances?), accompaniment (together with?), role (what as? what into?), matter (what about?), angle (says who?). Thus, ‘circumstances’ cover a diversity of categories: temporal, spatial, means, quality, comparison, reason, purpose, behalf, condition, concession, default, comitative, additive, guise, product. We here give an example of the tripartite f-structure of a clause, with experiential meaning included too. And yet textual

Theme

interpersonal

Adj

during all those years

had never

found out the name of the priest

Rheme Adjunct

Resexperiential

she

Subject

Fin/

Mood Circumstance

Senser

Pred

Compl.

-idue Process

Circ.

In this example, the Process is realized by a verbal group that refers to an event (‘had never found out’) consisting of a particular quantum of experience unfolded through time. The Participant (‘she’) is the entity involved in the Process; Circumstances have a less direct involvement in the Process, being dependent on it. The following sections of this chapter will focus on the specificity of each clause type, in terms of Processes and participants, only leaving the treatment of circumstances until later (cf. 4.4.5.).

1 SFG calls the ability of lasting through time and being located in space permanence. By contrast, each Process is most likely a unique occurrence. SFG names this quality transience. Thus, change involves both transience and permanence (IFG, 177).

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

113

4.4.1. Transitivity Structure of Material Cla use s A material clause is a clause of doings (actions) and happenings (events). By definition, a material clause assumes some input of energy from the part of an inherent participant – the Actor. ‘This participant brings about the unfolding of the process through time, leading to an outcome that is different from the initial phase of the unfolding’ (IFG, 180). In the following example, the material clause represents a happening (intransitive clause in traditional grammars). A little hand-mirror hung above the washstand Actor Process: material Circumstance

The material clause represents a doing when the unfolding Process moves towards a participant other than the Actor, usually the Goal. The equivalent in traditional grammars is the transitive clause. Then Ginger Mooney Actor

lifted Process: material

her mug of tea Goal

The representation of a Material clause can have two forms: operative (traditionally active voice) (#a), or receptive (traditionally passive voice) (#b). (a) textual interpersonal

A complete Wordsworth Theme Subject

experiential

Mood Actor

(b) textual interpersonal Experiential

The books on the white wooden shelves Theme Subject Mood Goal

stood Rheme Finite/ ‘(past)’

at one end of the lowest shelf

Pred ‘stand’ Residue Process: active were arranged

Adjunct Circumstance

from below upwards

Rheme Finite Pred Adjunct Residue Process: Circ passive

according to bulk.

Adjunct Circumstance

The main difference between the two representations consists in their interpersonal mapping. Whereas in the operative variant the Actor is

114

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

conflated with the interpersonal function of Subject, in the receptive variant the Goal is conflated with the Subject. Further distinctions can be made between material clauses in terms of the nature of the unfolding of the material Process through time. This is the source of the distinction that SFG draws between (i) ‘creative’ clauses, where Actor/Goal is brought into existence by the unfolding Process and (ii) ‘transformative’ clauses, where Actor/Goal is only transformed by the unfolding Process (IFG, 184). The outcome of a transformative Process can be described as elaboration, extension, or enhancement. A few examples are in order here, quoting Halliday (IFG, 185-6, 190). Limestone can form in many ways. (creative operative happening) I started writing short stories while I was at Yale. (creative operative doing) These are formed by chemical precipitation... (creative receptive doing) I served in World War II. (transformative operative happening) He runs The Washington Post. (transformative operative doing) all the marble fittings of the interior were painstakingly dismantled. (transformative receptive happening)

Other participant roles are Scope, Recipient, Client, and Attribute. With other participants, these may occur as Subject, but they are closer to circumstance than other participants, a fact reflected in their potential to be marked by prepositions. Scope is the only participant that is not affected by the unfolding Process. It construes the domain over which the Process takes place. The deceased had been in the habit of crossing the lines late at night from platform to platform I will not attempt to play tonight the part that Paris played on another occasion.

In both clauses, ‘the lines’ and ‘the part that Paris played on another occasion’ exist independently of their respective Processes, that is the act of ‘crossing’ and the act of ‘playing’, and they indicate the domain over which the Process takes place. Therefore, they are both participants functioning as Scope. In certain contexts, Scope can also be another name for the unfolding Process. This is the case with verbs like ‘play’ or ‘sing’, which often cooccur with nouns that lack independent content, such as ‘game’ or ‘song’. Such verbs are described as acts – ‘the act of playing’ or ‘the act of singing’– and are considered lexically empty verbs. Therefore, one can say that in clauses of the type they played games, or she sang a song, ‘games’ and ‘song’ are participants functioning as Scope defined as Process (IFG, 193). The first clause below illustrates the Scope as entity, the second, Scope as Process.

SYSTEMIC FUNCTIONAL GRAMMAR: A READER We Actor

crossed Process: material

The other men played Actor Process: material

115

the Liffey Scope: entity game after game Scope: process

Scope occurs typically in intransitive clauses, thus only Actor is present as participant. Since no Goal occurs, Scope can easily be assumed to be the Goal. In order to distinguish Scope from Goal, various criteria are available. First, Scope is not affected by the unfolding Process, but rather defines it or explains it. Second, Scope can never have a resultative Attribute (see infra). Third, Scope cannot be a personal pronoun. Fourth, it cannot normally be modified by a possessive (IFG, 194). Recipient and Client both construe a benefactive role, but differ in that Recipient benefits from the receipt of goods (in a transforming doing clause), while Client benefits from receipt of services (in a creative clause). such person may choose to alternatively give you [Recipient] a replacement copy she always gave the visitor [Recipient] one or two slips from her conservatory ... conscious of having created an effect on his audience [Client]...

Attribute belongs mainly to the realm of relational clauses, but can appear also as alternative function for a participant in a Material clause. A depictive Attribute construes the state of the Actor or Goal during a process; a resultative Attribute construes the state of the Actor or Goal after the process has been completed. the cars came scudding (Attribute: depictive) in towards Dublin I bore my chalice safely (Attribute: resultative) through a throng of foes

An Attribute in a Material clause is an optional specification, as opposed to an Attribute in a Relative clause, which is obligatory (cf. 4.4.3.). 4.4.2. Transitivity Structure of Mental Clauses Mental clauses construe a quantum of change in the flow of events taking place in our own consciousness, they belong to the world of our inner awareness. The central element of a mental clause is the Process of sensing that encodes the personal experience (emotive, desiderative, cognitive or perceptive). Verbs such as (dis)like, intend, think, or taste, respectively, occur characteristically in mental clauses. A mental Process implies the presence of two participants – the entity involved in a conscious processing (such as affection, desire, cognition, or

116

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

perception), and the content of that processing. SFG calls the main participant of a mental clause the Senser. Such an entity has to be endowed with consciousness; it is applied figuratively when the given entity is not expected to possess this ability (e.g., usually toys and animals in children’s books). The second participant is called the Phenomenon. Phenomenon is a participant that can be either a real entity or an abstract one. Gabriel recognised the prelude [Phenomenon: thing: real] he recognised familiar faces [Phenomenon: thing: abstract]

Unlike the participant in material clauses, or the Senser in mental ones, clauses that function as Phenomenon can refer to acts (macrophenomenal) or to facts (metaphenomenal). ‘An act is a configuration of a process, participants involved in that process and possibly attendant circumstances’ (IFG, 204). In macrophenomenal clauses, the verb is typically non-finite and the Phenomenon is restricted to perception. I saw him walking slowly away from us towards the near end of the field Senser Process Phenomenon: act my uncle was seen turning the corner Senser Process Phenomenon: act

A fact is an abstract, semiotic-like phenomenon: ‘a proposition (or sometimes a proposal) construed as existing in its own right in the semiotic realm, without being brought into existence by somebody saying it.’ (IFG, 205) In metaphenomenal clauses, the verb is typically finite and the Phenomenon is restricted to emotion, ‘construed as impinging on the Senser’s consciousness.’ (IFG, 205) The problem of how he could pass the hours till he met Corley again Phenomenon: fact

troubled Process

him Senser

... he thought what changes those eight years had brought ... Senser Process Phenomenon: fact [my aunt] hoped it was not some Freemason affair ... Senser Process Phenomenon: fact

4.4.3. Transitivity Structure of Relational Clauses In SFG, relational clauses correspond to ‘copula’ constructions in traditional grammars. Relational clauses seem complex because they are less

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

117

explored in traditional grammars and have a rich potential for ambiguity. Relational Processes construe experience, both material and mental, as ‘being’ rather than as ‘doing’ or ‘sensing’, hence the most frequent verbs that appear in relational clauses are ‘to be’ and ‘to have’. Since relational clauses serve to characterize or identify, two different types are established: attributive clauses and identifying clauses. Typically, the verb that realizes relational Processes, both attributive and identifying, is ‘to be’, but other verbs can occur too. Here resides the potential for confusion but also the potential to differentiate between an attributive Process and an identifying one. The lexical verb of the verbal group realising the Process in an attributive clause belongs to the ‘ascriptive’ class (e.g., appear, qualify as, look, weigh, cost, matter, differ, vary, confirm). In the case of an identifying clause, the lexical verb belongs to the ‘equative’ class (e.g., play, act as, mean, suggest, equal, comprise, represent, illustrate, express, choose). Participants in attributive clauses (Carrier and Attribute) are related by class membership, in identifying clauses (Identified and Identifier) by identity. Whereas the participant roles of the former belong to the same order of abstraction, the participant roles of the latter are of different order of abstraction, being related only symbolically. Distinguishing between the two classes of relational clauses can also be facilitated by the kinds of nominal element that realize the participant. Thus, the nominal group functioning as Attribute is typically indefinite and construes an entity (Carrier) that has some quality assigned to it (Attribute), be it a class of thing (intensive), a possessive relationship (possessive), or a circumstance (circumstantial).1 His face Carrier

was Process: intensive

very truculent , grey and massive Attribute

The nominal group functioning as Identifier is typically definite and construes an entity (Identified) that has an identity assigned to it (Identifier), be it a value (intensive), a relation of possession (possessive) or a circumstance (circumstantial).

1 Although ‘Attribute’ is mainly realized by nominal groups (though rarely proper nouns), the Attribute can also be realized by a prepositional phrase in the case of a circumstantial attributive clause.

118

Maria Identified

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

is Process: intensive

my proper mother Identifier

The participant roles of identifying clauses can be swapped (clauses are reversible without changing the meaning), but this is not possible in attributive clauses. my proper mother Identifier

is Process: intensive

Maria Identified

Another test to allow distinguishing between relative clause types is the interrogative probe. For an attributive clause as his face was very truculent the question is how was his face? For an identifying clause as Maria is my proper mother the question is who is Maria? Thus, the question for an attributive clause refers to the quality, whereas the question for an identifying clause refers to the identity. Attributive clauses should not be confused with existential clauses, which have only one participant – the Existent (cf. 4.5.4.). Although there is some overlap between relational and mental clauses, the two are not be confused either. Mental clauses typically have two participants (4.4.2), of which the equivalent of the Phenomenon appears in attributive clauses only as circumstantials.1 The relational clauses system in English operates with two distinct modes of being (attributive and identifying), each having three types of relation – ‘intensive’, ‘possessive’ and ‘circumstantial’, producing six categories of relational clauses.2

For other indicators to differentiate mental clauses from attributive clauses in relations to the submodifiers used in the nominal group, and the tense of the verb, see IFG, 224. 2 Relational clauses vary more than is shown in Table 4.5; this is mainly due to the diversity of lexical verbs that realize the respective Processes but also to the complexity of the relationship between the participants too. 1

types of relation

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

‘intensive’ x is a ‘possessive’ x has a ‘circumstantial’ x is at a

119

modes of being ‘attributive’ ‘identifying’ a is an attribute of x a is the identity of x Sarah is wise. Sarah is the leader. (The leader is Sarah.) Peter has a piano. The piano is Peter’s. (Peter’s is the piano.) The fair is on Tuesday. Tomorrow is the 10th. (The 10th is tomorrow.)

Table 4.5: Categories of relational clauses (IFG, 216, Table 5(12))

Attributive clauses of the intensive type define the Attribute by reference to an entity or quality (IFG, 220-22). Thus, they are very useful for characterizing the entities who serve as the Carrier. Mrs. Mooney Carrier He Carrier

was Process: intensive

was Process: intensive

a butcher's daughter. Attribute: entity

a little man, with a white, vacant face. Attribute: quality

Identifying clauses of the intensive type define the Identifier by reference to an Identified, therefore useful for producing definitions, in both everyday discourse and academic discourse.1 The relationship between the Identified and the Identifier revolves around the identity of the former that the latter describes. Therefore, ‘a veritable peace-maker’ in you are a veritable peace-maker does not describe an identity but rather a quality, that is pacifism. Consider the following two clauses from Joyce, of which the second is interrogative, thus the WH- particle taking the Identifier role. we Identified

are Process: intensive

the Three Graces Identifier

1 Modern English makes use of many other sub-types of identifying clauses, such as equation, equivalence, role-play, definition, symbolization, exemplification, and demonstration (IFG, 234-5). Since such typology is genre related, they can be found in the scientific, commercial, political and bureaucratic contemporary discourse, as well as in the ancient equivalent texts.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

120 What Identifier

are Process: intensive

the goloshes? Identified

Since in identifying clauses the participants refer to the same thing and the verb to be does not have a passive form, confusion can occasionally arise about the exact function of its participants. Clauses like Alice is the clever one are ambiguous in this way. Is ‘Alice’ the Identified or the Identifier? For that reason, Halliday proposes a supplementary pair of terms to define the participants and conflate with the Identified/Identifier pair. Thus, as the Identified can be either Token or Value so the Identifier can conflate with either Token or Value. The rule is that, in an operative clause, Subject is always also Token, and in a receptive clause Subject is always also Value. Once this obstacle is taken away, there is only need to distinguish an operative clause from a receptive one. Although lexical verbs such as mean, represent, play or spell do not raise problems regarding this matter, the main verb to be stands as the most problematic. The test offered by Halliday to identify a receptive representation of an identifying clause stipulates replacing the verb to be with the best suited verb from the group of lexical verbs preferred to represent a Process in an identifying clause. Few examples will make this discussion more relevant. They are laid out with the relevant questions that may have prompted them, their subsequent mood structure and its receptive representation (cf. IFG, 231). Which are you? (Which part do you play?) We are (=play) the Three Graces Subject Process: Complement Identified/T intensive Identifier oken /Value

Its marked version would be: The Three Graces Subject Identifier /Value

are (= are played by) Process: intensive

us. Complement Identified /Token

Attributive clauses of the possessive type construe the relationship between the two participants in terms of ownership either as attribute or as Process. If possession is construed as attribute, it takes the form of a possessive nominal group.

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

the piano Carrier

is Process: intensive

121

Peter’s Attribute: possession

Syntagmatically, such a clause is not different from an identifying clause. ‘Peter’s’can describe both a quality of the piano and its identity (belongs to Peter). If possession is construed as Process by means of possessive lexical verbs (e.g., need, deserve, feature, contain, include), it can generate two situations in which the Carrier can either be possessor or possessed (IFG, 245). Peter Carrier: possessor

has Process: possessive

a piano Attribute: possessed

the piano Carrier: possessed

belongs to Process: possessive

Peter Attribute: possessor

In the first example, ownership is an attribute ascribed to Peter; in the second, ownership is an attribute ascribed to the piano. Identifying clauses of the possessive type construe possession as a relationship between two entities as a feature of the participants (#a), or of the Process by means of possessive lexical verbs (e.g., possess, own, comprise, afford, cf. #b). (a) the piano Identified/Token: possessed

is Process: possessive

(b) The wild garden behind the house Identified/Token: possessor

Peter’s Identifier/Value: possessor

contained Process: possessive

a central apple-tree Identifier/Value: possessed

Attributive clauses of the circumstantial type construe the relationship between the participants in terms of various circumstances (e.g., time, place, manner, cf. 4.5.5.), which are ascribed as Attributes to some entity (#a), or expressed by the lexical verb (e.g., take, last, range, concern, #b) (cf. IFG, 240-41).

122 (a) my story Carrier (b) my story Carrier

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

is Process: intensive

about a poor shepherd boy Attribute: circumstantial

concerns Process: circumstantial

a poor shepherd boy Attribute

Identifying clauses of the circumstantial type construe the relationship between the participants in terms of circumstantial features by which they are related (#a), or expressed by the lexical verb (e.g., follow, support, match, contradict, discuss, #b) (cf. IFG, 242-44). (a) tomorrow Identified/ Token (b) the fair Identified/ Token

is Process: intensive

the tenth Identifier/ Value

occupies Process: circumstantial

the whole day Identifier/ Value

4.4.4. Transitivity Structure of Other Types of Cla uses To the three principal Process types discussed previously, there are three more secondary types to be added: behavioural at the boundary between material and mental, verbal at the boundary between mental and relational, and existential at the boundary between relational and material. Behavioural clauses ‘construe (human) behaviour, including mental and verbal behaviour, as an active version of verbal and mental processes.’ (WFG, 109) Being a border area, behavioural clauses blend the material and mental into a continuum. Thus, here are included psychological Processes (e.g. twitch, shiver, tremble, sweat, etc.), social Processes (e.g. kiss, hug, embrace, chat, talk, gossip, etc.), or physiological Processes (e.g. cough, spit, stutter, etc.). Among the lexical verbs that serve as Process in behavioural clauses, some verbs are closer to the mental type – look, watch, stare, listen, think, dream, gossip, talk, argue, cry, laugh, hiss, nod, breathe, cough, burp, faint, sleep – and others are closer to the material type – sing, dance, lie, sit. Behavioural Processes do resemble mental Processes in that they endow the main participant with human consciousness (called, in this case, Behaver). However, they resemble material Processes by their inability to

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

123

project. Although typically a behavioural clause consists of Behaver and Process only, a secondary participant, analogous to the Scope of a material clause, may also appear. A participant such as ‘a great yawn’ in he gave a great yawn is called Behaviour (IFG, 251). Among the preferred circumstances, those of Matter and Manner are more common (IFG, 250-1). That night I Behaver He Behaver

slept badly. Process: behavioural Circ: Manner

was trembling now with annoyance. Process: behavioural Circ Circ: Manner

Behavioural Processes represent ongoing activities and cannot be reported as being thought. To differentiate them from verbal Processes one should check if the Process in question can report another clause. Thus, Processes realized by speech verbs are not necessarily verbal clauses; some (not reporting speech) are behavioural ones. They Behaver

talked volubly and with little reserve. Process: behavioural Circumstance

Miss Ivors had praised the review. Behaver Process: behavioural Target

Verbal clauses are realized by Processes of saying, such as praise, insult, slander, flatter, blame, speak, talk. They may include also semiotic Processes that are not necessarily verbal, such as tell, report, announce, explain, ask, question, enquire, require, promise, threaten, implore, etc. Probably the most familiar lexical verbs to realize the Process of a verbal clause are the neutral verbs such as say, tell (IFG, 255). Other lexical verbs are more frequent in academic discourse: point out, suggest, claim, assert. Obviously, ‘saying’ has to be interpreted in a broader sense, including symbolic exchange of meaning also. The main participant in a verbal clause is the Sayer. Other participants may also occur. The Receiver is the participant to whom the saying is directed, typically realized by a nominal group denoting a conscious being. My aunt said Sayer Process: saying

to him Receiver

The Verbiage is the function that corresponds to what is said, either the content of what is said or the name of the saying. It may refer to a

124

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

generic category (e.g. story, fable, etc.), a speech functional category (e.g. lie, question, etc.), or a lexicogrammatical category (e.g. word, phrase, etc.). they Sayer

were talking Process: saying

about feats of strength Verbiage

he Sayer

would narrate Process: saying

the incident Verbiage

to the boys Receiver

More rarely, a Target may also appear, which is a function that refers to the entity targeted by the Process of saying. The reading of the card Sayer

persuaded Process: saying

me Target

that he was dead Verbiage

Verbal clauses have the ability to project, i.e. to produce a clause that is not a direct representation of experience but a representation of a (linguistic) representation (IFG, 441). Projection can take the form either of a locution (direct quotation, #a) or of an idea (indirect quotation, #b). In the former case, the main clause carrying the verbal lexical verb is said to be quoting, whereas in the latter, it is reporting. (a) “There was no row,” Quoted (b) I said Sayer Process Reporting

said Process Quoting

Gabriel Sayer

moodily Circ: manner

I wouldn't Reported

Verbal Processes are not to be confused with mental Processes, though. The main test to differentiate between them is by checking whether a Receiver role is possible (the Process has projection), in which case a verbal Process is used. I Senser He Sayer

hope Process: mental proposed Process: saying

you'll spend an evening with us Phenomenon that we should charge them Verbiage

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

125

Besides, the Sayer does not require to be a conscious participant as a Senser demands. it Sayer

was reported Process: saying

that he had a vocation for the priesthood Verbiage

The Existential clause is a mixed and rare type. Various lexical verbs can realize the Process of an existential clause (e.g. exist, remain, arise, occur, follow, sit, hang, emerge). This happened morning after morning. He emerged from under the feudal arch of the King's Inn. the distribution of the bottles of stout took place amid general merriment.

When the lexical verb be is used, it resembles relational clauses. This time, existential clauses employ the word there, which is neither participant nor circumstance; it serves only to indicate existence and functions as Subject (#a). This ‘there’ is to be differentiated from the circumstantial relational (#b). (a) there was a yacht piano in the cabin (b) there's poor Nannie

Unlike relational clauses, though, existential clauses have only one participant, called the Existent. Frequently an existential clause can take a circumstantial element, though. There was Process: existing

a pause Existent

in the very breathing of the clerks. Circumstance: Location

On the borderline between material and existential there is a category of Processes that realize meteorological events. In English, such events may be realized in four different ways: as a material Process (the wind is blowing), existential Process (there was a storm), relational attributes (it’s foggy), or as a limiting case of a material clause consisting of Process only (it’s raining) (IFG, 258-9). 4.4.5. Circumstantial Elements Circumstantial elements are the most peripheral elements of the experiential structure of a clause. All circumstantials are defined in relation to the unfolding of their respective Process. They are typically realized by prepositional phrases, adverbial groups, or prepositional phrases whose Head is an adverb or a preposition.

126

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Prepositional phrases that provide the ‘setting’ for the Process are functioning as ‘circumstances’. When prepositional phrases introduce clauses that have no independent meaning, they are functioning as part of a dependent clause. If there are any Adjuncts with reference to location or time, exploited rather textually, as anaphoric or cataphoric referents, they are to be considered conjunctive Adjuncts (cf. 4.3.1.). There are nine types of Circumstances (IFG, 262 ff.). Extent refers to the distance in space, or the duration or frequency in time, over which the Process unfolds. Extent thus answers questions such as ‘how far?’, ‘how long?’, ‘how many times?’. Typically, Extent is realized by a nominal group with a quantifier, either definite (‘ten miles/years’, ‘ten times’) or indefinite (‘a couple of miles/years’, ‘several times’). Location refers to static placement in space/time, but also to the source, path and destination of movement in space/time. Location answers questions of the type ‘when?’ and ‘where?’1 In English, location circumstantials can be realized without the input of prepositions, being made eligible for a Participant function in a similar fashion to those of extent. Manner refers to the way in which the Process is actualized. It can refer to the Means whereby a Process takes place, the Quality of the Process, the Comparison or Degree that describe the Process. The syllables of the word Araby were called to me through the silence [Circumstance: Manner: means] in which my soul luxuriated and cast an Eastern enchantment over me. ... (the preacher) slowly [Circumstance: Manner: quality] surveyed the array of faces Like illumined pearls [Circumstance: Manner: comparison] the lamps shone from the summits of their tall poles upon the living texture ... It hardly [Circumstance: Manner: degree] pained him now to think how poor a part he, her husband, had played in her life.

These circumstances usually respond to questions such as ‘how?’, ‘what with?’, ‘what like?’. Typically, Quality and Degree are functions realized by adverbial groups, while Degree offers a general indication of measurement. Cause indicates the reason for or purpose of a Process, or specifies its beneficiary. It comprises three sub-categories, typically realized by prepositional phrases: Reason, Purpose, and Behalf. Reason relates the Process to the prior situation leading or bringing about its actualization. Purpose relates the Process to a planned or intended situation towards Tourist guides and history books use Circumstantial elements of this type with great success. 1

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

127

which it is directed. They answer the questions ‘why?’ and ‘what for?’ respectively. Behalf is the circumstantial participant for whose sake the action is undertaken, and answers the question ‘who for?’. ... before the summer was over he'd go out for a drive [Circumstance: Cause: purpose] ... these two young men (who were cousins) were also in good humour because of the success of the French cars. [Circumstance: Cause: reason] ... Mahony's big sister was to write an excuse for him [Circumstance: Cause: behalf] ...

Contingency focuses on a situation or event on which the actualization of the Process depends. There are three types of Contingency: condition, concession, and default. Condition refers to a situation or event which must be actualized if the Process itself is to be actualized, i.e., if the situation concerned is not actualized, the Process will not take place. Concession refers to a situation accompanying the Process that may rather have been expected to prevent or hinder it. Default refers to the non-actualisation of an event that would have been expected to occur. It was only in case of corporal injuries [Circumstance: Contingency: condition] I'd have to report it at the station. (Joyce, Ulysses) ... in spite of the cold [Circumstance: Contingency: concession] they wandered up and down the roads of the Park for nearly three hours. Today at the City of Dublin Hospital the Deputy Coroner (in the absence of Mr. Leverett) [Circumstance: Contingency: default] held an inquest on the body of Mrs. Emily Sinico ...

Accompaniment describes an entity that joins participation in the Process. Its two sub-types – comitative and additive – are expressed by prepositional phrases. They both have a positive and a negative aspect, depending on whether they answer the question ‘who/what else?’ (positive polarity) or ‘but not who/what?’ (negative polarity). The main difference between a comitative and an additive circumstance is that the former represents the Process as a single instance with two entities involved, whereas the latter represents the Process as two instances and both entities sharing the same participant function. ... they used to play every evening with other people's children [Circumstance: Accompaniment: comitative]. She said she would do without any tea [Circumstance: Accompaniment: comitative: positive] ... I like that idea very much but wouldn't a comfortable spring bed do them as well as a coffin [Circumstance: Accompaniment: additive: positive]? He was so enraged that he wrote Bernard Bernard instead of Bernard Bodley [Circumstance: Accompaniment: additive: negative] and had to begin again on a clean sheet.

128

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

The two sub-types of Accompaniment differ also in the set of prepositions used as Head of the prepositional phrases. Thus, prepositions like ‘with’ or ‘without’ serve the comitative function, and prepositions ‘as well as’ or ‘instead of’ and the like serve the additive function. Role construes the meanings ‘be’ and ‘become’ circumstantially. Its two sub-types – guise and product – answer the question ‘what as?’ and ‘what into?’ respectively. He had started as a deck boy [Circumstance: Role: guise] at a pound a month on a ship of the Allan Line going out to Canada.. ... they had been cut into long thick even slices [Circumstance: Role: product] and were ready to be handed round at tea.

Matter refers to something that is described, referred to, narrated, etc. It parallels the verbal Processes already considered, and stands as the circumstantial equivalent of the Verbiage. Matter circumstantials answer the question ‘what about?’. Prepositions such as about, with reference to, concerning occur at the Head of the relevant prepositional phrases. He had told me stories about the catacombs and about Napoleon Bonaparte ...

Angle describes the origin of information (source), or the attitude or outlook held by one of the participants (viewpoint). Source angle is expressed by prepositional phrases such as ‘according to’, ‘in the words of’, or the like for. Viewpoint angle is indicated by phrases such as ‘in the view/opinion of’ or, simply ‘to’, accompanied by an appropriate head noun phrase such as ‘my way of thinking’ or even the simple pronoun, as in ‘to me, this was a disaster’. ... according to the terms of the contract [Circumstance: Angle: source], she should receive the sum originally stipulated for ... But granted even that, it is, to my mind [Circumstance: Angle: Viewpoint], a princely failing ...

4.4.6. Ergativity as an Alternative Interpretation to Transitivity The phenomena of experiential grammar can be more diverse than previously detailed. One good reason for expanding the knowledge at this level owes to the fact that (what traditional grammar would label as) voice has not as yet been brought into our discussion. Halliday proposed an alternative model to that of transitivity, called ergativity. Ergativity is not the name of a new system, but a property of the system of transitivity, which

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

129

includes both the ‘transitive’ model and the ‘ergative’ model (IFG, 281 cf. footnote). The variable that the transitivity/ergativity distinction indexes is one of causation; the question is: is the Process brought about from within or from outside. This is different from the transitive/intransitive distinction, which indexes the extension of the Process. In traditional grammars, ergativity is a pattern of marking transitive subjects, as opposed to absolutivity that is used for marking intransitive subjects (Kroeger 2005, 105-107). Medium is the entity without which there would be no Process at all. The Participant that realizes the external cause is the Agent. Medium corresponds to Actor in a happening clause or to Goal in a doing clause. In behavioural clauses, Medium is equivalent to Behaver, in mental clauses to Senser, in identifying clauses to Carrier, in existential clauses to Existent. In a verbal clause without an Agent, Medium conflates with Sayer, but when an Agent is present it conflates with Target. Medium is the only entity never introduced by a preposition (IFG, 291-2). material transitive interpretation ergative interpretation

Tom’s eyes Actor

closed Process: happening

Medium

verbal transitive interpretation ergative interpretation

the teacher Sayer Medium

said Process: saying

From an ergative perspective, a happening is said to construe a Process whose actualization is represented as being, a self-endangered or self-caused ability (the boat sailed) that does not describe a doing and assumes the contribution of another Participant, external to the core combination of Process + Medium (Mary sailed the boat). Thus, in more general terms, an ergative model of a clause takes into consideration the ability of an unfolding Process to be caused by another entity than the main Participant (Medium). On the contrary, a transitive/non-transitive traditional model would consider the ability of an unfolding Process to extend beyond the Actor to another entity (e.g., the tourist hunted – non-transitive; the tourist hunted the lion – transitive). Other participants may also appear as part of the ergative structure, but only as secondary Participants. Their place in relation to the other component is best explained when reference to the imaginative model is

130

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

made. Contrary to the linear transitivity model, the ergative model is imagined nuclear. Consequently, the secondary Participants occupy the circle at the periphery of the nucleus, representing Process + Medium, leaving the Circumstantial elements at the most peripheral circle. When present, the Agent is the initiator (external cause) of the Process. It can conflate with Phenomenon (mental clause), Attributor (attributive clause), Assigner (identifying clause), or Sayer (verbal clause with agency). Alongside the Agent, among the Participants of the extra-nucleus enclosure, one can also find a Beneficiary and/or a Range. Beneficiary is the one to whom or for whom the Process is said to take place. Since it can appear only in material, verbal and a few relational clauses, it conflates with the function of Recipient/Client, Receiver, or Beneficiary respectively. material mood interpretation

he Subject Mood Actor Medium

transitive function ergative function

would Finite

give him Pred Compl Residue Process: Recipient material Beneficiary

such a whipping Complement Goal Range

Range specifies the domain of the Process and appears in all clauses but existential ones as Scope (material), Behaviour (behaving), Phenomenon (mental), Verbiage (saying), Attribute (attributive), Value (identifying). behavioural mood interpretation transitive function ergative function

he Subject

stared Fin/Predicator

blankly Adjunct

down the staircase Complement

Behaver Medium

Process: behaving

CManner

Behaviour Range

mental

he

could not feel

her

mood interpretation transitive function ergative function

Subject

Fin/Predicator

Complement

near him in the darkness Adjunct

Senser

Process: sensing

Phenomenon

CLocation:

verbal mood interpretation transitive function ergative function

Medium he Subject Sayer Medium

place Range

described Fin/Predicator Process: saying

some of the practices Complement Verbiage Range

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

131

In order to facilitate the correct identification of ergative Participants, a few additional comments may provide some insight. Agent, Beneficiary, and Range are semantically mixed constituents, sharing features with both Participants and Circumstances. Thus, they can occur with a preposition to function as part of a prepositional phrase. Such a tendency is noted whenever a Participant other than Medium takes the place of prominence in the message (IFG, 295). The way ‘voice’ works in Halliday’s SFG assumes that a clause without ‘agency’ is neither active nor passive; it is middle. One with ‘agency’ is nonmiddle, i.e. effective. In turn, an effective clause is either active (Agent and Subject are conflated), passive (Medium and Subject are conflated), or even medio-passive (when Range or Manner and Subject are conflated). Consider the following samples of material clauses (IFG, 289-90). middle clause mood interpretation transitive function ergative function

The paper of his cigarette Subject Actor Medium

effective active clause mood interpretation transitive function ergative function

Mr. O'Connor Subject Actor Agent

effective passive clause

Her blue serge skirt Subject Goal Medium

mood interpretation transitive function ergative function

broke Fin/Predicator Process: happening

tore Fin/Predicator Process: doing

a strip off the card Complement Goal Medium

was held

at the waist

Fin/Pred Process: doing

Adjunct CLocation: place

effective medio-passive clause mood interpretation transitive function ergative function

they Subject Goal Range

were held up Fin/Predicator Process: doing

effective medio-passive clause

he

had been seen

mood interpretation transitive function ergative function

Subject Behaviour Range

Fin/Predicator Process: behaving

by a belt of black leather. Adjunct Actor Agent by Mary Jane Adjunct Actor Medium by many of his friends Behaver Medium

132

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Thus, in Halliday’s opinion, each clause can receive two interpretations to its experiential f-structure: a transitive one and an ergative one, even though some clause types seem not to support this option. But some degree of indeterminacy in analysis is recognized as allowable. One of the features of English grammar that ergativity enables us to understand is the matter of causality. Traditional grammars suggest that clauses such as the police exploded the bomb should be analysed as a complex sentence wherein the verb make is implied – the police made the bomb explode. The problem is that such causative renderings are not always possible (cf. the lion chased the tourist cannot be converted into the lion made the tourist chase, unless the verb is changed with another, e.g. run). For a transitivity structure to have sense in such a case, SFG introduces a new function to take account of the executive role. That is the Initiator in a material clause, an Attributor in an attributive clause, and an Assigner in an existential clause (IFG, 299). material transitive function ergative function

the police Initiator Agent

exploded Process: doing

attributive transitive function ergative function

His new office Attributor

identifying transitive function ergative function

A medieval doctor Assigner

Agent

Agent

made Process: being

the bomb Actor Medium him Carrier

professionally interested ... Attribute

Medium

Range

would have called Process: being

him Id/Token

saturnine Ir/Value

Medium

Range

4.5. LOGICAL S TRUCTURE At the ideational level of meaning, above the clause, SFG posits the existence of another grammatical unit, namely the clause complex. It stands for what ‘sentence’ represents in written material. The clause complex structure realizes logical meaning, externally augmenting a clause by means of another clause, as part of a clause complex (e.g., A happened and then B happened’). A clause complex consist of a chain of clauses, a pair at a time, related by interdependency (taxis). Such a pair of clauses is known as clause nexus. The basic structure of this pairing is binary, labelled HEAD + MODIFIER.

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

133

Since taxis can define two types of relation between clauses, the Head and its Modifier vary accordingly. Parataxis is the relation between two like elements of equal status; hypotaxis is the relation between a dependent element and its dominant. Thus, the Head of a clause nexus, also called the primary clause, could be either the initiating clause of a paratactic nexus or the dominant clause of a hypotactic nexus. In the latter case, the dominant clause does not need to be the first one of the chain. A practical method for analysing the clause complex starts out with identifying boundaries between clause complexes. In most modern languages, formal orthographic signs, such as a ‘dot’, mark the end of a clause complex. The identification of boundaries between ranking clauses follows. The clauses thus identified are written out each on a different line, with the necessary analysis on the left margin. This procedure may be called pragmatic mapping; it allows one to see at a glimpse the content, the context and even the relationship between clauses. Exceptions to this ideal situation are noted and discussed below. 4.5.1. Interdependency When two clauses in a complex are interdependent, the meaning of one clause depends on that of another. Some clauses can function independently. These are the primary clauses. Others lack this potential; these are secondary clauses. Primary clauses are always easy to discern by means of a question tag. As opposed to secondary clauses, which are normally introduced by a connective, a primary clause can stand without a connective as a functioning unit. 1 2

We waited on for a quarter of an hour more, ¨ but still there was no sign of Leo Dillon. ¨|

primary paratactic clause secondary paratactic clause

α β

How my heart beat ¨ as he came running across the field to me! ¨|

secondary hypotactic clause primary hypotactic clause

Occasionally, these two systems interact with each other in producing clause complexes that are less linear than the sequences of related clauses seen above. This can happen when one of the clauses connected in the nexus develops a sub-complex of its own. Such an internal bracketing is called nesting (IFG, 376).

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

134

I pretended to pray ¨

1 2

α

but I could not gather my thoughts ¨

β

because the old woman’s mutterings distracted me. ¨|

primary paratactic clause secondary paratactic clause/primary hypotactic clause secondary hypotactic clause

Taxis is the system concerned with the degree of interdependency relations between clause elements forming a complex. There are two options within the system of Taxis: parataxis (equal status) and hypotaxis (unequal status), standing for the traditional terms coordination and subordination, respectively. A different system of clause complexing is the logico-semantic. It consists of two options inspired by the transitivity system of the clause, namely expansion and projection. The former is most frequently found in relational clauses, the latter in verbal or mental clauses. Expansion generates expanded secondary clauses that relate phenomena as being of the same order of experience with the one in primary clause. Projection generates projected secondary clauses that relate phenomena of a higher order of experience than that of the primary clause. Both expansion and projection are realised as one of several subtypes; these will be illustrated in the following sections. The notation to be used is as follows: for subtypes of expansion, elaborating (=), extending (+), and enhancing (×), for subtypes of projection, locution (“) and idea (‘). Texts can normally be laid out as linear sequences of related clauses. A paratactic chain is marked as 1 ^ 2, whereas a hypotactic chain is marked as α ^ β, or β ^ α. The end of each clause is marked by ¨, whereas the end of the clause complex is marked by ¨|. There are alternative representations of interdependencies in a clause complex (IFG, 388-91; WFG, 169-70), but, due to its ease of construction, the model preferred here is the one derived from Halliday’s box diagram. 4.5.1.1. Parataxis Parataxis ‘is the relation between two elements of equal status, one initiating and the other continuing’ (IFG, 375). Units in a paratactic relation have the potential to stand independently. Consequently, their position in relation one to another is both symmetrical and transitive. ‘A and B’ also implies ‘B and A’ (symmetry); ‘A and B’ and ‘B and C’ together imply ‘A and C’

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

135

(transitivity). This relation is marked by Arabic numerals in sequence (1, 2, 3). 1 2

The cold air stung us ¨ and we played till our bodies glowed. ¨|

primary paratactic clause secondary paratactic clause

Internal bracketing, also known as nesting, may occur with paratactic clauses, without dependency being involved. 1 2

1 2

I wished to go in a¨ and looked at him b¨ but I had not the courage to knock. c¨|

Clause (a) connects both to clause (b) and to clause (c). Thus, the best way to explain the relationship between the above clauses is to propose that clause (b) is nested. Conjunctions generally used to mark the secondary clause in a paratactic nexus belong to the group of linkers (and, or, but – only with expansion relation) or correlatives (either ... or, not only ... but). 1 2

Either you believe, ¨ or you don't ¨

primary paratactic clause secondary paratactic clause (Joyce, Ulysses)

4.5.1.2. Hypotaxis As opposed to the paratactic relation, units within a clause complex linked in a hypotactic relation are unequal; there is always a dominant clause and at least one dependent clause. As a result, their relation is neither symmetrical nor transitive. Therefore, ‘A when B’ is not the same with ‘B when A’ (asymmetrical), ‘A when B’ and ‘B when C’ does not necessarily imply ‘A when C’ (intransitive). In this case, the preferred notation makes use of the Greek characters, with a ascribed to the dominant clause.1 Conjunctions marking the secondary clause in a hypotactic nexus are called binders. 1 If the clause nexus structure is interpreted from a Thematic perspective, the dependent clause in a α ˆ β (also called progressive) sequence is given rhematic status. Similarly, the dependent clause in a β ˆ α (also called regressive) sequence has thematic status. Conditional clauses and purpose clauses are by definition thematic β-clauses. This clause type can also be used in the staging of a narrative to re-orient the plot development, or to connect the text initiated with the previous one (IFG, 393).

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

136

α β

As Gabriel never ate sweets, ¨ the celery had been left for him.¨|

secondary hypotactic clause primary hypotactic clause

Some correlatives are also known, such as ‘if ... then’, or ‘because ... so’. β α

If he were a painter ¨ he would paint her in that attitude. ¨|

Subject is more frequently elided in the secondary clause of a paratactic nexus, but can be elliptical in a primary clause also. 1 2

I was surprised at this sentiment ¨ and [I] involuntarily glanced up at his face. ¨|

A dependent clause can enclose a dominant clause or be enclosed by it. Enclosure is marked by >. β

“Every boy,” >, “has a little sweetheart.” ¨| he said, ¨

The secondary clause of a hypotactic clause nexus does not need to be finite. Frequently, the dependent clause precedes the dominant clause (β ˆ α). β α

on my aunt’s nodding, ¨ [she] proceded to toil up the narrow staircase before us. ¨|

4.5.2. Logico-sema ntic Types: Expansion In terms of logico-semantic relationships, SFG proposes two types: expansion and projection. Expansion assumes that secondary clauses expand the primary clause by elaborating it (marked with =), extending it (+), or enhancing it (×). Projection assumes that the secondary clause expands the primary clause either as locution (marked with “), or as an idea (‘). 4.5.2.1. Expansion through Elaboration Elaboration stands for that relation between clauses by which the secondary clause continues with a further specification or description on the meaning of its primary counterpart. Elaboration may apply to both paratactic and hypotactic nexuses. There are two main differences between paratactic elaboration and hypotactic elaboration. In terms of meaning, the former involves exposition, exemplification, and clarification, whereas the latter involves description. In terms of realization, the main difference resides in

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

137

the fact that the latter either has a non-finite verb, or is introduced by a WH-element (IFG, 397). Of the three paratactic elaboration types, exposition restates the thesis of the primary clause in different words and potentially in more words. Conjunctions or conjunctive expressions such as or, in other words, that is to say, I mean, or the abbreviation i.e., can make explicit this relationship, but the occurrence of such a connective is not obligatory. 1 =2

He had been a very charitable priest; ¨ in his will he had left all his money to institutions and the furniture of his house to his sister. ¨|

Paratactic elaboration by exemplification develops the thesis of the primary clause by becoming more specific about it. Explicit renderings of this relation make use of conjunctives such as for example, for instance, in particular, or the abbreviation e.g. 1 =2

he is different; ¨ he goes in for games. ¨|

Clarification as paratactic elaboration backs up the thesis of the primary clause with some form of explanation. Expressions such as in fact, actually, indeed, and the abbreviation i.e. or viz support this relation. 1 =2

The tone of her voice was not encouraging; ¨ she seemed to have spoken to me out of a sense of duty.¨|

Hypotactic elaboration takes the form of a description, expressed in a non-defining relative clause, which functions as a gloss to the primary clause. When he smiled he used to uncover his big discoloured teeth and let his tongue lie upon his lower lip--a habit which had made me feel uneasy in the beginning of our acquaintance before I knew him well. ×

β

α

α =β

While he expressed these sentiments a¨ which bored us a little b¨ we kept silent. c¨|

Here, clause (b) is nested, because the main line is formed by clause (a), which stands as a temporal circumstance for its dominant – clause (c). This strategy is frequently used for introducing information into the discourse background. Hypotactic relative clauses can refer back to one of its constituents.

138

α =β

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS Sometimes he used to put me through the responses of the Mass ¨ which he had made me learn by heart ¨

Sometimes the hypotactic relative clause splits the dependent clause. His face, was of the brown tint of Dublin streets.

A complicated structure at Joyce follows. Note, though, that clause (c) elaborated by means of ‘when’. Her eyes, had a habit of glancing upwards a¨ when she spoke with anyone, c¨ which made her look like a little perverse madonna. d¨|

4.5.2.2. Expansion through extending Extension takes the form of addition, variation, or alternation; the secondary clause provides further details on the meaning of the primary clause, by adding, replacement, or the offering of an alternative. Paratactic extending is most frequently expressed in English by and, nor, or, but; other connectives used are in addition, also, moreover, on the other hand, although, instead, on the contrary. Two paratactically complexing clauses may adjoin each other without any causal or temporal relationship between them. In such cases, SFG speaks of addition. 1 +2

The cat escaped once more ¨ and Mahony began to throw stones at the wall Æshe had escaladed.Ø ¨|

When the secondary clause totally or partially replaces the primary clause, SFG prefers the term ‘variation’. 1 +2

Ernest had been her favourite ¨ but she liked Harry too. ¨|

Sometimes, the secondary clause presents an exception to what has been said in the primary clause. Then, the relation can be explained as ‘alternation’ (clauses (c) and (d)). 1 +2 +3 +4

I used to take them out, man, on the tram somewhere a¨ and pay the tram b¨ or take them to a band or a play at the theatre c¨ or buy them chocolate and sweets or something that way. d¨|

Hypotactic extending is not as common as the paratactic one. Their main characteristics are the lack of independence of the secondary clause,

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

139

and the use of initiating conjunctions. As with paratactic extensions, hypotactic constructions can be realized as addition, variation or alternation. Specific conjunctions are employed in each case: e.g. whereas, while, besides for hypotactic clauses of addition, except that, but that, instead of for variation, and if not ... then for alternation. α +β

One hand played in the bass the melody of Silent, O Moyle, ¨ while the other hand careered in the treble after each group of notes. ¨|

α +β

Is this what you read ¨ instead of studying your Roman History? ¨|

α +β

Anyway, if it's not your business ¨ it's my business ...¨|

4.5.2.3. Expansion through enhancing Enhancement offers details of a temporal or causal nature. This makes possible the integration of circumstantial clauses within the clause complex, in which case, the main function of the secondary clause is to qualify the meaning of the primary clause. This is done by reference to time, place, manner, cause or condition. Certain types of enhancement occur with both parataxis and hypotaxis. Paratactic enhancement of the temporal type connects two clauses in terms of temporality; they either share a common temporal background (simultaneity), or refer to it in succession (sequentiality). 1 × 2

There was a good deal of confused talk, ¨ and then Mr. Browne got into the cab. ¨|

Such circumstantial clauses have an important role in building up chronological sequences. Spatial enhancement of the paratactic type allows the secondary clause to introduce another event or happening, at the same location as that of the primary clause. 1 × 2

[I] slipped out the back into the garden ¨ and there was the poor fellow at the end of the garden, shivering. ¨|

Paratactic enhancement of the manner type construes the meaning of the secondary clause in reference to the means for the realization of its Process (#a), or in comparison with the Process of the dominant clause (#b)

140 (a) 1 × 2 (b) 1 × 2

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

could pull the indispensable wires ¨ and thus combine business with pleasure. ¨ We could find no dairy and so we went into a huckster's shop ¨ and so we went into a huckster's shop ¨

Clauses related paratactically as causal-conditional are of three types: cause-effect, condition, and concession-consequence. [cause ^ effect] 1 × 2

I knew that I was under observation ¨ so I continued eating ... ¨|

[effect ^ cause] 1 × 2

He seemed to be fairly old ¨ for his moustache was ashen-grey. ¨|

[condition: positive] 1 × 2

We'll have just one little smahan more ¨ and then we'll be off. ¨|

[concession ^ consequence] 1 × 2

They looked high up and low down; ¨ and still they couldn't see a sight of him anywhere. ¨|

[consequence ^ concession] 1 × 2

I lingered before her stall, ¨ though I knew my stay was useless, ¨

In paratactic environments, several conjunctions are found such as and, thus, so, then, but, though, sometimes even when (IFG, 415-6). In English, long sequences of clauses are more likely to be constructed paratactically than hypotactically. Hypotactic enhancement produces clauses of time, place, manner (quality, means, comparison), cause (reason, purpose, result), concession, and condition, known in traditional grammars as adverbial clauses. These may be finite or non-finite. Unlike the paratactic enhancement clause, the hypotactic do not move the discourse forward, but they move further away from the discourse (IFG, 416-7). Several examples follow.

SYSTEMIC FUNCTIONAL GRAMMAR: A READER [time] β Α ×

Whenever any job was vacant ¨ a friend was always ready to give him the hard word. ¨|

[place] β α ×

Wherever they gather ¨ they eat up the nation's vital strength. ¨| (Joyce, Ulysses)

[manner: quality] α × β

I wanted to see him, ¨ as it happens. ¨| (Joyce, Ulysses)

[manner: comparison] α × β

But the ball rolled down to her, ¨ as if it understood. ¨|

[manner: means] α × β

her goodman husband would not let her death ¨ whereby they were all wondrous grieved.¨| (Joyce, Ulysses)

[cause: reason] α × β

I could not gather my thoughts ¨ because the old woman's mutterings distracted me ¨|

[cause: purpose] α × β

Wherefore make unto yourselves friends out of the mammon of iniquity ¨ so that when you die they may receive you into everlasting dwellings.¨|

[cause: result] α × β

every two or three minutes a mist gathered on his glasses ¨ so that he had to take them off ¨|

[concession] β α ×

Though I was angry with old Cotter for alluding to me as a child, ¨ I puzzled my head to extract meaning from his unfinished sentences. ¨|

[condition: positive] α × β

If my uncle was seen turning the corner ¨ we hid in the shadow ¨|

141

142

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

[condition: negative] α × β

[I think] Christmas is never really Christmas ¨ unless we have the snow on the ground.¨|

Hypotactic spatial enhancement makes use of conjunctions such as wherever, where, etc. Such clauses are not to be confused with hypotactic elaborating clauses, which describe the dominant clause or one of its constituents. A hypotactic clause enhanced spatially is not an epexegetic clause that gives details about a constituent, but stands for a Location (space Circumstantial) which has been expanded to a clause. Consider the following example, which can be summarized as ‘He glanced there’ (viz. to where somebody was moving). α × β

He glanced behind him ¨ to where a face with dark thinking eyes followed towards the cardinal’s mausoleum.¨| (Joyce, Ulysses)

4.5.2.4. Embedded expanded clauses Embedding occurs when a given constituent has a function at an inferior level, as part of an inferior rank group. This is also called rank shift. Embedded clauses are down-ranked clauses that connect to a constituent belonging to the main clause, therefore functioning with it at the clause level as part of the clause structure, within the group (nominal or adverbial), but not with other clauses. A relative clause is a typical example of a rankshifted clause, and it is marked with Æ...Ø (IFG, 426).1 nominal group as Head (clause: finite)

ÆWhat he learnt from his other wife Myrto (ABSIT NOMEN!), Socratididion's Epipsychidion,Ø no man, not a woman, will ever know. (Joyce, Ulysses) nominal group as Postmodifier (clause: finite) but the window-panes of the houses Æthat looked to the westØ reflected the tawny gold of a great bank of clouds. nominal group as Postmodifier (clause: non-finite) Kate, Æbeing too feeble to go about much,Ø gave music lessons to beginners 1 In English, an embedded clause can function as Head or Postmodifier in nominal groups, but only as Postmodifier in adverbial groups (cf. IFG, 427, Table 7(12)).

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

143

nominal group as Postmodifier (phrase) the old gentleman had a horse Æby the name of JohnnyØ adverbial group as Postmodifier (clause: finite) He felt better after having eaten Æthan he had felt beforeØ adverbial group as Postmodifier (clause: non-finite) [he] liked nothing better Æthan to find himself at the head of a well-laden table. Ø adverbial group as Postmodifier (phrase) He would love that, he said, better Æthan anything in this world Ø

The relation of embedded clauses to their dominant units follows the three-fold typology of expansion already described for the relation between clauses: elaborating, extending, and enhancing. These clauses can have either a finite or a non-finite verb, and may be either operative or receptive (IFG, 428-37). In the present analysis, we confine ourselves to those factors relevant to our studies in Hebrew, sufficient for the identification of such clauses and differentiate them from similar syntactic constructions. Embedding by elaborating occurs typically with defining relative clauses. These are clauses introduced by who, which, that (sometimes even without them) that are restating their nominal antecedent. But he played too fiercely for us Æ=who were younger and more timid Ø. catching sight of the cat Æ=which had escaped himØ my heart was beating quickly with fearÆ= that he would seize me by the anklesØ.

Embedding by extension is possible only in the sense of possession, introduced by whose, which of (IFG, 432). And yet during all those years she had never found out the name of the priest Æ+whose yellowing photograph hung on the wallØ In one of these trimly built cars was a party of four young men Æ+ whose spirits seemed to be at present well above the level of successful Gallicism Ø

Embedding by enhancing produces a variety of clauses, some having the circumstantial sense located in the embedded clause (relative or adverbial clauses), others having it in their nominal antecedent (cf. IFG, 432-7). Some such nouns that are known as nouns of expansion can take only non-finite postmodifying clauses (e.g., purpose, point, aim, result, cf. IFG, 435).

144

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

relative clause I looked over at the dark house Æ× where she lived Ø. adverbial clause I continued eating Æ× as if the news had not interested meØ. circumstantial features in the nominal antecedent One time there used to be a field thereÆ× in which they used to play every evening with other people's childrenØ. they had good reason Æto be fussy on such a nightØ

Embedded clauses are to be distinguished from hypotactically dependent, non-defining relative clauses (cf. 4.5.2.1.). The former provide information (expansion) in relation to the identity of a participant (never in relation to the Process). In English, hypotactic (ranking) clauses are delimited by commas; these are absent in the case of embedded (down ranked) clauses. Embedded clauses must also be distinguished from enclosed clauses. An enclosed clause is one that breaks another clause into two parts, itself occurring in between the fragments. He would love that, , better than anything in this world...

Nominalization of Processes can also realize embedded clauses that can be treated as elaborations. Since a nominalized Process of this sort represents the name of an action, event, or other phenomenon, SFG calls it an act (cf. 4.4.2.). It can be realized by means of various Process types. ÆThe reading of the card Ø persuaded me Jack Mooney, the Madam's son, Æ...Ø, had the reputation Æof being a hard case.Ø

4.5.3. Logico-sema ntic Types: Projection ‘The logical-semantic relationship whereby a clause comes to function not as a direct representation of experience but as a representation of a (linguistic) representation’ stands for projection. At the discourse level, projection has various uses: constructing dialogue in narratives (#a), framing questions in conversation (#b), representing views (#c) and indicating sources (#d). (a) Everybody said: ¨ “O, here's Maria!” ¨| (b) I wonder ¨ where did they dig her up ¨ (c) I don't think ¨ that that's a fair question ¨

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

145

(d) he told me ¨ that the fathers of the Church had written books as thick as the Post Office Directory¨|

According to SFG, there are three systems involved in projection: the level of projection (idea vs locution), the mode of projection (reporting vs quoting) and the speech function (proposition vs proposal). Projection appears only with mental and verbal clauses (cf. 4.4.2, and 4.4.4.). As a result, the level of projection identifies only two locations. As the representation of a verbal clause, the projected clause is a locution, projecting wording. As the content of a mental clause, the projected clause is an idea, projecting meaning. “I have my own theory about it,” he said. A graceful image of his, he thought, and a just one.

Halliday prefers the association of these two kinds of projection with the conventions of cartoons: locutions are represented in ‘balloons’, whereas ideas are represented in ‘clouds’ (IFG, 443). The general convention is to mark locutions with quotation marks and the ideas with commas. In terms of modes of projection, taxis and embedding combine with projection. Parataxis projects as quotation, whereas hypotaxis projects as reporting. This distinction corresponds with the traditional distinction between direct and indirect speech. Level of projection and mode of projection intersect producing four kinds of projection nexus, as indicated in the table bellow (IFG, 444). locution idea

paratactic (quoted) 1 “2 Brutus said, “Caesar is ambitious” 1 ‘2 Brutus thought, ‘Caesar is ambitious’

hypotactic (reported) α “β Brutus said that Caesar is ambitious α ‘β Brutus thought that Caesar is ambitious

An embedded rendering of similar information would be: Brutus’ accusation Æthat Caesar is ambitiousØ created confussion in the Senate.

As for the speech functions of projection, it is to be noted that projecting clauses project questions, proposals, and even minor clauses, as well as statements.These are always connected paratactically with their projecting clause.

146

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

“Are you trying to get inside me?” 1¨ he asked. 2 ¨| “You're not to put the weight of your body behind it.” “Damn it!” 1¨ said Lenehan boldly 2 ¨|

“1 1

¨ Play fair,"

“1 =2

¨ he said. 2¨|

Surveying a text sample of 6100 clauses, IFG noted that out of 623 projection nexuses 95% of all projected ideas are reported (hypotactic), whereas projected locutions are evenly distributed between quoting and reporting. A more detailed analysis of the main projected clauses follows. 4.5.3.1. Projected Parataxis (Quotation) Quoting speech employs a paratactic relation between clauses, allocating one of the clauses as the quote of the other, which has its Process realized by a lexical verb of speech (verbal clause). Modern languages, including English, signal the content of direct speech by quotation marks. The projected clause can precede or follow the projecting one, but it can also be enclosed. Three possible arrangements result then: He had often said to me: 1 ¨ “I am not long for this world,” 2¨ “Everything in Paris is gay,” 1 ¨ said Ignatius Gallaher. 2 ¨| “Every boy,” “has a little sweetheart.” 1¨|

Various verbs are used for quoting. IFG groups them into four categories: verba declarandi specific for giving information (e.g., tell, remark, observe, report, etc.), or for demanding information (e.g., ask, demand, inquire, etc.), verbs with additional circumstantial features (e.g., reply, explain, protest, continue, etc.), or verbs with manner specifying connotation (e.g., insist, complain, cry, shout, etc.) (IFG, 448 table 7(19)). Quoting thought employs a paratactic relation between clauses, allocating one of the clauses as the quote of the other, which has its Process realized by a lexical verb of sensing (mental clause). Although the content of direct thought is signalled by quotation marks, the quotation is no more than a thought, not the real wording. The projected quoting thought clause is rather rare in English, but when it appears, verbs such as think, wonder, surmise occur in the projecting clause, as also verb expression’s such as ‘to say to oneself’. He just says to himself: 1 ¨ ‘The old one never went to see these wild Irish.(...)’ ‘2¨

4.5.3.2. Projected Hypotaxis (Reporting) Reported speech, traditionally called indirect speech, employs a hypotactic relation between clauses, allocating one of the clauses as the quote of the

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

147

other, which has its Process realized by a lexical verb of saying (verbal clause). Reported speech is abbreviated as α ˆ ″β. Indirect speech is frequently introduced by that, especially when the reported mood is indirect declarative. α ″ β ″ γ

She had been told 1 ¨ that Freddy had come 2¨ and that he was nearly all right. 3 ¨|

When the mood of the reported clause is interrogative, indirect speech can be introduced by whether, or WH- particles. α ″ β

He asked us 1 whether we had read the poetry of Thomas Moore or the works of Sir Walter Scott and Lord Lytton. 2 ¨|

α ″ β

I asked him 1 ¨ why he had brought it 2 ¨

Reported thought employs a hypotactic relation that has the Process of the projecting clause realized by a lexical verb of sensing (mental clause). The projecting clause refers to a phenomenon (the nature of someone’s mental state), whereas the projected clause refers to a metaphenomenon (the content of the mental state) by indirect statement or question. SFG abbreviates reported thought as α ˆ ′β. Typically, the verbs that construe the Process of projecting clauses are either of the cognitive (e.g., believe, guess, think) or desiderative class (e.g., want, like, intend), producing two kinds of projected clause, ‘like’ and ‘please’ respectively. ‘like’ type α ′β

I believe 1 ¨ they mean well, too. 2 ¨

α ′β

She thinks 1 ¨ you’ll marry her. 2 ¨

‘please’ type α ′β ′γ

Often he had wanted 1 ¨ her to go 2¨ and live with them 3¨

148

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

α ′β

it just struck him 1 ¨ that Fitz, Æ...Ø merely drove the car for the actual perpetrators of the outrage 2¨

Reported thought has multiple functions in texts, such as representing speaker’s or addressee’s thinking in dialogue, proving a character’s consciousness, indicating speaker’s angle or the source of a given opinion. Usually, an undecided state of mind is realized by an indirect projected question. α ′β

I did not know 1 ¨ whether I would ever speak to her or not 2¨

A final word is necessary on free indirect speech, which is a blend between the two main modes of projection – quoting and reporting. Although it preserves from the quoting clause the paratactic relation with the projecting clause, the free indirect clause is a report rather than a quote. This means that its time and person reference are shifted from those expected from a quotation, making it difficult to identify who is referred to in the indirect quotation. Usually third-person pronouns are used in reference to the reported speaker. ′β α

Melancholy was the dominant note of his temperament,1 ¨ he thought 2¨

α ′β ′β α

She thought 1 ¨ she would have to stand in the Drumcondra tram 2 ¨ A graceful image of his, > and a just one 1 ¨ >2 ¨

α ′β

Then she thought 1 ¨ what else would she buy 2¨

All these nexuses and other similar to them, state the projected clause as an independent clause, but as a reported, not a quoted one. The direct speech alternative would have been: “Melancholy is the dominant note of my temperament,” ″1 ¨ he said 2 ¨ She wondered 1¨ “Will I have to stand in the Drumcondra tram!” ″2 ¨ “He is a graceful image of myself,” “and a just one” ″1¨ Then she asked 1¨ “what else would I buy?” ″2¨

Conversely, their indirect speech variant would have looked like this:

SYSTEMIC FUNCTIONAL GRAMMAR: A READER He said α¨ that melancholy is the dominant note of his temperament ″β¨ She pondered α ¨ whether she would have to stand in the Drumcondra tram He declared α ¨ that that one is a graceful image of him and a just one ″β¨ Then she asked herself α ¨ what else would she buy? ″β¨

″β

149

¨

Frequently, the projected clause of the free indirect speech type lacks its projecting counterpart. In such circumstances, the context provides the required clues indicating the clause to be free indirect speech. Consider the exclamative clause in the following paragraph. I crammed my mouth with stirabout for fear I may give utterance to my anger. Tiresome old rednosed imbecile! It was late when I fell asleep.

4.5.3.3. Embedded projected clauses Besides being able to enter in tactic clause nexuses, projected clauses can be embedded, that is they function as qualifiers within a nominal group. For such a construction to qualify for the embedded status, it has to be agnate with a clause nexus projection. The following clause has an expanded nominal group (underlined) that includes a dependent clause introduced by that. He watched her leave the room in the hope Æthat she would look back at him Ø,║ but he was disappointed. ¨|

The agnate form of this embedded construction would be: ... he hoped ¨ that she would look back at him ¨ ...

At times, embedded clauses can refer anaphorically to material previously encountered, by means of this, that, it, or other text references. I kept her brown figure always in my eye ¨ and, , I quickened my pace ¨ and passed her. ¨ This happened morning after morning. ¨

Nouns that project can produce propositions or proposals, each giving rise to the options of locutions and ideas. Propositions include affirmative and interrogative formulations. These are realized by means of conjunctions that, and if/whether, respectively. Similarly, proposals can cover not only commanding but also offering formulations. The conjunctions to or of are instrumental in projecting proposals. but the thought Æthat a poetic moment had touched him Ø took life within him the few people in the hall began to express their desire Æto be entertained.Ø

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

150

In addition to clauses projected by verbal or mental Process, or by nouns of projection, there is another group of embedded clauses, those projected by fact nouns. There are four classes of such nouns in English: cases, chances, proofs, and needs. The first three classes refer to propositions, whereas the latter refers to proposals. Few more examples will be appropriate here (IFG, 470-71, 478). stating

locutions

propositions

ideas

questioning

locutions

offering

ideas locutions

proposals

ideas commanding

locutions ideas

projection nouns report, news, rumour, claim, assertion thought, belief, knowledge, suspicion, sense question, query, inquiry, argument, dispute doubt, question offer, suggestion, proposal, threat, promise intention, desire, hope, inclination, decision order, command, instruction, demand wish, desire, hope, fear

fact nouns ‘cases’: fact, case, rule, lesson ‘chances’: chance, possibility, likelihood, probability ‘proofs’: proof, indication, evidence, demonstration ‘cases’: issue, problem, conundrum ‘chances’: uncertainty

‘needs’: requirement, need, rule, obligation, necessity, duty, expectation

cases The fact Æthat Lear never even alluded to that at the end Ø is a sign Æthat he didn’t learn very much through the course of the play Ø. chances there’s a good chance Æthat it will settle down enough. ..Ø proofs there is evidence Æthat the ozone hole is enlarging and spreadingØ needs You’ve said that one of your editorial rules is Ænot to publish your buddiesØ Table 4.6: Nouns of projection and nouns of fact in English (IFG, 469, Table 7(25))

SYSTEMIC FUNCTIONAL GRAMMAR: A READER

151

Although, impersonal attributive and identifying Processes prefer embedded projection, there are others that can use it, such as mental, or verbal (IFG, 470-78). [impersonal attributive – proposition] As a matter of fact it is of no importance Æwhether Benedetto Marcello found it or made itØ. (Joyce, Ulysses) [personal attributive – proposition] "I am afraid Æyou didn't enjoy yourself at all,Ø" ¨ said Mary Jane hopelessly. ¨| [identifying – proposition] "The fact is," > Æ"I have just arranged to go―"ب| [existential – proposition] There is a word throstle Æthat expresses thatØ. (Joyce, Ulysses) [mental – proposition] More than he resented the fact Æthat he had been victimisedØ ¨ he resented ¨ such low playing of the game. ¨| [verbal – proposition] "You see, ¨ we may as well all admit Æwe're a nice collection of scoundrels, one and allØ.¨| [identifying – proposal] ... it was necessary for him Æto move his body from the hipsØ.

4.6. S UMMARY According to SFG, clause constituents have meaning at three different levels, their function indexing the specific role they play at each level. The same proposition can be perceived as text, exchange and experience. The textual meaning of a clause is indexed by its information structure and its thematic structure. Whereas the former is incorporated within the metafunction of New, the element selected to carry the information focus, the latter is incorporated within the metafunction of Theme, the element the Speaker selects to start a clause. The interpersonal meaning of a clause is indexed by its mood structure and is incorporated within the metafunction of Subject, the most salient element of the verbal transaction between speaker and addressee.

152

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

The experiential meaning of a clause is indexed by its transitivity/ergativity structure and is incorporated within the metafunction of Actor/Agent that identifies the active participant in the Process of an ongoing experience. Supplementary to these three strands of meaning, the logical meaning is indexed by the clause complex structure. Interdependency and logicosemantic types interact, complexing clauses.

5 A SYSTEMIC FUNCTIONAL GRAMMAR OF S TANDARD HEBREW Before experiencing the shift towards becoming a Subject-initial language, SH used to be for centuries a Verb-initial language (Givón 1974). In the case of many other Semitic or non-Semitic Verb-initial languages, the Verb occupies the most salient position of the clause. Therefore, the constituent that occupies the initial slot, when rightfully it belongs to the verb, is considered marked, and the Process is known as defamiliarization. Compared to POETRY, PROSE is expected to behave in a more regular way since the intricacies of language are not complicated further by aesthetics, as happens in POETRY. After setting out the theoretical basis of SFG in the previous chapter, one can embark on applying these principles on SH and produce a systemic grammar of SH. Upon completing such an exercise, one will be able to engage in an informed discussion of HPy and discern its specificity in relation to its more regular counterpart that one can find in PROSE (cf. Chapter 6). The structure of this chapter follows the three levels of analysis that were detailed previously, referring specifically to the thematic (including information structure), mood and ideational (including transitivity/ergativity and clause complex) structures.

5.1. THEMATIC S TRUCTURE The largest part of this section deals with the Theme+Rheme system of network as it appears to be used in SH. At the outset, a few comments are in order in relation to the Given+New system. Phonologically, the atempt to approach the SH texts without any reference to the Masoretic accents is a very difficult enterprise, based entirely on reconstructions and comparison with cognate languages, of which most are either in a poor condition (Ancient Semitic languages) or too late (Arabic). If the intonation reconstructed by the Masoretes is taken in consideration, which is the more sensible alternative, one has to make sure that the Masoretic accents indeed 153

154

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

carry a clear phonological value. Had one admitted with Revell (1980, 1981), Dresher (1990) and Greenstein (1992) this quality of the Masoretic accents, one would have to embark on a debatable terrain with more expertise in the field than is currently available to us, for results that have less relevance and are not justified by the limits of this dissertation. Even so, one cannot escape entirely the prosodic observation of the text, because the Given+New message line is realized by tone, the greatest pitch movement in the tone group realizing the New element. The study of the information structure may prove more relevant for our discussion of the SH clause, and more useful for chunks of texts than for individual clauses. Functionally inspired information structure analyses have already been conducted by many scholars on SH narrative (Heimerdinger 1999; Shimasaki 2002; Floor 2004) and Aramaic narrative (Buth 1987). The information structure theory was also applied to HPy, such as Isaiah 40-55 (Rosenbaum 1997), Isaiah 40-66, Psalms, Job 3-14, Proverbs 1-9, Song of Songs, and Numbers 23-24 (Lunn 2004). Of particular value here is Lunn’s thesis, who has indicated conclusively the priority of line A over line B in any information structure analysis, and the dissimilarity between marked information focus (MKD) and mere defamiliarized information focus (DEF) as opposed to regular, canonical information focus (CAN) (cf. supra 2.2.4.). Before dealing in depth with the thematic structure, a few samples can illuminate in passing the case for information structure in SH from a SFG perspective. Clauses that open new texts contain New information exclusively. Example (1) introduces the hero to the reader, zooming in after matters of general historical context were presented in the previous verses (Judg 10.6-18). (1) Judg 11.1 w=y]pT> h~G]lu*d' h*y> G]BBor j~y]l 1║ Jephta the Gileadite was a mighty warrior - - - - - - - - - - - - - - - - - - - - New

Fortunately, cases where the first clause of a given narrative in SH is not entirely recoverable are rare. Regularly, each information unit coextensive with a clause possesses a Given element and a New element, the latter towards the end of the information unit.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

155

(2) Judg 11.1 w=hWa B#n -a]V> zon> 2║ w~yol#d G]lu*d a#t-y]pT> 3║│ and he [was] the son of a prostitute Given - - - - - - - New

fathered Given

Gilead

Jephta New

In other clauses, the author left the known material to the end of the clause, in order to give emphasis to the New element (focus). Although the information unit in #3 comes at the very beginning of the HB, ‘God’, ‘heaven’ and ‘earth’ are retrievable from the cultural context of the audience. Only ‘in the beginning’ and ‘created’ are unexpected and irretrievable. (3) Gen 1.1 B=r}v't

B*r*a a$Oh'm a}t h~V*m~y]m w=a}t h*a*r#x

in the beginning created God New (focus)

the heaven Given

and the earth

Since anaphoric elements and deictic elements do not carry information focus, they are part of the Given element. (4) Gen 40.20 w~yh' B~yom h~V=l'v' yom h%L#d#t a#t-P~ru{h it happened on the third day Given

[was] the birthday of Pharaoh New

The study of information units is more valuable when texts are analysed as larger units in order to ease the referential Process a reader needs to do. Even though SFG admits the organization of message at the information structure level, this is not particularly its strength. In our current study, we will refer to the Given + New system only where whole poems are analysed. To summarise, the prosodical marking is employed at the level of information structure on the foot carrying the tonic prominence, regularly placed at the end of a clause. The salient information of a clause is also marked positionally at the level of thematic structure, by placing it in the most salient slot, which in Hebrew as well as in English is at the head of the clause. In SH, unmarked Theme is negotiated in relation to the system of Mood and displays several peculiarities. For a definition of mood categories employed in the second column, which are to be distinguished from the

156

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

traditional grammatical categories used in the first column, see 4.3.1. and 4.3.2.). The follwing table shows that the Finite is expected to occupy the first slot of a clause only in declarative clauses with a finite predicate, in clauses with infinitives as non-finite predicates, or in volitive clauses. The Predicator appears at the head of declarative clauses of existence. Declarative clauses with participles as non-finite predicates and declarative clauses of identification prefer the Subject in front of the Predicator. Classification clauses give precedence to the Complement. Interrogative and exclamative clauses, whilst preserving the order of their respective declarative counterparts, have at the head of the clause the interrogative/exclamative element.1 mood type declarative: finite predicate declarative: non-finite predicate: inf. abs./cstr. declarative: non-finite predicate: ptcp declarative: verbless: identification declarative: verbless: classification declarative: verbless: existence declarative: exclamative interrogative: finite: polar interrogative: non-finite: inf. abs./cstr.: polar interrogative: non-finite: ptcp: polar interrogative: verbless: identification: polar interrogative: verbless: classification: polar interrogative: verbless: existence: polar interrogative: WHvolitive

unmarked Theme Finite^Predicator Finite^Predicator Subject Subject Complement Predicator WH- element interr. element^Finite interr. element^Finite interr. element^Subject interr. element^Subject interr. element^Compl interr. element^Pred WH- element Predicator

Table 5.1: The unmarked choice of mood in SH

5.1.1. Theme in declarative clauses Although SH uses some verbs as auxiliaries, it seems that conflating Finite with Predicator functions is the rule in most cases of finite verbs in a declarative clause (#5). Constituents that are in focus in our analysis will be 1 Such complexity of thematic unmarkedness seems to have put off some scholars such as Floor (2004, 199), who refused to engage with it although he acknowledged it as such.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

157

doubly underlined in the transliterated script. An atypical Theme occupying the most salient slot of a clause is considered marked. Subject can function as Marked Theme of a SH declarative:finite clause (#6). (5) Gen 37.1 w~yy}v#b y~u&q)b B+A#r#x m+gWr? settled Theme

Jacob Rheme

A*b'w B+A#r#x

K+n*u~n

in the land of the sojourn of his father, in the land of Canaan

(6) Gen 37.3 w=y]car*a}l a *h~b a#t-yos#p m]KK"l - B*n*yw but Israel loved Joseph more than all his sons Marked Theme Rheme

The exclamative clause is a subset of the declarative clause that has the exclamative material as part of the thematic structure (cf. Table 5.1). As in English, in SH the auxiliary particle (interrogative, WH-, or h]N#h particles) stays at the head of exclamative clauses. Contrary to the expected structure in English, however, the exclamative clause in SH displays a similar thematic structure to that of the interrogative clause, with the verb (Finite/Predicator) following the interrogative/WH- particle. (7) 1 Sam 19.24 h&G~m v*aWl B~N=b'a]m Even Saul [is] among the prophets! Theme Rheme

Most frequently, other interrogative particles are preferred at the head of exclamative clauses. Theme includes both the Finite verbal operator and the auxiliary element. (8) Gen 28.17 M~h-nor*a h~M*qom h~ZZ#h how awesome [is] this place! Theme Rheme (9) Gen 16.6 h]N#h v]pj*t}k B=y*d}k behold your servant [is] in your hand! Theme Rheme

Although it is possible, the literal rendering of #10 does not make any sense if it is in the interrogative mood. If this is the expression of an authoritative person, the exclamative mood gives more credit to the

158

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

initiator. In this case, the WH- particle is conflated with the Subject function (cf. also 2 Sam 23.15). (10) 2 Sam 15.4 m' - y=c]m}n' vop}f B*a*r#x Oh, [if only somebody] would make me judge in the land! Theme Rheme

Declarative clauses with non-finite Predicators imply either an infinitive absolute, an infinitive construct or a participle (IBHS, 72, 623-8). Clauses with a participle as Predicator prefer a Subject^Predicator word order. When other material occupies the salient slot, then that becomes a marked Theme (#12). (11) 1 Sam 9.11 h}M> u)l'm B=m~u*l}h h*u'r they were going up the ascent of the city Theme Rheme (12) Gen 3.5 K' y)d}^U a$Oh'm because knows God Marked Theme Rheme

Declarative clauses with an infinitive as Predicator do not follow the preference of participles to follow the Subject. When infinitive constructs and infinitive absolutes function as Predicators, they play a similar role to the one performed by the finite verbs. If infinitive constructs do not have an explicit Subject, a pronominal suffix that functions as Subject accompanies them (cf. #13).1 Declarative clauses having their Predicator realized by an infinitive construct with any preposition except l*m#d can serve as subordinate clauses for an independent clause that follows (cf. IBHS, 604-5). When a pronominal suffix is attached to the verbal form, the former functions as Subject for the Predicate that is the latter (#14).

Clauses with an infinitive absolute with a conjunction following a clause with a finite verb seems to be preferred in LH, in books like Jeremiah, Zechariah, Esther and Nehemiah (Rubinstein 1952). Rubinstein quotes the following passages: Gen 41.43, Exod 8.11, Josh 9.20, Judg 7.19, 1 Sam 2.27-28, 1 Kgs 9.25, Is 8.6, 37.19, Jer 3.1, 7.18, 13.16 (Qere), 14.5, 19.13, 22.14, 32.33, 44, 36.23, 37.21, Ezek 23.36, 47, Amos 4.5, Hag 1.6, Zech 3.4, 7.5, 12.10, Eccl 8.9, 9.11, Esth 2.3, 3.13, 6.9, 8.8, 9.6, 12, 1618, Dan 9.5, 11, Neh 7.3, 8. 8, 9.8, 13, 1 Chr 5.20, 16.36, 2 Chr 28.19. 1

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

159

(13) Exod 8.11 w=h~kB}d #t-l]Bo and [he] hardened his heart Theme Rheme (14) 1 Sam 11.6 B=v*m=uW a#t-h~D=b*r'm h*a}L#h while hearing those words Theme Rheme

One should not forget the possibility that some existence particles can function as Predicator, in which case the preferred word order is Predicator^Subject. Although it is not very common elsewhere, the Subject is formed by the preposition l*m#d with a pronominal suffix or a noun. (15) Gen 33.9 y\v l' r*b There is to me plenty Theme Rheme

Many circumstances wherein other material fronts the existence particle, like the ones in the following text, differ from the numerous more standard occurrences. (16) Judg 19.19 w=g^m-T\b\n G^m-m]sPoa y}v l^j&mor?nW ¨ Both straw and fodder there is to our donkeys Marked Theme Rheme w=g^m l\j\m w*y^y]n y\v - l' w=l^a&m*t\k* w=l^N^u^r [...]¨ also bread and wine there is to me and to your maid and the youth Marked Theme Rheme a?n m^jsor K"l-D*b*r ¨ there is no lack of anything Theme Rheme

Declarative:verbless clauses differ in terms of their respective word order due to the relation between Subject and Predicator. In such a case, an adjective, adverb, a noun, pronoun, or adverbial prepositional phrase can carry the Predicator role (IBHS, 72-3). Due to scarcity of examples to confirm it as a rule, adverbs and adverbial prepositional phrases should not be considered as standards for the realization of Predicator but rather as exceptions (Marked Theme).

160

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

When the Predicator identifies the Subject, one speaks of an identification clause and its expected word order is Subject^Complement (cf. IBHS, 130-2).1 (17) Num 13.16 a}L#h v=mot h*a&n*v'm these [are] the names of the men Theme Rheme (18) Lev 18.15 a}v#t B]nk* h]wa [is] the wife of your son she Marked Theme Rheme

A specific application of this type of clause is the narrative disclosure of names. Following the standard formula for asking for someone’s name, the name itself corresponds to the WH- element and functions as Complement, whereas ‘name’ functions as Subject.2 Thus, in the affirmative mood, the expected order would be Subject^Complement.3 (19) Gen 38.1 Wv=mo j'r> and his name [was] Hira Theme Rheme (20) Exod 34.14 K' yhwh q~N*a v=mo because [is] YHWH-Jealous his name Marked Theme Rheme

When the other participant classifies the Subject, i.e. refers to a general class of which the subject is a member, one speaks of a classification clause

Andersen (1970, 32) confirms a Subject first position for identification clauses, those clauses wherein ‘the predicate has total semantic overlap with the subject; that is, each has exactly the same referent’ and offers many samples following a similar basic such as pattern #1 (‘I am the Lord’ - Exod 6.2). 2 Cf. Gen 32.28; Exod 3.13; Prov 30.4. 3 Giving/changing names follows a different paradigm (√qWm/√c'm/√s*b~b (a#t) v=mo/v=m*h X): √qWm a#t-v=mo (e.g. Judg 13.24; 1 Kgs 7.21), √qWm a#tv=m*h (e.g. Gen 30.21), √c'm a#t-v=mo (e.g. Judg 8.31; Neh 9.7), √s*b~b a#tv=mo (e.g. 2 Kgs 23.34; 24.17). Cf. also Gen 17.15; Exod 15.3; Job 1.1. Notice that similar constructions with demonstrative pronouns follow the same order (e.g. Gen 2.19; Exod 3.15). 1

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

161

and the subsequent word order is Complement^Subject (#21) or Attribute^Subject (#22) (cf. IBHS, 132-4).1 (21) Exod 29.14 j~F*at hWa [is] a sin offering it Theme Rheme (22) Gen 25.30 K' u*y}p a*n{k' [am] hungry I Theme Rheme

Adjectives can function as Predicators in a declarative:verbless clause, producing an assertion about the subject of the clause (IBHS, 260-1), in which case the word order is Predicator:Adjective^Subject.2 (23) Gen 6.5 r~BB> r*u~t h*a*d*m B*a*r#x [was] great the evil of the mankind on the earth Theme Rheme (24) Num 13.28 w=h#u*r'm B=x%rot G=d)Ot m=u)d the fortified cities [are] great exceedingly Marked Theme Rheme

Also, a demonstrative adjective can fulfil the role of Predicator. IBHS fails to distinguish the variation Subject^Predicator, which is considered by its authors only as a possible alternative. (25) Gen 5.1 z#h s}p#r Tol=d)t a*d*m [is] this the book of the history of Adam Theme Rheme

Declarative:verbless:existence (possession) clauses follow the typology of a declarative:finite clause but the Predicator role is played by vyE / !yae. The Possessor is expressed by means of l*m#d. Andersen (1970, 32) confirms this sequence for classification clauses, those clauses wherein ‘the predicate has partial semantic overlap with the subject; that is, it refers to the general class of which the subject is a member’ and offers many samples following a similar basic pattern such as #94 (‘he is unclean’ - Lev 13.36). 2 Other examples of Marked Theme in a declarative:verbless clause are Genesis 18.27 and Exodus 17.12. 1

162

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(26) 2 Kgs 4.2 m~Y#v l*k B~BB*y]t […] what is there to you in the house? Theme Rheme a?n l=v]pj*t=k* k"l B~BB~y]t there isn’t to your maid anything in the house Theme Rheme

Out of the two previously quoted clauses, the first one is an interrogative: existence: WH- clause. Its Theme is realized by the WHelement conflated with the Subject. More examples of this type will be discussed below. Thus, its Predicator is pushed backwards as part of the Rheme. A proper reply to the question follows, revealing that the Possessor cannot boast any possessions. This time the Predicator stays as Theme. In a declarative clause, the Mood constituent of the clause stays typically at the head of the clause. The Predicator precedes the Subject in the case of clauses with a finite verb, a non-finite:infinitive verb, or verbless:existence clauses. Clauses with a non-finite: participle verb, verbless:identification and exclamative clauses prefer to start with the Subject. The Residue constituent varies in length according to the number of Complements or circumstantial Adjuncts. 5.1.2. Theme in interrogative clauses SH implies the WH- element when asking questions, but the polar interrogation is realized by the deictic h], placed in front of the first word of the clause in which it occurs, i.e. the Finite verb in the case of a declarative:Finite clause. (27) Gen 18.14 h&y]PP*l}a m}yhwh D*b~r ¿is difficult for the Lord anything? Theme Rheme

Nevertheless, it is possible that other material (e.g. Subject, Complement, Circumstance) occurs between the interrogative particle and the verb. In such a case, the option is Marked. (28) Gen 18.25 h&v)p}t K"l-h*a*r#x l)a y~u&c#h m]vP*f ¿the judge of all the earth not do judgement? Marked Theme Rheme

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

163

(29) Gen 3.11 h&m]n-h~u}x a&v#r x]ww'T'k* l+b]lT' a&k*l-m]mm#NW a*K*l=T* ¿from the tree which [I] commanded you not to eat from it [you] ate Marked Theme [[ embedded clause ]] Rheme

Since the unmarked realization of a declarative:non-Finite:participle clause (Subject^Predicator) is closely followed in the interrogative mood, its modification signifies markedness. (30) 1 Kgs 18.17 h^a^TT> z#h u)k}r y]cr*a}l ¿Are you the one disturbing Israel? Theme Rheme (31) Gen 4.9 h&v)m}r *j' a~n)K' ¿[am] the keeper of my brother I? Marked Theme Rheme

Unlike English, Hebrew does not use question tags, but the equivalent is found. When al{h] appears at the beginning of the interrogative clause, it functions like a question tag. A positive answer is the expected reply to such questions (Brongers 1981, 178-80). The expected word order follows the rule of the respective mode the clause has after removing the interrogative particle. For example, in a clause with a Finite verb, the Predicator will come next to the interrogative particle (#32). (32) 2 Sam 4.11 h&Oa a&b~Q}v a#t-D*mo m]Y#dk#m ¿not [I] demand his blood from your hand? Theme Rheme

In a clause with a non-Finite verb (#33), the Subject will appear between the interrogative particle and the Predicator. In a verbless identification clause, the Subject will follow the interrogative particle (#34), but the Predicator is expected instead in the case of a verbless classification clause. Any departure from the preceding rules indicates markedness (#31). (33) Deut 32.34 h&Oa - hWa K*m%s u]M*d’ ¿not he stored up with me? Theme Rheme

164

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(34) Exod 4.11 h&loa a*n)k' yhwh ¿not I [am] the Lord? Theme Rheme (35) Gen 19.20 h&loa m]xu*r h]wa ¿[is] not a small thing she? Marked Theme Rheme

An analysis of a large corpus of HB samples indicates that in interrogative clauses the Subject manifests a strong tendency to follow immediately after the interrogative particle. Nevertheless, many declarative: finite clauses with a non-Marked Theme do not have an explicit Subject.1 However, very few declarative:finite clauses with a Marked Theme lack an explicit Subject.2 Therefore, the tendency of interrogative particles to attract the Subject to their proximity should be interpreted as a preference of such constructions to focus on the doer. Hence, the derived function of this clause as an irrefutable truth or an indirect accusation. One of the commonest quote-formulae is rendered as an interrogative clause with a declarative: non-finite: participle clause (only once in Joshua 10.13, but eighteen times in Kings). (36) 2 Kgs 21.17 h&loa- h}m K=tWb'm u~l-s}p# D]br? H~Y*m'm ¿not they [are] written in the book of Chronicles Theme Rheme

WH- interrogative questions are realized by means of interrogative pronouns or adverbs: the animate pronoun ymi (‘who?’) and its forms, or the inanimate pronoun hm' (‘what?’) and its forms, as well as others: yt'm ' (‘when?’), hm'l' (‘why?’), hz< hm'l' (‘why then?’), [:Wdm' (‘wherefore, for what reason?’), hm'-d[; (‘how long?’), yae / hYEa; / hk'yae (‘where?’), hpoyae / hn'a ' (‘where to?’), !yIa; / !a' / !yIa'me (‘where from?’), %yae / hk'yae (‘how?’), yt'm' / yt'm'd[; / yt'm'l. (‘until when?’) (Joüon 1996, I: 332-3). 1 One can find clauses with a Finite verb and a Non-marked Theme in the following texts: Gen 29.25; Num 12.14; 22.37; 23.26; Judg 6.14; Ruth 3.1; 1 Kgs 18.13; 2 Kgs 19.25. 2 Clauses with a Finite verb and marked Theme can be traced in the following

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

165

(37) Gen 20.9 m~h - u*c't> L*nW ¿what [you]did to us? Theme Rheme

5.1.3. Theme in volitive clauses As opposed to English, which has only one distinct grammatical form for the imperative (the second-person volitional), SH has a set of volitional forms, the cohortative, the imperative, and the jussive. Since they are rather independent forms that can have different volitive nuances, and share usages in the declarative mood realm too, they are considered rather a functional class than a mood (IBHS, 564-5). Therefore, it is necessary to distinguish between a volitive form and a volitive sense. Most verbal roots do not display any morphological distinction between jussive and non-jussive forms. Although in earlier stages of Northwest Semitic jussive was used with all three persons, in SH it is restricted to the third person and second person negative (IBHS, 567). Since the uses of the Jussive follow the status relations of the speaker and addressee, they will be reviewed later (see infra ‘clause as exchange’). Suffice it to say that the jussives are much involved in the poetic texts conveying functions such as benedictions, exhortations, or invitations. From a SFG perspective, the jussive too is expected at the head of the clause. Any unexpected variations are susceptible of markedness. (38) Num 17.3 w=y]hyW l=aot l]bn? y]cr*a}l let them be as signs to the children of Israel Theme Rheme (39) Gen 14.24 h}m y]qjW j#lq*m they let take their share MTheme Rheme

SH employs the negative adverb la; with the jussive verbal forms in order to produce a negative command. Prohibition can be realized by means of a non-perfective form with al{ as well.

texts: Num 12.2; 23.12; 24.12; Judg 11.24.

166

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(40) Gen 22.12 a~l-t]vl~j y*d=k* a#l-h~N~h~r do not lay your hand on the boy! Theme Rheme (41) Gen 22.12 Oa-t~y&n#h b=r}u&k* u}d v*q#r do not do give against your neighbour false witness Theme Rheme

The imperative prefers the most salient slot of the clause too. Imperative grammatical forms are always used positively, i.e. to express a direct command (#38). If any material precedes the imperative, markedness explains its unfamiliar usage. IBHS (p. 572) does not see that the first clause in Gen 24.60 has an unfamiliar usage of the vocative preceding a pronoun, thus functioning as a Marked Theme (#43). Both jussives and cohortatives can take the place of an imperative in a volitive sequence. Even y]qf)l’s can appear in a volitive sequence, but they will be charged with volitive tenor (IBHS, 577-8). (42) Deut 5.1 v=m~u y]cr*a}l a#t-h~j%Q'm w=a#t-h~MvP*f'm listen, o Israel, [to] the decrees and the laws Theme Rheme (43) Gen 24.60 a&j)t}nW a~TT= h&y' l=a~lp? r=b*b> ¨ our sister, you may be thousands of myriads! MTheme Rheme w=y'r~v z~ru}k a}t v~u~r c)n=a*yw ¨ and may inherit your seed the gate of your enemy! Theme Rheme

As the volitive for the first person, SH prefers the cohortative, a grammatical form regularly formed by attaching a h '- suffix. Such volitives may be accompanied by the precative particle aN'- (IBHS, 578-9). Cohortatives prefer to front the clause in which they are present with the exception of a negated volitive when la; fronts them.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

167

(44) 2 Sam 24.14 n]PP=l>-N*a b=y~d - yhwh ¨ K' r*BB'm r~j&mo [K] ¨ let [us] fall, please, in the hand of the Lord because [is]great his compassion Theme Rheme Theme Rheme Wb=y~d a*d*m a~l - a#PP)l> ¨ and in the hand of man may not [I] fall Marked Theme Rheme

SH can even use infinitive absolutes to carry a command in which the request is by no means negotiable, especially divine or divinely inspired commands. (45) Exod 20.8 z*kor a#t-yom h~V~BB*t l=q~DD=vo Remember the day of rest to sanctify it Theme Rheme

One concludes by noting that the familiar usage of volitive locates them in the most salient slot of the clause. Unfamiliar formulations of volitive clauses should be validated only after considering the quality of the fronted material. 5.1.4. Textual and interpersonal Them es Textual Themes will be discussed first with their varieties: continuatives, conjunctions and conjunctive Adjuncts. In SH, some of the well known continuatives are al{, la;, hNEhi/hNEhiw>, hT'[;w>, !kel',1 w> (as now in waw-X-q*f~l constructions) and Hh'a] (or similar interjections). hT'[;w> in particular, introduces a shift in the argumentative tack, without changing the subject or reference.2 The continuative character of waw is obvious in the case of waw consecutive but the form is very stable in SH (see infra). (46) Gen 3.22 w=u~T> P#n - y]vl~j y*do and now lest [he] will stretch his hand Cont Conj Textual Rheme Theme

1 2

Cf. Jongeling (1981) and Gross (1994, 209-11). Cf. IBHS (p. 667) and Gross (1994, 211-2).

168

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

The interjection hNehi can function as a discourse signaller, highlighting the presence of a new character (Gen 24.15), a statement that is contrary to the listener’s expectation (Gen 20.16), or a procedure with a high degree of certainty (Lev 13.5) (Slager 1989, 54, 57, 66). Slager maintains that hNEhi prefers the direct quotation, whereas hNEhiw> prefers narrative texts. In the following example, a familiar declarative:non-finite:participle clause structure places hNEhiw> at the front, a fact that does not change anything in the Thematic structure. Nevertheless, the continuative particle succeeds in signalling to the reader the appearance of a new character on the stage, namely Rebecca. The technique is similar to zooming, as known in filmmaking. (47) Gen 24.15 w=h]N}h r]bq> and look Rebekah Cont Subject Textual Topical Theme

y)x}at came Rheme

Although they were long regarded as not differing in use (Labuschagne 1973), the particles hNehi and !he display distinctive characteristics. Unlike hNeehi, the particle !hee records distinctive usage (Garr 2004, 343). Syntactically, !he is restricted to clause-initial position, but hNehi can appear within the clause as well. Pragmatically, !he is never used to call attention to information that is contrary to the listener’s expectation. Although they are both used as deictics, !he is not locative or presentative as hNehi is. Finally, in terms of the quality of the information provided, !he does not introduce new information as hNehi does, but reiterates old (known) information. Brongers (1981, 181) proposes that, in several contexts, the meaning of al{h] should follow the translation LXX suggests, being read as h]N}h, therefore calling the attention to the specified Subject (Gen 13.9; Judg 4.14; 1 Sam 10.1; 1 Sam 20.37). (48) Gen 13.9 h&Oa k"l-h*a*r#x l=p*n#yk* ¿[is]not all the earth before you? Cont Subject Textual Topical Rheme Theme

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

169

(49) Gen 20.15 h]N}h a~rx' l=p*n#yk* Look my land [is] before you Cont Subj Textual Topical Rheme Theme

Some of the most common paratactic conjunctions in SH are: w> ‘and, but’, ~aiw> ‘and if’, lb'a] ‘but’ (more frequent in LH), ~G' ‘moreover’, ~ai yKi ’because, when, if, although,1 then,2 but3’, al{h] ‘rather’4. The declarative clause in #50 starts by lb'a], whose Predicator is fronted by a Subject, hence Marked Theme. The conjunction connects this clause with the previous one that, in this case, contradicts it. (50) 1 Kgs 1.43 a&b*l a&d{n?nW h~M#l#k-D*w]d but our Lord, king David Cont Subject Marked Theme

h]ml'k a#t-v=Om{h enthroned Solomon Fin/Pred Complement Rheme

Frequently met hypotactic conjunctions are: ~ai ‘if’, (van Leeuwen 1973), aloh] ‘¿not?’ (Brongers 1981, 185-7),5 yK ‘that’ (Schoors 1981),6 rv,a] ![;y; ‘because that’ (Mulder 1973), ![;m;l. ‘in order that’ (Brongers 1973), ~r,j,B., qr; ‘but, only’, (Jongeling 1973), !kea' ‘therefore’. Therefore, although in most cases a subordinate relative clause is introduced by the particle rv,a], the prepositional phrases constructed with rv,a] function conjunctively – rv,a]K; ‘as, while’, rv,a]me, rv,a]B;, rv,a]l; ‘wherever’.7

1 These are the primary and more frequent meanings of the conjunction in its function of introducing adverbial clauses (causal, temporal, conditional, and concessive, respectively). Cf. Schoors (1981, 264-73). 2 At the head of the apodosis of conditional sentences (Schoors 1981, 250). 3 After a negative clause (Schoors 1981, 251). 4 Brongers (1981, 182) quotes as illustration Jer 26.19, Ezek 18.23, 25, 29. Also Gross (1994, 207-9). 5 Examples quoted here suggest that the clause introduced by al{h] functions as an apodosis: Exod 33.16; 1 Sam 29.4. 6 At the head of an object clause (Schoors 1981, 254-6) or a subject clause (Schoors 1981, 259-60). 7 In LH, the relative pronoun can also introduce a vocative (Cant 1.7 cf. IBHS, 336).

170

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Example (51) displays two clauses, of which the second is dependent on the first. In both cases their thematic structure complies with the familiarity requirements, thus both their Themes are unmarked. Waw is not always a simple coordinative conjunction, but can also be adversative or contrastive. (51) Exod 11.9 Oa - y]vm~u a&l?k}m P~ru)h ¨ l=m~u~n

not listen Top Theme

r=bot

to them

mopt~y B=a#r#x m]xr*y]m ¨

Pharaoh in order that will multiply the miracles in the land of Egypt Textual Topical Rheme Rheme Theme

Some of the well known conjunctive Adjuncts in SH are: rx;a; ‘after’, !ke-yrex]a; ‘afterwards’, hLah xa;; ‘after these’, !ke-l[; ‘therefore’, yK pap,, ‘however’. According to Brongers (1981, 186-7) even al{h] can assume such a function (cf. 1 Sam 9.21; Prov 26.19). (52) Num 13.28 a#p#s K' - u~z h*u*m h~Y)v}b B*a*r#x howbeit [is] strong the people that lives in the land Text Top Rheme Theme

A discussion of the interpersonal Themes will follow. This will include modal Adjuncts, volitives and Finite verbal operators. Even though in English there is a large variety of modal adverbs, SH lacks this diversity. IBHS (pp. 656 ff.) accounts for three classes of adverbs in Hebrew: clausal adverbs, item adverbs, and constituent adverbs. Of interest to us at this level of discussion are only those that refer to the content of the message as a whole not to particular words in a given clause, particularly the emphatic adverbs (IBHS, 662-8). Such examples of SH modal adverbs may be: lb'a ] ‘verily’ (AH), %a; ‘surely’, ~N'xi ‘for nothing’, ~aot.Pi or [t;P, ‘suddenly’, aw>v' ‘vainly’, hn'm.a' ‘truly’, aloh] ‘certainly’,1 yKi ‘surely’2. In a declarative:non1 For this rather unorthodox usage of the particle, Brongers (1981, 182-5) quotes: Gen 44.5; Josh 22.20; Judg 5.30; 1 Sam 21.12; 2 Sam 3.38; 11.10; Is 8.19; Jer. 38.15; Ezek 24.25; 38.14; Amos 5.20; Mic 1.5; Jonah 4.2; Job 22.12. 2 In this respect, Schoors (1981, 243-8) quotes the asseverative function of the conjunction as displayed in the following texts: Gen 18.20; 2 Sam 23.5; Is 1.29, 2.6, 3.1, 7.9, 15.1, 9; 32.12; 45.16; Amos 3.7; Ps 18.29-30; 49.16; 77.12; refrains 102.11; 118.10-12; 120.7; 128.2; Lam 3.22; Job 5.2; 28.1; Prov 30.1; Eccl 7.7. From among the large number of asseverative yKi, proposed by Dahood in Psalms, Schoors (1981, 245-6) retains only the following: 68.36; 71.23; 76.11; 85.9; 90.7; 104.13; 116.16;

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

171

finite:ptcp such as the one in #53, the f-structure is familiar, therefore Theme is unmarked. (53) Gen 17.19 a&b*l c*r> a]vT=k* but Sara your wife Theme

y)l#d#t l=k* B}n will bear you a son Rheme

In SH, the use of Vocatives is very flexible. Vocatives can appear in front of the Predicator, after it or at the very end of the clause. Whenever it fronts the Predicator, the Vocative stays as a part of the interpersonal Theme, therefore not being charged with thematic markedness. For example, in the volitive clauses below, a Vocative appears in front of the Predicator (#54),1 or after it (#55).2 (54) Gen 27.8 w=u~TT> b=n' v=m~u B=q{l' and now, my son, listen to my voice! Theme Rheme (55) Judg 9.7 v]muW a}l~y B~u&l? v#k#m ¨ listen to me citizens of Shekhem Theme Rheme

In interrogative clauses, Vocatives typically follow the interrogative particle, sometimes being placed at the far end of the clause (#56),3 but occassionally they can front the Predicator (#57).4

135.5; 139.4; 141.5. 1 Similar occurrences are found in Gen 19.2; 27.43; Josh 7.19; 10.12; 16.28a; 19.23; Judg 5.10; 16.28; Ruth 3.11; 1 Sam 14.41; 24.12; 2 Sam 1.24; 7.25; 1 Kgs 8.25, 26; 18.16, 36; 2 Kgs 19.19. 2 The following examples are relevant here: Gen 23.6; 24.18; 27.26, 31, 34, 38; Judg 5.3 (twice), 12 (four times) 21, 31; 9.7; 16.28; 1 Sam 3.9; 4.9; 2 Sam 7.18; 14.4; 16.7; 1 Kgs 12.16b; 2 Kgs 2.23 (twice); 19.16 (twice). Concerning the use of l*m#d to mark the vocative, there is no decisive evidence that it may have been used in SH, although its presence is accepted in Ugaritic (Pardee 1975, 371; 1976, 326). For a detailed discussion on this matter, see 5.2.2. 3 Other occurrences are found in Gen 21.17; 27.18, 20, 37; Judg 21.3; 1 Sam 17.55, 58; 24.17; 1 Kgs 1.13; 19.9; 21.20. 4 As can be found in Gen 20.4; 29.4.

172

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(56) Gen 27.20 m~h - ZZ#h m]h~rT* l]mx)a B=n' what [is] this [you] hasten to find, my son? Theme Rheme (57) Gen 16.8 h*g*r v]pj~T c*r~y a?-m]zz#h b*aT O Hagar servant of Saray, wherefrom [you] come? Theme Rheme

In declarative: verbless clauses, the Vocative can stay either in front of the Predicator (#58),1 or after it (#59).2 (58) 2 Sam 7.28 w=u~TT> a&d{n*y yhwh a~TT> - hWa h*a$Oh'm and now, Sovereign Lord, you [are] the one GOD Theme Rheme (59) Judg 16.9 P=l]vT'm u*l#yk* v]mvon the Philistines [are] upon you, Samson! Theme Rheme

Only in seven instances out of the nineteen examples of vocatives in declarative: finite clauses collected from the corpus, Vocatives follow their respective Predicator,3 and precede it in the rest. 4 This situation is also confirmed by Rosenbaum (1997, 117 ff.), who noted that in 82 per cent of the cases found in Isaiah 40-55, the vocative follows the predication. A vocative can appear alone in a minor clause (#60),5 and it can be reiterated (#61). Minor clauses can also be found where vocatives are associated with the particle of entreaty yBi.6 Also in 2 Sam 14.19; 1 Kgs 1.20. Similar occurrences can be traced in Gen 33.9; Judg 3.19; 6.12; 16.9, 12, 20; 20.7; 2 Sam 1.26; 1 Kgs 12.16; 19.4; 20.4;2 Kgs 3.23; 9.5; 19.19. 3 Consider also the following: Gen 22.8; 1 Sam 14.44; 2 Sam 7.19, 20, 22; 1 Kgs 8.28, 53. 4 Similar occurrences are present in Judg 11.36; 1 Sam 25.26; 2 Sam 7.24, 27, 29; 1 Kgs 1.17, 18, 24; 2 Kgs 1.9, 11, 13; 19.17. 5 Other occurrences may be: Gen 24.12; Exod 3.4; 1 Sam 3.10; 1 Kgs 8.23; 14.30; 2 Kgs 2.12; 4.18; 6.5; 11.14; 13.14; 19.15. 6 cf. Gen 43.20; 44.18; Exod 4.10, 13; Num 12.11; Josh 7.8; Judg 6.13, 15; 1 Sam 1.26; 1 Kgs 3.17, 26. 1 2

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

173

(60) Gen 22.1 w~Y)am#r a}l*yw ¨ a~br*h*m ¨ said to him: “Abraham!” Fin/ Pred/Subj Vocative Topical Rheme Theme minor clause (61) Gen 22.11 a~br*h*m a~br*h*m Abraham! Abraham! minor clause

Finally, other clauses in our database display a vocative preceded by a negation adverb (#62).1 A variant of this type uses the negation adverb specific to the volitive supplemented occasionally by particle aN'- (Gen 19.18; 2 Sam 13.25). (62) 1 Sam 1.15 Oa a&d)n' ¨ a]V> q=v~t-rW~j a*n)k' ¨ no, my lord! a woman of a sorrowful spirit [am] I Theme Theme Rheme

SH can use a non-finite verb as a modal comment Adjunct (e.g., the intensifying infinitive absolute as in #63). It implies a grammatical form built on the same root as the Finite verb that it accompanies prepositively, rendered into English by an intensifying adverb whose nuance is drawn from the context (IBHS, 585). (63) Judg 20.39 a~k n]Ggop n]GG* hWa l=p*n?nW K~M]lj*m> h*r]av)n> surely definitely is smitten he before us like in the battle of first Theme Rheme

Constituents fronting the infinitive absolute or delaying the infinitive absolute function as Marked Theme. (64) 1 Sam 9.6 K)l a&v#r - y=d~BB#r Boa y*boa everything – [he] says – surely happens Marked Theme Rheme

In SH, the realization of modality differs by having the auxiliary conjugated and the finite verb retaining a fixed form (mostly infinitive 1

Check also Gen 23.11; 42.10; 1 Sam 2.24; 2 Kgs 6.12.

174

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

construct). lkoy' (to be able) is one such.1 Gianto also quoted the auxiliary role of some lexical verbs indicating modality: vq;B' ‘to seek’ in Gen 37.15, vr;D', #pex', la;v' ‘to seek, require, request’ in Is 58.2, hk'x' ‘to expect’ in Is. 30.18, hw"q' ‘to expect’ in Is 60.9, h['B' ‘to attempt’ in Is. 21.12, rx;v' ‘to seek, to wish’ in Is 26.9, or @[;v' ‘to desire’ in Job 7.2 (Gianto 1998, 184, n. 2). (65) Gen 37.4 w=l) y*k=lW D~B=ro l=v*Om and not [they] could speak to him peaceably Theme Rheme

Aternatively, the irregular imperative form of the verb $lh, may serve as Finite especially when used alongside other action verbs in a volitive sequence ($le). In such circumstances, it expresses a sense of urgency or immediateness in relation to the first verb of the volitive sequence. An infinitive absolute accompanying a same-root y]qf{l verb may function as Finite too, as shown in the previous example, but when fronted with other constituents the structure is Marked (#66). (66) Exod 5.11 a~T#m l=kW q=jW l*k#m T#b#n you go take for you straw MTheme Rheme

Some epistemic particles %a;, yl'Wa, yKi, hNehi, aAlh; and deontic particles Wl, aleWl, !TeyI ymi, an- can also function as modality particles (Gianto 1998, 184, n.2). It seems that modality in SH relies more on particles than on verbal morphology (Gianto 1998, 188). On the certainty scale, !he introduces utterances that contain irefutable information (Garr 2004, 328) or a reliable hypothesis (pp. 334-9). hNEhi can also introduce certain information but the stress is more on its surprising quality, or its vivid immediacy (IBHS, 675).2 (67) Gen 38.13 h]N@h j*m'k u)l#h t]mn*t> behold your father-in-law goes up to Timnah Theme Rheme

Its usage as modal is admitted by IBHS (507 n.26). At the complex clause level, !he can introduce a subordinate clause, as recently explored by Garr (2004, 341). He concluded that ‘when the !he utterance contains information that is peripheral to another clause, the marked clause is effectively interpreted as subordinate’. 1 2

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

175

According to IBHS (pp. 506-9), SH renders modality by the nonperfective usage of the y]qf{l verbal forms such as capability, permission, possibility, obligation, and desire. The non-perfective usage of capability implies the subject’s capacity to do the action expressed by the root. (68) 2 Sam 22.39 w*a&k~L}m ¨ w*a#mj*x}m¨ w=Oa y+qWmWn ¨ [I] consumed them [I] smote them not [they] will arise Theme/Rheme Theme/Rheme Theme Rheme (69) 1 Sam 13.19 w+j*r*v l)a y]M*x}a B=k)l a#r#x y]cr*a}l a blacksmith could not be found in all the country of Israel MTheme Rheme

The non-perfective of permission implies the speaker’s permission for the subject to do the action expressed by the root. (70) Gen 42.37 a#t-v=n? b*n~y T*m't both my sons [you] may kill Marked Theme Rheme

The non-perfective of possibility implies the subject’s ability to do the action expressed by the root. (71) Josh 1.3 K"l-m*qom a&v#r T]dr)k K~p - r~gl=k#m Bo every place which will tread the sole of your feet on it Marked Theme [[ embedded clause ]]

l*k#m n=t~T'w to you [I] will give Rheme

The non-perfective of deliberation implies the speaker’s or subject’s deliberation as to whether the situation expressed by the root should take place. Regularly it assumes the form of a question. (72) Gen 27.45 l*m> a#vK~l G~m-v=n?k#m yom a#j*d why should [I] bereave both of you same day Theme Rheme

The non-perfective of obligation implies the speaker’s consideration of what should be the subject’s obligation or the subject’s own consideration of what he/she is to do.

176

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(73) Gen 31.39 a*n)K' a&j~F#N> I bore the loss MTheme Rheme

The non-perfective of desire implies the subject’s expression of desire as formulated by the root. (74) 1 Sam 21.10 a]m - a)t*h T]Q~j l=K* if it [you] want to take for you Text Top MTheme Rheme

Without proposing it as yet as a firm rule, from the previous examples it emerges that Marked Theme is a preferred function accompanying the modal usage of the y]qf{l verbal forms. This theory will be tested on the fstructure displayed by the clauses of HPy, in couplets where the QYYQ verbal sequence is present. 5.1.5. Equative Constructions Identification clauses were studied previously in the section on the thematic structure of declarative: verbless: identification clauses. A thematic equative is more than an identification clause. One should note the identification role played by the personal pronoun of the second clause (matrix clause), by means of which the two clauses are connected. Another important construction is hy'h'w> ... rv,a] this time as part of the first clause (fact clause).1 (75) Num 16.7 w=h*y> h*a'v a&v#r - y]bj~r yhwh hWa h~Q*dov will the man which chooses the Lord he will be holy Theme [[ embedded clause ]] Rheme

Identifying a thematic equative has its pitfalls. It can be confussed with an identification clause enlarged by a relative clause qualifying a precedent noun. Take for example the clause in #76.2 Here the relative clause refers Genesis 24.14 and 1 Kings 18.24 (‘the god who answers through fire will be the GOD’) provide new examples. Thematic equative can also be realized by a different mechanism, namely, by delaying the Object of a finite verb in an apparent relative clause. Such a structure appears twice in the prophecy of Malachi (3.19, 22). 2 Occurrences like this can be multiplied: Gen 10.12; 41.28; Num 21.16; Deut 18.22; 1 Kgs 1.45; Ezrah 1.3; Ezek 39.8, etc. 1

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

177

back to one of the constituents of the previous clause, a noun (see it underlined). (76) Exod 16.15 hWa h~L#j#m a&v#r n*t~n yhwh l*k#m l=a"kl> this [is] the bread which gave the Lord to you for eating Subj Predicator Theme Rheme

Further, thematic equatives are not to be confused with clauses in which the Subject is conflated with the Topical Theme (Marked). (77) Gen 3.16 w=hWa y]mv*l-B*k and he will rule over you Text Top MTheme Rheme

Situations of predicated theme can be found even in Hebrew, which offers an alternative explanation for the phenomenon of identifying clauses where the Subject is seen in focus (IBHS, 297-8). (78) Deut 8.18 K' hWa h~N)t}n l=k* K)~j l~u&co j*y]l because he [is] the one who gave you strength to produce wealth Text Top MTheme Rheme

The previous clause can be reduced to a classification clause whose Predicator is realized by an active participle and whose Subject stands as Topical Theme. Although a nominative usage of the participle is tempting here, it does not make sense in translation: he is the giver to you of strength to produce wealth. A different rendering that takes into account both Subject’s Markedness and the verbal function of the participle would be as follows: it is he who gives you strength to produce wealth.1 A more famous example of Predicated Theme is the most quoted monotheistic formula. A tentative translation would be it is Yhwh who is God

1

etc.

Similar constructions can be found in Gen 2.11, 13, 14; Deut 9.3; Ps 100.3;

178

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

in heaven above and on earth below.1 These cases are known in traditional Hebrew grammars as cleft sentences with pleonastic pronoun.2 (79) Deut 4.39 K' yhwh hWa h*a$Oh'm B~V*m~y]m m]M~u~l w=u~l-h*a*r#x because the Lord, he [is] the God of the heaven from above and of the earth Text Top MTheme Rheme a?n there isn’t Top Theme

uod other Rheme

5.1.6. Theme in Bound, Minor and Elliptical Clauses Since the relation between clauses is going to be detailed below (cf. 5.4.), few examples from SH will suffice here for each of the aforementioned categories. Finite dependent bound clauses are typically introduced by rv,a]K.; (80) Gen 37.23 w~y=h' K~a&v#r - B*a yos}[ a#l-a#j*yw ¨ w~Y~pv'fW a#t-yos}p when arrived Joseph to his brothers [they] stripped Joseph Top Rheme (dominant clause) Theme

In SH, non-finite dependent bound clauses typically make use of infinitive constructs (#81). (81) Gen 12.4 w=a~br*m B#n-j*m}v v*n'm w=v]bu'm v*n> B=x}ato m}j*r*n [and Abram was seventy-five years old] when exiting from Haran (dominant clause) Theme Rheme

As an illustration for SH minor clauses, one should consider the interjection sh; (Hush!). In earlier biblical documents, the particle can appear by itself, contrary to the usage preferred in later documents (Hab 2.20; Zep 1.7, 2.17), where it is accompanied by many other constituents. Similar occurrences of shorter length are found in other texts (Deut 4.35; 7.9; 1 Kgs 8.60; 18.39; 2 Kgs 19.15). 2 Geller (1991, 21-3) provides a contemporary linguistic approach of this matter, suggesting various functions of this structure in their respective contexts, and a few more examples: Gen 2.19; 9.18; 21.29; 25.16; 40.12; Gen 45.20; Exod 16.36; 1 Kgs 18.39 inter alia. 1

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

179

(82) Judg 3.19 w~Y)am#r ¨ h*s ¨ [he] said: “quiet!” Theme Rheme Minor clause

Several examples below will illuminate the discussion of Hebrew elliptical clauses. The first example provides with a complex meaning by means of a very simple wording. The reiteration of a very short nominal group, ‘My head! My head!’, can be interpreted in this case as intensity, i.e. ‘My head hurts terribly!’ (#83). An example with a minor clause containing a negative adverb follows (#84). (83) 2 Kgs 4.19 w~Y)am#r a#l-a*b'w [he] said to his father: Fin/ Predicator Theme Rheme

¨

r)av' ¨ “my head! Subject Rheme

r)av' ¨ my head!” Subject Rheme

(84) Gen 19.2 w~Y)am=rW ¨ Oa ¨ K' b*r=job n*l'n ¨ [they] said: “no!” but in the street we spend the night The Rheme minor clause Marked Theme Rheme

Next we have a clause complex starting with an independent clause followed by an elliptical clause, with an Expletive functioning as Theme [minor clause], a situation frequently found in the prophetic literature. Again, the verb of the elliptical clause needs to be supplied, in which case an identifying clause will result: ‘(This is) the oracle of the Lord’. (85) Num 14.28 a$m)r a&l}h#m [you] say to them: Top Theme Rheme

h~y-a*n' n=a%m-yhwh “as I live – Lord’s oracle – ...” [minor clause] Topical Theme

The ellipsis of the Subject is more frequent, because of SH’s preference to conflate Subject with Finite/Predicator function. (86) Judg 3.26 w=a}hWd but Ehud Text Top MTheme

n]ml~f u~d h]tm~hm=h*m ¨ slipped away while [they] lingered Rheme

180

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

w=hWa and he Text Top MTheme

u~b~ a#t - h~PP=c'l'm ¨ passed the marking stones Rheme

w~Y]M*l}f h~C=u'r*T> ¨ and escaped to Seir Theme Rheme

After two independent clauses with the Subject conflated with the Marked Theme, another independent clause follows, but its Subject is elided.1 To summarise, the thematic structure of the clause in SH contains a Theme and a Rheme. As expected in a verb-initial language like Hebrew, typically the verb occurs first, hence unmarked, in declarative clauses with a finite verb or with an inf abs/cstr as finite verb as well as in their exclamative and interrogative counterparts. Moreover, the situation remains unchanged in volitive clauses, in verbless: classification and verbless: existence clauses. When the verb fronts the Subject in declarative: non-finite: participle clauses and in verbless:identification clauses, it functions as Marked Theme. In such cases, any other constituent that takes the place occupied normally by the Subject, functions as Marked Theme. In WH- interrogative and exclamative clauses the WH- element comes first in the clause. In polar interrogative clauses, the interrogative particle is part of the unmarked Theme. Similarly, textual Themes such as continuatives, conjunctives, conjunctive Adjuncts, modal comments and finite verbal operators belong to the unmarked Theme and take their place at the head of the clause. Whenever they appear before the Predicator, Vocatives and Expletives are considered part of the interpersonal Theme. SH does not lack equative constructions, but predicated Theme should not be confused with Marked Theme. Although they are not as frequent as in informal texts or in conversational language, bound, minor and elliptical clauses appear in the PROSE texts of HB too. 1 It is usual for verbs to be ellided in the secondary line of a verse and understood from their prior iteration in the primary line. Such verbs are traditionally known as double-duty verbs.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

181

5.2. MOOD STRUCTURE The four semantic categories that realize the Mood at the lexicogrammar level, and the various types of initiating and responding clauses they produce, can be ilustrated in SH too. initiate giving goodsand-services = offer

Exod 2.7-8

expected response = acceptance

H~a}l}k w=q*r*at' l*k a]V> m?n#q#t m]n h*u]br]Y)t Shall I go and call to you a woman of suckling amongst the Hebrew women? l=k' Go!

initiate giving goodsand-services = offer discretionary response = rejection

h~a~KK#h a~KK#h a*b' Shall I kill them, shall I kill them, my master? Oa t~KK#h Do not kill!

2 Kgs 6.21-22

initiate demanding goods-and-services = command

w=a#t-G=b'u' G=b'~u h~K#s#p T*c'm B=p' a~mT~j~t h~Q*f{n w=a#t K#s#p v]bro Then my cup, my silver cup, put at the mouth of the sack of the youngest as well as the silver for his grain. w~Y~u~c K]db~r yos}p a&c#r D]BB}r [He] did according to the word Joseph said.

Gen 44.2

initiate demanding goods-and-services = command discretionary response = refusal

l=k b=k)h&k* z#h w=hov~uT* a#ty]cr*a}l Go in this strength and save Israel! B' a&d)n*y B~M> aov'~u a#ty]cr*a}l But, Lord, wherewith shall I save Israel?

Judg 6.14-15

initiate information statement

a]m a$Oh'm hWa y*r#b lo K' n*t~x a#t-m]xB+jo If he is a deity, he can defend himself when one breaks his altar. y*r#b Bo h~BB~u~l K' n*t~x a#tm]xB+jo Let Baal contend with him because he broke his altar.

Judg 6.31-32

expected response = undertaking

giving =

expected response = acknowledgment

182

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

initiate information statement

a&h*h a&d)n*y yhwh K'-u~l -K}n r*a't' m~la~k yhwh P*n'm a#lP*n'm Ah, Sovereign Lord, I have seen the messenger of the Lord face to face! v*lom l=k* a~l -T'r*a Oa T*mWt Peace to you! You will not die!

Judg 6.22-23

initiate demanding information = question expected response = answer

B~t-m' a~TT= h~g'd' n*a l' Whose daughter are you, please tell me!

Gen 24.23-24

initiate demanding goods-and-services = question discretionary response = disclaimer

a}Y#h h~C#h l=uol*m Where is the lamb for the burnt-offering?

discretionary response contradiction

giving =

=

B~t-B=tWa}l a&n)k' B#n-m]lK> a&v#r y*l=d> l=n*jor Daughter of Betuel-ben-Milka, which was born to Nahor. Gen 22.7-8

a$Oh'm y]ra#h-Lo h~C#h l=uol*m B=n' God will provide it, the lamb for the burntoffering, my son.

The order of constituents varies according to the type of verbal forms that realizes the Predicator, that is finite, non-finite (infinitive absolute, construct, participle), or even verbless (identification, classification, existence/possession). Examples from SH will be provided later on, when the Mood structure will be discussed according to the main Mood type clauses, such as declarative, interrogative and volitive (Table 5.2). mood type declarative: finite declarative: non-finite: inf. abs./cstr. declarative: non-finite: ptcp declarative: verbless: identification declarative: verbless: classification declarative: verbless: existence declarative: exclamative

Order of constituents Fin/Pred^Subj Fin/Pred^Subj Subj^Fin/Pred Subj^Compl Compl^Subj Pred^Subj WH- element^Subj

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

mood type interrogative: finite: polar interrogative: non-finite: inf. abs./cstr.: polar interrogative: non-finite: ptcp: polar interrogative: verbless: identification: polar interrogative: verbless: classification: polar interrogative: verbless: existence: polar interrogative: WHvolitive

183

Order of constituents interr. element^Fin/Pred interr. element^Fin/Pred interr. element^Subject interr. element^Subject interr. element^Compl interr. element^Pred WH- element^Subj Pred^Subj

Table 5.2: The Order of constituents according to the mood type of clauses in SH

5.2.1. The Mood Element In SH, Finite is conflated with Predicator in most cases. For example, w~Y=b*r#k (#87) contains conflated both the Finite (the idea of ‘past tense’ in this case) and its respective Predicator (a Pi of the lexical verb √$rb). References to verbal family (conjugation) in SH belong to the Residue section.1 (87) Gen 47.10 w~Y=b*r#k y*u&q)b a#t-P~ru)h blessed Jacob Pharaoh Fin/ Pred Subject Complement Res-idue Mood

Whereas the Finite operator stays within the Mood component of the f-structure, because, along with the Subject, it realizes the selection of mood in the clause, the Predicator does not. Even in the case of non-finite verbal forms (infinitives and participles), the Finite operator is still conflated with the Predicator. (88) Gen 48.21 h]N}h a*n)k' Look, I Expletive Subj Mood

m}t am dying! Fin/ Pred Residue

1 An analysis of the verbal group in Hebrew would have been useful here. Unfortunately, the risk of being evasive and superficial is greater than its value for this transitory aspect and, therefore, it will not be approached here.

184

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Some infinitive absolutes have a notable ability to accompany same root finite verbal forms and play the role of a Finite for them (cf. Joüon & Muraoka 1993, §123). (89) Gen 2.17 K' B+yom a &k"lk* m]M#NW mot T*mWt in the day of your eating from it [you] will die Adj Adjunct Adjunct Finite Predicator Mood Residue

Where verbless clauses are considered, Finites tend not to appear, enabling the Predicator to stand alone for the verbal group. Similarly, volitives do not pinpoint a particular time reference, and therefore lack a Finite operator. (90) Gen 2.18 Oa - fob h$yot it is not good being Adj Pred Compl Residue

h*a*d*m l=b~Do the man by himself Subj Adjunct Mood

The Subject supplies the information by reference to which the proposition can be affirmed or denied. It is regularly picked up by the mood tag, as in the following reconstructed example: The woman said to the serpent, didn’t she? (91) Gen 3.2 w~TT{am#r said Fin/ Pred ResMood

h*a]vv> a#l-h~N*j*v the woman to the serpent Subject Complement -idue

Part of the Mood element of the clause is also the mood Adjunct, typically realized by an adverbial group expressing temporality, modality, or intensity, of which the first two have a strong tendency to function as Theme. In regular SH clauses, adjuncts of temporality appear after the Predicator, like dA[ when it functions adverbially.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

(92) Gen 8.21 Oa a)s]p not I will repeat Adj: Fin/ Pred mod Mo-

185

l=q~L}l uod a#t-h&a*d*m> B~u&bWr h*a*d*m cursing again the land on account of humankind Compl Adj: temp Compl Adj: circumstantial -od Residue

Adjuncts of modality cover the space between ‘yes’ and ‘no’, providing intermediate degrees of probability/usuality (for propositions) or obligation/inclination (for proposals) (cf. 4.3.1.). The particle T*m'd ‘always’ is one of them. (93) 2 Sam 9.13 K' u~l-v%lj~n h~M#l#k T*m'd Because at the table of the king always Adj: Adj: Circumstial Adj: Mood Cont Mood Residue

hWa he Subj

a)k}l was eating Fin/ Pred

In SH, the usage of positive mood Adjuncts lacks the frequency of Rabbinic or Modern Hebrew, wherein many particles are involved. On the contrary, SH prefers to reiterate, at least partially, the part of the initiation statement that is accepted as correct. (94) 2 Sam 12.19 h&m}t h~Y#l#d [...]¨ m}t ¨ ¿ Did die the child? [he]died! Fin/ Pred Subject Fin/ Pred Mood Residue Mood Res

When the reply consists of a negative statement, aOl is used adverbially and functions as a mood Adjunct. In #95, the negative answer, elliptical though it is, receives the back-up of another positive declarative: finite clause by which the truth of the previous contradiction is restated.

186

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(95) Gen 18.15 Oa x*j~qT' ¨ [...] Oa ¨ K' x*j~qT= ¨ not [I] laughed no but [you] laughed Adj: Fin/Pred/Subj/ Adj: Fin/Pred/Subj mood mood Res Mood Res Mood Mood

In SH, comment Adjuncts are rather rare (e.g., !kea'). (96) Ex 2.14 a*k}n nod~u truly is known Adj: Fin/ Pred mood Res Mood

h~DD*b*r the matter Subject

SH makes use of different conjunctions for the role of textual connectives, such as w>>, ~G', hT'[], or any association of the previous ones with various particles. Constructions such as w~w-X-q*f~l (Niccacci 1990, 36, 48),1 or clauses with Marked Theme preceded by a conjunction are frequently used with conjunctions with a disjunctive role. (97) Gen 33.17 w=y~u&q)b n*s~u s%Kkot> However Jacob went to Sukot Adj: Subj Fin/ Pred Adj: circ conj Mood Residue

Although scholars are divided over what indexes the Finite with Hebrew verbal forms, it is accepted that auxiliary verbs are missing in SH and the few verbal forms extant can have various time and aspect references. But the consensus stops here. Some even proposed that q*t~l Niccacci suggests that such a construction is used for recalling known information (e.g., Gen 39.1 and 1 Sam 28.3) as indicating the beginning of a new story. In the case of Gen 33.17, though, Jacob’s departure for Sukot represents new information. Therefore, the role of such a construction is to throw light on a previously met character, namely Jacob, after other characters have occupied the foreground scene, that is, Esau. 1

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

187

and y]qt{l verbal forms have temporal functions in PROSE, but aspectual ones in POETRY.1 Since such a decision is still difficult to make without considering the context, and in the vacuum of precise criteria to define the time and aspect references a verbal form would have in PROSE or POETRY, we are in the position to postpone pursuing the delimitation of Finite, until further clarifications are in place. In SH, modality is frequently implied from the context, as is the case with the habitual use of finite or non-finite verbs (cf. the verb of the embedded clause in #98). (98) 2 Kgs 3.11 P)h a$l'v*u B#n-v*p*f a&v#r y*x~q m~y]m u~l-y=d? a}l]Y*hW Here [is] Elisha-ben-Shaphat Æwho [used to] pour water on the hands of ElijahØ Pred Subject Adj: Fin/ Pred Compl Adjunct Conj Res Mood Mood Residue

Suggesting a new approach to the Hebrew verb, Cook (2001, 134-5) proposed a modal use for the w=q*f~l verbal forms. Although not entirely new in the field of linguistics, grammaticalization can explain the variations, noted within the limits of a given grammar, that could not be accounted for from a traditional synchronic viewpoint. w~Y]qf{l and q*f~l could have developed along the path of resultative > perfective > simple past, whereas the progressive participle could have developed as y]qf{l has along the path of progressive > imperfective. Hence, an overlap between these pairs of verbal forms can be seen. Unlike discourse analysts, Cook did not see any analogical association between w~Y]qf{l and w=q*f~l, neither in grammaticalization nor in semantics. As a result, he preferred to analyse w=q*f~l as the perfective q*f~l functioning modally. At the clause level, such a verbal form expressed to him directive deontic meaning in prescriptive texts (sample 19b in Cook 2001, 134). (99) Exod 40.9 w=l*q~jT* a#t-v#m#n h~M]vj> and [you] are to take the oil of the annointing Fin/ Pred Complement Mood Residue

Segert’s observation on Ugaritic literature (Segert 1985, 88) is confirmed by Rata (2004) on Job. 1

188

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Another substantive proposal towards modality in SH prose came from Joosten, who argued that y]qf{l mainly expresses modality (1997, 58). In his general language system, the indicative would be mainly expressed by a participle (Joosten 1989). Even when it refers to the past, ‘y]qf{l signals an action that is not real, implying either that it could yet be realised subsequently, or that, on past experience, one may expect it to be realized subsequently’ (Joosten 1999, 25). The whole article defended this position by surveying the cases where y]qf{l indeed referred to the past.1 In a later article, Joosten returned to this issue by arguing against the application of the Aktionsart theory to SH, suspected by Joosten (2002, 67) of being a category Arabic grammarians unnaturally imposed on Hebrew grammar. This time, Joosten refuted the idea that y]qf{l could express real present and concomitant actions with the main event, assigning the participle to this purpose. He admits, though, that y]qf{l could express past repetitive actions as well as habitual and general present (pp. 60-64). Nonetheless, ‘the most frequent function of y]qf{l is that of actions that are not yet begun’ (p. 64). A good summary of the type of actions mainly expressed by y]qf{l, that is actions that are not yet begun, is non-reality (irrealis).2 In SH the domain of commands and suggestions (proposals) is covered by volitive with the support of entreating particles such as hn'-, hN'ay or yBi (cf. Joüon & Muraoka 1991, §105c). An imperative realizes a requirement (#100),3 but hn'- downgrades it to something one is supposed to do (#101). Politeness and emotivity are typically indexed by the use of such a particle (Shulman 1999). (100) Num 24.11 w=a~TT> B=r~j - l=k* a#l-m=qom#k* but now flee to you to your place Adj: Cont Pred Adj Adj Residue

Of a similar persuasion is B. Isaksson in his study on Ecclesiastes (Studies in the Language of Qoheleth: With Special Emphasis on the Verbal System, Uppsala: Almquist and Wiksell, 1987). 2 Over 80 per cent of the cases in Joosten’s database of y]qf{l found in Genesis 1-11 express futurity/modality (Joosten 2002, 64). 3 Fassberg (1999) entertains the view that the lengthened imperative (hl' j.q' ) is employed in SH as a marked imperative. 1

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

189

(101) Judg 19.6 hoa#l-n*a ¨ w=l'n ¨ w=y]t~b l]BB#k* ¨ would you be content and lodge and please your heart Fin/ Predicator Pred Pred Complement Mood Residue1 Residue2 Residue3

The cohortative (first person volitive) can suggest asking for permission, or allowing someone to do something. (102) Gen 19.20 a]M*l=f> N* v*M> let me escape there Fin/ Pred Adj: circ Mood Residue

5.2.2. The Residue elem e nt Residue consists of three functional elements: Predicator (only one), Complement (one or two), and circumstantial Adjunct(s) (indefinite number). Traditionally treated as Predicate, SFG divids the verbal group into Finite and Predicator. If it is not displaced through ellipsis, the Predicator is present in all major clauses. (103) Judg 3.12 w~Y)s]pW B=n? y]cr*a}l added the sons of Israel Fin/ PredSubject ResMood

l~u&cot h*r~u B=u?n? yhwh to do evil in Lord’s eyes -icator Complement -idue

SH typically employs the use of the particle ta, to express the Complement, but on numerous occasions, it does not appear. Freedman and Geoghegan (1999) have suggested that HPy tends to avoid the use of this particle. (104) Exod 40.33 w~Y]TT}n a#t-m*s~k v*u~r h#h*x#r he gave the curtain [to] the entrance of the court Fin/ Pred Compl Complement Mood Residue

In SH, there are numerous occurrences of Adjuncts realized by means of prepositional phrases. Among the most frequently used prepositions are

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

190

the inseparable prepositions (B., l., !mi, K.) and independent prepositions such as d[;, l[;, ~[i, ynEp.li, ta, (‘with’), etc. (105) Lev 1.10 w=a]m -m]n-h~x)an q"rb*no m]n-h~KK=c*b'm o m]n-h*u]ZZ'm l=uol> and if of the flock [is] your gift of the lambs or of the he-goats for the burntoffering

Adj: conj

Pred

Subject Mood

Adjunct

Adjunct

Adjunct

Residue

Unlike English, SH prefers the Vocative at the end of the clause. (106) 1 Sam 3.6 Oa - q*r*aT' not called, Adj: Fin/ Pred mood Mood Res

b=n' my son Vocative

Some have suggested that the use of lamed vocativum, following its use in the lyric Ugaritic literature, is peculiar to HPy (e.g., Aistleitner 1974, 163; Miller 1979, 619-23; Segert 1985, §58.1; Tropper 2000, §84.12). Since its identification by W.F. Albright, the case for a vocative function of l*m}d in SH was strongly supported by Mitchell Dahood and followed up by other researchers (cf. Miller 1979, 617-9, mainly note 3). Miller noted that in Ugaritic the preference for such a usage comes along with the preference to immediately follow an imperative and to mark a name/epithet (Miller 1979, 622). After surveying the data provided by Dahood from the Psalter, Miller concludes that there is not enough evidence to support the presence of lamed vocativum in the HPy of the Psalms. With the exception of Psalms 7.8, 33.1, and 119.126, which do come closer to the typology of lamed vocativum (due to inconclusive evidence, hence, Miller overrode them too), all of the nearly 50 examples proposed by Dahood are rejected.1 Dahood insists on promoting this function of the preposition, but his arguments do not change. Still lacking the ability to root his preference more objectively, he continues to rely heavily on the literary meaning of the phrase, invoking chiasmus (Dahood 1978) or sarcasm (Dahood 1981), which cannot avoid the charge of subjectivity. The latter example attracted Pope’s attention (1988), who reopened the case of lamed vocativum. Pope invoked as arguments Jerome’s translation of the OT, the direct address to the Lord in 172 of the 176 verse-lines 1

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

191

Expletives are equally rare in SH texts (cf. #107). (107) Judg 6.22 a&h*ha&d)n*y yhwh K'-u~l-K}n

Alas, Sovereign Lord,

certainly

Expletive

Adj: mood Mood

r*a't'

m~la~k yhwh

P*n'ma\lP*n'm

the angel of the Lord Complement

face to face

[I]have seen Fin/

Pred

Adjunct: circ

Residue

5.2.3. Mood in declarative clauses Since SH is a verb-initial language, typically, the Finite is expected to precede the Subject. (108) Gen 38.6 w~Y]Q~j y=hWd* a]V> l=u}r B=koro took Judah a wife to Er his firstborn Fin/ Pred Subject Compl Adjunct Res... idue

Mood

Ocassionally, the Subject is not explicitly present, but is understood instead. (109) Gen 38.13 w~Y%GG~d l+t*m*r [they] told to Tamar Fin/ Pred Adjunct Mood

Residue

Marked Themes can interfere with the Mood constituents but they still belong to the Residue unit.

Psalm 119 has, and on a liturgical interpretation of Matthew 21.15. Althann (1997, 109-12) defended Dahood by pointing out the parallelism in Ps. 3.8-9 that drives a vocative interpretation of the second tetragrammaton, and to two cases from Ben Sira (41.1, 2). The evidence for enclitic -mem as a vocative particle seems to be stronger in Ugaritic than in Hebrew (Pope 1951; Althan 1997, 112-9).

192

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(110) Gen 39.4 w=k"l-y#v-lo and all it was his Complement

n*t~n B=y*do ¨ put in his hand Fin/ Pred Adjunct Mood Residue

Unlike English, in SH polarity is not ambiguous, being always realized by negation adverbs (part of the Mood unit, #111). (111) Gen 38.14 w=h]wa Oa n]T=n> lo and she not was given to him Adj Subj Adj Fin/ Pred Adj Mood Residue

l=a]V> as wife Adj

When the verb is non-finite, the order of the constituents follows the standardized sequence as expected in a declarative clause, namely Finite/Predicator^Subject. In #112, there are two Adjuncts, the first one conjunctive (Textual Theme) and the latter one circumstantial. Note that the verb is an infinitive construct. (112) Deut 7.20 u~d a&b)d until will perish Adj: Fin/ Pred conj ResMood

h~N]va*r'm w=h~N]cT*r'm m]PP*n#yk* the remaining and the hiding ones from your face Subject Adjunct -idue

The order of elements is altered when the verb is realized by a Participle, the Subject preceding the Finite/Predicator unit. The Mood constituent is still located at the head of the clause. (113) Deut 24.18 u~l-K}n a*n{k' thus I Adj: Subj conj Mood

m=x~ww=k* l~u&cot a#t-h~DD*b*r h~ZZ#h command you to do this thing Fin/ Pred Compl Complement Residue

In a declarative:identification clause, the Complement has the role of identifying the Subject and, therefore, comes before it.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

193

(114) Lev 23.4 a}L#h mou&d? yhwh These [are] the feasts of the Lord Pred Subject Residue Mood

Not so with the declarative:classification clause where the Complement refers to a general class, of which the Subject is a member. In this case, the Subject comes first. (115) Gen 24.16 w=h~N~u&r> fob~t m~ra#h m=a{d and the girl [was] good looking very Subject Predicator Mood Residue

In a declarative:existence/possessive clause the Subject is again preceded by the Predicator (y#v). (116) Gen 44.20 y#v - l*nW a *b z*q}n w+y#l#d z=q%n'm there is to us an old father and a child of old age Pred Subj Compl Complement Mood Residue

In SH, the exclamative clause uses an interrogative particle and the word order is not altered.1 (117) Num 31.15 h~j]Y't#m K"l-n=q#b> ¿[you] saved all female! Fin/ Pred Complement Mood Residue

The WH- particle is part of the Mood or Residue depending on the function it plays in the f-structure.

1 Similar examples are quoted by Joüon (1945, §161b) in Gen 3.11, Num 20.10, 1 Sam 2.27, 1 Kgs 18.17, 20.13, 21.19, 21.29, 22.3 and also in Jer 3.6, 7.9, Ezek 8.12, 8.15, 8.17, 47.6, Amos 5.25, Jonah 4.4, Hag 2.19 and Ruth 1.19.

194

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(118) 2 Sam 15.4 m' y+c]m}n' [oh if] someone would put me WHFin/ Pred/ Compl /Subj Mood Residue

v)p}f B*a*r#x judge in the land Compl Adjunct

Practically, the Subject is understood from the pronominal suffix attached to the verbal form as if it were a Complement, a situation resembling the Subject in an existence/possession clause.1 An exclamative clause can also be produced by means of interrogative negative clauses (alh]) as seen in #119.2 Exclamative clauses can also be constructed by means of other particles such as: hNEhi,3 %yee,4 hk'yae,5 yAh,6 yAa,7 and even the noun lAq.8 (119) 2 Sam 11.3 h&Oaz)at B~t-v#b~u B~t - a$l'u*m a}v}t aWr]Y> h~j]TT' ¿not [is] this Bat-Sheba daughter of Eliam wife of Uriah the Hittite! Adj Pred Subject Resi due Mood

5.2.4. Mood in interrogative clauses As in English, SH has the ability to construe an interrogation simply by means of rising tone9 or by changing word order.10 The interrogative

Similar samples could be 2 Sam 23.15, Exod 16.3, Num 11.4, 18, Deut 28.67. Joüon (1945, §161c) himself quotes some more examples: Judg 4.6; 1 Sam 20.37; 23.19; Deut 11.30; Josh 10.13; 1 Kgs 11.41; 14.29. Most Bible translations render Ezek 21.5 as ‘Isn’t he just telling parables?’ whereas its meaning as an exclamative clause may be even more illuminating: ‘He is just telling parables!’ 3 E.g. 2 Sam 1.18; 1 Kgs 14.19; 2 Kgs 15.11 cf. Joüon (1947, §161c). 4 E.g. 2 Sam 1.25, 27 cf. Joüon (1947, §162b). 5 Cf. Joüon (1947, §162b). 6 E.g. 1 Kgs 13.30 cf. Joüon (1947, §162d). 7 Cf. Joüon (1947, §162d). 8 E.g. 1 Kgs 18.41 cf. Joüon (1947, §162e). 9 E.g. 1 Kgs 1.24; 2 Kgs 9.11. 10 E.g. 1 Sam 16.4; 1 Kgs 2.13 cf. Joüon (1947, §161a). 1 2

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

195

particle can be omitted when an interrogative clause follows a w~w that introduces an opposition.1 Although English can construe the first type of clause by auxiliary verbs, SH prefers to use the interrogative particle for that purpose. There is a construction, though, that comes close to English polar questions, but even in that case, the use of interrogative adverbs is still necessary. (120) Gen 27.36 h&k' q*r*a v=mo ¿not called his name Adj:mo Fin/ Pred Compl od Mood Residue

y~u&q)b Jacob? Compl

In #120, the verb takes two Complements, its Subject being implicit, ‘the father’ being most likely the announcer of the child’s name. Relatively rare in the biblical text, the particle ykih] gives way to a more common adverb, aOlh], producing a typical tag interrogative clause. (121) Judg 4.6 h&Oa x]ww> ¿not commanded Adj: Fin/ Pred/ mood MoRes

yhwh a$Oh?-y]cr*a}l the Lord God of Israel? Subject -od

Although relatively frequent, the interrogative usage of aOlh} is supposedly restricted to its rhetoric function. Therefore, it is only when a hint of irony can be traced in the words of the speaker that its interrogative usage can be proposed with confidence.2 Otherwise, it is quite difficult to distinguish such an interrogative clause from an exclamative clause. By far, the most common interrogative particle in SH is h} . Abraham Even-Shoshan in his Concordance divides the occurrences of the interrogative particle into four categories: direct questions,3 questions expressing doubt,4 indirect questions,5 and disjunctive questions.6 From a formal perspective, E.g. Judg 11.23; 14.16; 2 Sam 11.11 cf. Joüon (1947, §161a). This is especially the case with LH texts, cf. Job 7.17-21; 22.2-5; 39.9-12; 40.24-31; 41.1-6; Ezek 15.2-5; Micah 6.6-7; Amos 3.3-6; Is 66.7-9. 3 E.g. Gen 29.5; 2 Sam 18.32. 4 E.g. Gen 4.9; Judg 11.25; 2 Sam 7.5. 5 E.g. Gen 8.8; 24.21. 6 E.g. Josh 5.13; Eccl 2.19. 1 2

196

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Joüon (1947, §161) discusses only the direct, disjunctive and indirect interrogative clauses. From a functional perspective, an indirect interrogative clause is actually a subordinate clause introduced by an interrogative particle (adverb or pronoun). (122) Deut 1.8 l*d~u~t a#t-a&v#r B]lb*b=k*

knowing Pred

all that [is] in your heart Complement

Residue

a¨

h&t]vm)r

m]xw)t*yw

¿whether keeping his commands Pred Compl Residue

¨a]m

b

or Adj: conj

- Oa c¨

not Adj: mood Res

The last two coordinated clauses (of which type see below) are subordinated to the previous clause by means of the interrogative particle. A disjunctive interrogative clause is, again, a subset of the direct question at the level of clause complex, by means of which two clauses are coordinated by a disjunctive particle, which replaces the interrogative particle. It can appear with both direct and indirect interrogative clauses. (123) Josh 5.13 h&l*nW a~TT> a¨ a]m - l=x*r?nW b¨ ¿for ours [are] you or for our enemies? Adj: circ Subj Adj: Adj: circ conj Residue1 Mood Residue2

In the case of the second clause, one can take note of the ellipsis of Subject and the replacement of the interrogative particle with the conditional particle. Similar occurrences are found appear in 1 Kgs 22.15, Joel 1.2, and Job 21.4. Alternatively, the conjunction Aa can share the disjunctive function of ~ai.1 Joüon (1947, §161f) illustrates a disjunctive interrogative clause as part of an indirect question and clause complex by quoting Num 13.18 and Eccl 2.19. (124) Num 13.18 h#j*z*q hWa a¨ ¿strong [is] he Pred Subj Residue Mood

h&r*p#h b¨ h~m=u~t hWa c¨ ¿weak ¿ scarce [is] he Predicator Predicator Subj Residue Residue Mood

a]m - r*b d¨ or numerous Adj: Predicator conj Residue

E.g. Judg 18.19; Eccl 2.19; 2 Kgs 6.27 and also Job 16.3; 38.28, 31; Mal 1.8 cf. Joüon (1947, §161e). 1

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

197

The WH- element can be conflated with the Subject, because it refers to the identity of the Subject. Thus, it is still part of the Mood section. (125) Judg 6.29 m' u*c> h~DD*b*r h~ZZ#h ¿Who did this thing? Wh-/Subj Fin/ Pred Complement Mood Residue

The WH- element can be conflated with a Complement function as the following verbless: identification clause indicates: (126) 1 Sam 30.13 l=m' a~TT> Whose [are] you? Wh-/ Compl Subject Residue Mood

A WH- element conflated with an Adjunct function is part of the Residue alongside the Predicator as the following non-finite:participle clause shows: (127) Gen 37.16 a?p)h h}m Where they WHSubject /Adj Mood Residue

r)u'm are grazing? Fin/ Pred

Occasionally, the WH- element supplies information about what happened or what someone had done. In SH, such a question appears mainly in the context of confronting a proven wrongdoer. As opposed to English, SH prefers a more complex manner of enquiry, by means of two clauses, of which the second qualifies the Subject of the first. (128) Gen 12.18 m~h ZZ)t u*c't> What [is] this [you] have done Wh-/Compl Subj Fin/ Residue Mood Mood

l' to me? Pred Residue

Compl

198

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Complement can also be conflated with the WH- particle l*M> (#129). (129) Josh 7.7 a&h*h a&d)n*y yhwh l*M* h}u&b~rT* h~u&b'r a#t-h*u*m h~ZZ#h Alas! Sovereign Lord, why have [you] brought over this people? Expletive Vocative Wh-/Adj Pred Finite Complement Residue Mood

It should be noted that the interrogative pronouns manifest specific preferences as to what functions they conflate. ymi ‘who’ is conflated with the Subject, ymil. ‘whose’, hm' ‘what’ and hm'l' ‘why’ are conflated with the Complement. The list of interrogative pronouns that are conflated with the Adjuncts is larger, due to the diversity of circumstances such Adjuncts cover: hpoa,e hYEa;, yae ‘where’, hn'[' ‘whither’, hn'['-d[; ‘how long’, hM'K' ‘how much’.1 5.2.5. Mood in volitive clauses The usage of direct and indirect volitives implies a formal distinction rather than a functional one in SH. Direct volitive displays distinct verbal forms for imperatives (second person volitives), cohortatives (first person volitives), and less distinct for jussives (third person volitives) (cf. Joüon 1947, §114). Indirect volitives are those verbal forms with a conjunctive w^w but expressing the idea of purpose or consecution (cf. Joüon 1947, §116). First person volitives may also carry an optative meaning. When present, the particle an', carries a sense of deprecation. (130) 1 Sam 9.3 q~j-n*a a]TT=k* a#t-a~j~d m+h~N=u*r'm take with you one of the servants Fin/ Pred Adjunct Complement Mood Residue

The volitive forms of √%lh can play the role of a Finite, indicating urgency or immediacy in relation to the action of the main verb.2

1 Dahood (1979) proposed a rather uncommon use of the particle kî is in Psalm 90.11, Is 36.19 and Hos 13.9. His option is prompted only by his translation choice. 2 Similar occurrences are to be found in Genesis 11.4, 7; 47.15 and Exodus 1.10.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

199

(131) Gen 11.3 h*b> n]lB=n> l=b}n'm Let’s make bricks Finite Pred Compl Mood Residue

Volitives are negated by the adverb la;, with the exception of imperative, which cannot be negated, but a negated y]qf)l is used instead. When the precative particle is also present, it closely follows the negative adverb,1 sometimes the main volitive verbs being elided as in #132 (Gen 19.18, 33.10). Frequently, volitives are connected in chains, expressing a sequence of actions expected by the speaker from the receiver. Only the initiating verb is required to have a volitive verbal form (or the first two if the initiating verb is %lh) as in #133. (132) Gen 47.29 a~l-n> t]qB=r}n' B=m]xr*y'm Do not bury me in Egypt! Finite Pred/ Compl Adj: circ Mood Residue (133) Ruth 2.14 G)v'

h&l)m ¨ w=a*k~lT+ m]n-h~l#j#m ¨ w=f*b~lT+ P]TT}k

Approach here Pred Adj Residue

and eat Predicator Residue

from the bread Adjunct

and dip Pred Residue

B~j)m#x ¨

your morsel in the vinegar Compl Adjunct

Here, w~w functions conjunctively as an additive conjunctive Adjunct. As already mentioned, there are instances when the secondary volitive is logically subordinated to the initiating volitive, expressing, therefore, purpose or consecution. (134) Gen 23.4 T=n> l' Give me Pred Compl Residue

*j%zz~t - q#b#r u]M*k#m ¨ w=a#qB+r> m}t' a possession of sepulchre with you so that I will bury my dead Complement Adj Predicator Compl Residue

Gen 13.8; 18.3, 30, 32; 19.7, 18; 33.10; 47.29; Num 10.31; 12.11, 12; 22.16; Judg 6.18; 1 Sam 3.17; 25.25; 2 Sam 13.25; 14.18. 1

200

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

In the case of #134, the conjunction w~w suggests result.1 Usually such derogative linguistic constructions represent defamiliarized language intended to mark social inequality and protect the social status of the interlocutors, or to replace blunt offensive speech (euphemisms) as in #135. In SH, imperative clauses are usually construed as non-Mood clauses. This is not exactly the case here, even though the primary verbal form is an imperative. Due to the use of the auxiliary particle aN", modal in function, the command is moderated into a polite request. The Predicator (realized by a verbal group) is split into two by the modal material, which functions rather as Finite than Adjunct. Subject is understood again. (135) Gen 38.16 h*b>-N*a *boa a}l*y]k Let me enter at you! Finite Pred Adjunct Mood Residue

Again, the language is not congruent with its meaning. Judah is not asking to enter this woman’s house (Tamar undercover) for a cup of tea. As it becomes clear in the context, it was a specific encoded request for paidup intercourse with a professional prostitute. To summarise, volitive clauses build upon the structure of declarative positive clauses, with the verb in the initial position. Imperative clauses do not display a Mood constituent, as other volitive clauses with cohortatives and jussives do. The negation adverb for volitive clauses is la;. 5.2.6. Mood in minor and elliptical clauses Like English, SH displays a tendency to elide less salient material, but to repeat it in the reply if considered important. Unlike English, though, SH has at its disposal a collection of literary texts, carefully edited and lacking the spontaneity so common in informal conversations. Even so, the written language is as natural as the informal spoken language. Consider the following dialogue between Jacob and the shepherds from Haran. The contexts of speech of these verbal contributions were omitted on purpose. Similar examples of indirect cohortative can be found in Gen 12.2; 24.56; 27.4, 9, 25; 29.21; 30.25, 26; 42.34; 49.1; Deut 32.1; 1 Kgs 13.7 (after an initiating imperative); Gen 18.30; 1 Sam 27.5; Is 5.19; Jer 33.3; Mal 3.7 (after an initiating jussive); Exod 3.3; 1 Kgs 19.20 (after an initiating cohortative) (cf. Joüon 1947, §116, pp. 315). The discussion can be extended with examples of indirect jussive and indirect imperative (pp. 316-7). 1

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW (136) Gen 29.4 a~j~y m}a~y]n a~TT#m Brothers, where from [are] you? Vocative Wh-/Adj Subject Residue Mood

¨

201

m}j*r*n a&n*jnW ¨ [are] from Haran we. Adjunct Subject Residue Mood

This dialogue is sustained by verbless:classification clauses, the initiating one being interrogative, the reply – declarative. Although Subject could have been elided in the reply, the author preferred to retain it. Not so in the following exchange, where the Complement (identity of Laban) is left out. The reference to his identity is elided again in the next exchange (v. 6) as if Jacob was disturbing some murky waters. (137) Gen 29.5 h*y=d~uT#m a#t-l*b*n B#n-n*jor ¨ y*d*unW ¨ Do [you] know Laban son of Nahor? [we] know! Fin/ Pred Complement Fin/ Pred Mood Residue Mood Residue

Even if any element can be missing, a reply to a WH- question requires at least the presence of the WH- element. In #138, the reply is Subjectless. (138) Gen 32.28 m~h - V+m#k* ¨ y~u&q)b ¨ What [is] your name? Jacob! Wh-/ Subject Compl Compl Residue Mood Residue

Another example displays elision of all the elements except the correspondent of the WH- element from the initiation clause (in this case a complex noun group). The relative clause describes the Subject, although a subordinate clause in itself. (139) Gen 38.18 m> h*u}r*bon a&v#r a#TT#n-l*k What pledge shall [I] give to you? WhSubject /Compl Residue Mood

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

202

j)t*mk* Wp=t'l#k* Wm~F=k* a&v#r B=y*d#k* Your seal, your cord and your staff which [is] in your hand Complement Residue

Known exclamations in SH include woe cries (yAh, yAa, ylla;;, Hha]]), interjections of pain (xa', xa'h,), despair (vaAn), quiet (sh;, e.g., Judg 3.19), deprecation (hl'ylix'), and even oath formula (hwhy-yx;, yn'doa]-yx;, yvip.n;-yx;) (IBHS, 681-4). (140) 1 Sam 20.3 w=aWl*m

But

indeed

Adj: conj

Adj Mood

j~y-yhwh w=j? n~pv#k*

as the Lord lives and your soul lives Minor clause

K'

[there is] as Adj: conj

k=p#c~u B?n' Wb?n h~M*w#t

a step

between myself and death

Pred Adj Residue

Frequently, in SH, when personal names are called, they are repeated in order to suggest the urgency of the call. (141) 1 Sam 3.10 v=mWa}l ¨ v=mWa}l ¨ Samuel! Samuel! Minor Minor

In SH, greetings have the appearance of well-wishings. (142) Judg 6.12 yhwh u]M=k* G]BBor h#j~y]l the Lord [may he be] with you, mighty warrior! Fin/ Predicator Vocative Mood Residue

Sometimes the initiating greeting is rendered as an interrogative minor clause, with a positive declarative minor clause. (143) 2 Kgs 5.21-22 h&v*lom ¨ | v*lom ¨ [Is there] peace? [There is] peace! Minor clause Minor clause Residue Residue

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

203

From such an understanding of greeting probably derives the description of greeting as asking somebody of one’s peace (Gen 43.27; 2 Kgs 9.17), or ‘blessing’ somebody (2 Kgs 4.29). The question can refer to other persons too, mainly close relatives of the receptor. Greetings and their expected replies are recorded in Gen 29.6; 43.27; 1 Kgs 2.13; 2 Kgs 4.26. Lack of a proper reply appears in 2 Sam 20.9 (clandestine smiting disguised by the greeting), 2 Kgs 9.11, 18, 22, 31 (greeting denied by a derogatory reply). An example of alarm follows: (144) 2 Kgs 4.19 r)av' ¨ r)av' ¨ My head! My head! Minor Minor Mood Mood

The minor clause reiterated consists of a nominal group, which can be considered as the Subject of a clause with the Finite/Predicator ellided. It qualifies, though, as an appeal. When an alarm minor clause has an imperative verb, its Mood structure has at least a Residue. Note the minor clause ‘Run!’ (2 Sam 18.23). We are not aware of SH having continuative elements, mainly due to the less interactive character of the written language of HB. 5.2.7. Extensions of Mood analysis In SH, the article or the particle rv,a] is used to material already announced or detailed later (so called ‘cleft clauses’ in traditional grammars) (IBHS, 13.5d, 19.3). At times, the Subject is realized by a nominal group complex, in which the Postmodifier is a relative clause. (145) Gen 28.13 h*a*r#x a&v#r a~TT> v)k}r u*l#yh* the land that you lie down on it Subject Æembedded clauseØ Mo-

l+k* to you Adjunct Res-

a#TT=n#N> [I] will give Fin/ Pred -od -idue

Wl=z~ru#k* and to your seed Adjunct

In #146, the Adjunct receives special attention from the narrator, being construed as a nominal group in which the relative clause functions as Postmodifier, in reference to the peculiar tree. Both the Residue and the Mood are split in two. In this case, there is a further backreference, this

204

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

time to the ‘fruit’. Therefore, the clause can be translated ‘As for the fruit of the tree that is in the middle of the garden - said God - do not eat any of it’. (146) Gen 3.3 Wm]PP=r' h*u}x a&v#r B=tok-h~G*n ¨a*m~r a$Oh'm¨ Oa t)ak=lW

but from the fruit that [is] in the of the tree middle of the garden Adjunct Æembedded clauseØ Res-

said God

m] h~hWa a&v#r r=a't#m

[we] walked Process

all the wilderness Scope

the great

and dreadful one

that you saw

Here the Actor is implied from the previous context, the Israelites and Moses, but it is also implicitly expressed in the clause through the verbal form (first plural and second plural) in use. Communicating about playing instruments in SH follows a similar tendency to associate the verb ([wr ‘to sound’, !tn ‘to give [sound]’, vpt ‘hold’, !gn ‘to touch’, [qt ‘to blow’) with the name of instrument that is being played.

208

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(160) Josh 6.16 T*quW h~KK{h&n'm B~vop*rot blew the priests into the horns Process Actor Scope

SH does not lack material Processes that accompany Scope rather than Goal. Besides ‘singing a song’ and ‘crossing a field/water course’, one can also consider ‘make an agreement’ (literally ‘cut a covenant’). (161) Judg 10.9 w~Y~u~brW b=n?-u~Mon a#t-h~Y~rD}n crossed the Ammonites the Jordan Process Actor Scope: entity (162) Gen 26.28 w=n]kr=t> b=r't u]M*k let us cut a covenant with you Process Scope: entity Client

The previous example displays an implicit Subject (we) and the Predicator has a double Complement (covenant, you). At the experiential level, one of the Complements functions as Scope and the other as Client (beneficiary of the services offered to him) (see details infra). The participants identified as having the function of Scope qualify as such since they are not affected by the performance of the Process, and their existence is not hindered in any way by the Process. Recipient and Client construe a benefactive role. (163) Gen 43.24 w~Y]TT#n m]sPoa l~j&m)r}h#m [he] gave fodder to their donkeys Process Goal Recipient

In other instances, the original owner of the goods can be construed circumstantially as a locative source. (164) 2 Kgs 11.2 w~TT]gn)b a)to [she] stole him Process Goal

m]Ttok B=n?-h~M#l#k from among the princes Circumstance

In #165, an abstract Location Circumstance realizes the original owner and the goods transferred function as Goal.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

209

(165) 2 Sam 8.4 w~Y]lK)d D*w'd m]M#NW a#l#p Wv=b~u m}aot P*r*v'm captured David from him a thousand and seven hundred horses Process Actor Circumst Complement

Note the use of l*m}d to mark Client in SH. (166) Gen 33.17 w~Y]b#n lo B*y]t ¨ Wl=m]qn}hW u*c> s%KK)t ¨ [he] built for him a house and for his livestock [he] made shelters Process Client Goal Client Process Goal

As opposed to English, SH manifests a greater interest in using prepositions. In #167, l*m}d precedes a depictive Attribute (‘plentifully’, literally ‘by handfuls’). (167) Gen 41.47 w~TT~U~c h*a*r}x B=v#b~u v=n? h~C*b* l]qm*x'm produced the land during the seven years of plenty by handfuls Process Actor Circumstance Attribute

5.3.2. Transitivity structure of mental clauses Even though SH does not possess so many ‘mental’ verbs as English or other modern languages do, the four categories of sensing are well represented. Perceptive sensing includes √!yb – ‘to discern’, √har – ‘to see’, √!za – ‘to hear’, √~[j – ‘to taste’ and √xyr – ‘to smell’. In the category of cognitive sensing, the following ones are found: √[dy – ‘to know, recognize, understand’, √rkz – ‘to remember’, √bvx – ‘to think, imagine’ and √hvn – ‘to forget’. Not less represented is the group of desiderative sensing verbs, such as √rxb – ‘to choose’, √hkx – ‘to wait, await, long for’, √#w[ – ‘to counsel, plan’, √vq – ‘to seek, intend’, √[mv – ‘to agree’ and √!a – ‘to refuse’. Lastly, emotive sensing verbs include Processes such as √bh – ‘to love’, √anf – ‘to hate’, √xm – ‘to rejoice’, √db – ‘to grieve’, √dps – ‘mourn, wail’, √lb – ‘to lament’ or √ary – ‘to fear’. The two regular Participants of a mental clause are the Senser and the Phenomenon (#168). Occasionally, the clause can lack its Phenomenon (#169), or the Senser be only implicitly present, since Subject is conflated with Predicator (#170)

210

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(168) Gen 13.11 w~Y]bj~r- lo chose for himself Process: Circumst mental

lof a}t K"l-K]KK~r h~Y~rD}n Lot the whole plain of Jordan Senser Phenomenon

(169) 1 Sam 11.15 w~Y]cm~j v*m v*aWl w=k"l-a~nv? y]cr*a}l u~d-m=a{d rejoiced there Saul and all the men of Israel exceedingly Process: Circ Senser Circumstantial mental (170) Gen 31.1 w~Y]vm~u a#t-D]br? b=n?-l*b*n [he] heard the words of Laban’s sons Process: Phenomenon mental

Lexical verbs known traditionally as stative but expressing a moral judgment, such as [[;r' (to be evil/bad) or bj;;y' (to be good/well/pleasing) produce a more nuanced type of mental clause. In such cases, the formula ‘in the eyes of PN’ indicates Senser. (171) Gen 41.37 w~Y'f~b h~DD*b*r B=u?n? p~ru)h Wbu?n? K"l-u&b*d*yw was pleasing the matter to Pharaoh and to his officials Process: Phenomenon Senser sensing

Notice the material nature or reference of the Phenomenon involved in producing the following mental clauses. Whereas in #172 Phenomenon is realized by an infinitive construct, in #173 Phenomenon consists of an existing clause. (172) Judg 7.10 w=a]m-y*r}a a~TT> but if fear you Process: Senser mental

l*r#d#t to descend Phenomenon: act

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

(173) Gen 42.1 w~Y~r=a y~u&q)b understood Process: mental

Jacob Senser

211

k' y#v-v#b#r b#m]xr*y]m that there is corn in Egypt Phenomenon: fact

Both previous clauses can receive an agnate realization, charging the Phenomena with the function of Subject and Theme at the interpersonal and textual level, respectively: ‘if descending scares you’, ‘there being corn in Egypt came to Jacob’s understanding’. The metaphenomenal clause in SH proves to be more of a complex structure. Both the independence and the abstraction that characterize it are easily noted, as well as their transformation from acts to facts through propositionalization. In #174, Leah’s ‘suffering’ is replaced by ‘her being hated’. Similarly in #175, realization of Joseph’s brothers that Jacob prefers Joseph as his heir is rendered as ‘their father loving Joseph more than his brothers’. (174) Gen 29.33 K'-v*m~u yhwh heard the Lord Process: mental Senser

K'-c=nWa> a*n)k' that hated [am] I Process: mental Phenom Phenomenon: fact

(175) Gen 37.4a w~Y]ruW a#j*yw Understood his brothers Process: mental Senser

K'-a)to a*h~b a&b'h#m m]K"l-a#j*yw that him loved their father more than his brothers Phenom Process: mental Senser Circumstance Phenomenon: fact

A clause that represents an idea brought into existence by the mental processing indicates the ability of a mental clause to project (#176). (176) 2 Sam 14.16 K' y]vm~u h~M#l#k ¨ l=h~X'l a#t-a&m*to m]KK~p h*a'v ... ¨ will agree the king to deliver his maid from the hand of the man ... Process: Senser Process: Goal Circumstance mental material

Although the Process of ‘understanding’ can construe the sensing activity of reasoning (#176), it can also refer to a preliminary decision that king David may have reached on the juridical matter presented by the

212

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

woman from Tekoa. Probably a clearer illustration of such cases comes from the use of mental Processes that refer to the content of thinking (#177). (177) Deut 8.17 w=a*m~rT* B]lb*b#k* ¨ K)j' w=u)x#m y*d' u*c>

[you] will think

Process: mental

my power and the strength of my hand Actor

l'

produced

for me

Process: material

Circ

a#t-h~j~]l h~ZZ#h ¨

this wealth Goal

In this case, there are two clauses, of which the second is subordinated to the first, and reveals the content of Israelite consciousness, rendered by the mental Process of thinking and realized by the Semitic formula ‘say to one’s heart’. 5.3.3. Transitivity Structure of Relational Clauses In this section, we will survey the attributive and identifying clauses. In SH, attributive clauses tend to make very resourceful use of the domain of nonfinite clauses.1 Even though the verb is missing, a simple juxtaposition of two noun groups, functioning as Carrier and Attribute, can render an intensive relation of the Attributive mode. (178) 2 Sam 13.2 K' B+tWl> h'a [was] virgin she Attribute: Carrier quality

Since the familiar construction places the Attribute in front of its respective Carrier, reversed constructions are considered Marked. The disjunctive w~w typically accompanies the Carrier and functions as a supplementary sign for such an interpretation. (179) 2 Sam 13.3 w=yon*d*b 'v j*k*m m=a)d Jonadab [was] a man of shrewdness exceedingly Carrier Attribute: quality

Among those who tried to offer an exhaustive list of relative clauses is de Regt (1999), who named them ‘nominal clauses referring to participants’. 1

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

213

An alternative realization of attributive clauses makes use of the verb hyh. As expected in the case of a finite clause, the tendency is to allow the verb to occupy the most salient slot. (180) Gen 39.6 w~y=h' yos}p y=p}h-t)a~r w'p}h m~ra#h was Joseph fair of shape and good looking Process: Carrier Attribute: quality intensive

In other circumstances, especially with q*f~l verb forms of √hyh, the word order follows a different pattern, actually reversing the expected order of a classification verbless clause with the Subject in the most salient slot (Marked Theme). Therefore, the realization with q*f~l verbal forms of hyh can be said to represent a Marked variant of the one with w~y=h'. (181) Gen 10.9 hWa - h*y> g]BB{r - x~Y]d l]pn? yhwh he was mighty of hunting before the Lord Carrier Process: Attribute: quality Circ: Manner intensive

SH realizes the attributes of possession mostly by means of non-finite clauses with the particles vyE/!yIa;. Again, the familiar rendering of such attributive: possession clauses allocates the most salient slot to the possession particles, and expresses the Carrier: Possessor by means of a noun with l*m#d. (182) Gen 44.20 y#v l*nW a*b z*q}n have we an old father Process: Carrier: Attribute: intensive Possessor Possessed

Sometimes, by omitting the possession particle, the possession clause becomes elliptical. With such a construction, the author gives prominence to the Carrier instead (cf. also 2 Sam 13.3). (183) 2 Sam 13.1 Wl=a~bv*lom B#n-D*w]d a*jot y*p> [there was] to Absalom, son of David, a beautiful sister Carrier: Possessor Attribute: Possessed

214

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Lastly, the circumstantial realization of the Attributive mood manifests the Attribute by means of prepositional phrases or adverbial groups that refer to Circumstance. (184) Num 6.7 K' n}z}r a$Oh*yw u~l-r)av' the consecration of his God [is] on his head Carrier Attribute: circumst (185) 2 Kgs 6.11 m' m]V#L*nW a#l-m#l#k y]cr*a}l Which of you [is] for the king of Israel? Carrier Attribute: circumstantial

SH shares with English the ability to allow other lexical verbs, that build up the verbal group functioning as Processes in attributive clauses, to express the attributive relationship between participants. (186) Gen 45.16 w~Y'f~b B=u?n? p~ru)h Wb+u?n? u&b*d*yw [it] pleased Pharaoh and his officials Process: Circumstantial: Angle intensive

The Carrier refers to the news that had reached Pharaoh’s palace, namely that Joseph’s brothers had arrived, realized as the immediately preceding clause. Although a mental interpretation of this preceding clause may seem appropriate, the relational interpretation is to be preferred for several reasons: (1) mental clauses are coherent with submodifiers like so, very, too; (2) mental clauses admit the past tense; (3) mental clauses typically do not need supplementary entities to function as Circumstances (IFG, 224). Besides, its agnate construction does not imply the use of a finite verb. The formula B=u?n? + PN can easily be appreciated as a circumstantial entity (Angle: viewpoint). Example (187) offers an alternative analysis of this clause type. (187) Num 14.8 a]m-j*p}x B*nW if delights in us Process: Attribute circumst

yhwh the Lord Carrier

The verb √vry could be considered as another candidate for a lexical verb that realizes the attributive relationship between the participants, but in

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

215

SH, this verb is basically active, therefore material, because it implies a process of taking possession of something (Wright, NIDOTTE, H3769). By far, the largest category of SH verbs that fit in this class is the one traditionally called stative verbs. Formally speaking, alongside the traditional stative verbs, there are other types of verbs too. Thus, q*f}l and q*f)l are the most quoted class of stative verbs, with roots such as the following lexical verbs: ~vea' ‘be guilty’, !veD' ‘be fat’, !qez' ‘be old’, #mex' ‘be sour’, !bel' ‘be white’, amej' ‘be unclean’, and bjoy' ‘be good’, respectively. One could also add to them hl verbs (ha'l' ‘be weary’, hl'B' ‘become old’, hl'x' ‘be weak’, hn'z' ‘be a harlot’), [[ verbs (~m'x' ‘be warm’, %k;z' ‘be clear, bright’, dd;B' ‘be separate’, dd;x' ‘be sharp’, ~m;D' ‘be dumb’), or even wy[ verbs (vAB ‘be ashamed’, !Wh ‘be easy’, rWz ‘be stranger’, bWx ‘be guilty’). Of course, strong verbs can also be included, such as xr;a' ‘be corrupt morally’, #m;a' ‘be strong’, rb;G' ‘be mighty’, ~l;x' ‘be healthy’ and vp;x' ‘be free’.1 A clause such as the one in #188 may have an agnate realization as #189 indicates, where an adjective takes over the function of the stative verb. (188) Gen 18.20 w=j~F*at*m K' k*b=d> m=a)d their sin [is] grievious exceedingly Carrier Process: Circumstance intensive (189) Gen 13.2 w=a~br*m K*b}d m=a)d Abram [was] wealthy exceedingly Carrier Attribute: quality

In SH, the identifying clause is realized by different clause types, interrogative included. Unlike English, though, we do not know whether SH has the ability to switch stress simply by intonation and to make it visible in writting. Consequently, the conflation between Subject and Identifier or Identified is predictable and relates to the slot Subject occupies in a given clause: Subject/Identified/Theme and Subject/Identifier/Marked Theme.

For a supplementary discussion and more examples on the matter, see IBHS (pp. 372-3). 1

216

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(190) Gen 27.32 m' - a*TT> Who [are] you? WhComplement /Subj Identified Identifier /Token /Value

The previous question requires the identity of the receiver. Since the Value of the person is not known, an agnate question may be ‘what is your identity?’. Although an appropriate answer would be needed to reveal this secret, another attributive clause delays the familiar reply. The straightforward answer would have been I am Esau. Any of the other possible answers would not be narrow enough to qualify as identifier. Nevertheless, Esau produces the following reply: (191) Gen 27.32 a&n' B]nk* I [am] your son, Subject Compl Carrier

Attribute

¨

b=k)r=k* your firstborn, Compl

u}c*w ¨ Esau Compl

Identifier /Value

The first clause is just a classification clause and, thus, rather ambiguous, since there were two entities who could fit this description, both Esau and Jacob. Given the circumstances, a second clause necessarily follows, this time a classification one with implied Subject. Therefore, the final qualifications ‘first-born’ and ‘Esau’ are more than welcome. Another example of the juxtaposition of an identifying clause and an attributive clause comes from the most important theological sentence of the SB, that of the Shema. (192) Deut 6.4 yhwh a$Oh?nW The Lord [is] our God. Subject Compl Identified /Token

Identifier /Value

¨

yhwh a#j*d ¨ The Lord [is] one. Subject Compl Carrier

Attribute

The author could choose to shift the focus from the Identified to the Identifier as in #193.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

217

(193) 1 Sam 1.26 a&n' h*a'V> h~N]X#b#t u]Mk> B*ZZ#h l=h]tP~L}l a#l-yhwh [was] I the woman standing beside you here praying to the Lord Subject Complement Identifier /Value

Identified/Token

As in English, SH uses identifying clauses for different purposes, the giving of names having an important role. (194) Gen 32.28 m~h V=m#k* ¿what [is] your name? Wh-/Compl Subject Identifier /Value

Identified /Token

(195) 2 Sam 13.1 Wv=m*h and [was] her name Subject

T*m*r Tamar Complement

Identified/Token

Identifier /Value

We need to consider next the role-play sub-type of an identifying clause. (196) 1 Sam 29.3 h&Oa-z#h d*w]d u#b#r v*aWl m#l#k - y]cr*a}l This [is] David, the servant of Saul, King of Israel [, is he not]? Subject Complement Id/Token

Identifier/Value

Specific to the onirocritic ancient literature, there appears to be the symbolization sub-type of identifying clauses.1

1 Equivalence is the sub-type of identifying clauses that appears in the proverbial material of LH, in which comparisons are frequently involved to convey the meaning (e.g. Prov 16.14).

218

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(197) Gen 40.12 v=l)v#t h~C*r]g'm v=l)v#t y*m'm h}m the three tendrils three days [are] they Subject Complement Subject Identified/Token

Ir/Value

Id/Token

Frequently, SH prefers to conclude lists with a general statement that refers back to the list by means of an inclusive demonstrative pronoun. (198) Gen 36.43 a}L#h a~LWp? a$dom these [are] the chieftains of Edom Subject Complement Id/Token

Identifier/Value

Occasionally, SH may rely on a lexical verb, such as hyh, to express the possession relationship between its participants. (199) Josh 19.41 w~y=h' G=bWl n~j&l*t*m was (=consisted of) the border of their inheritance Process: Subject possession Identified/Token

x"ru> w=a#vT*aol w=u'r v*m#v Zorah, and Eshtaol, and Ir-Shemesh ... Complement Identifier/Value

5.3.4. Transitivity structure of other types of clauses Behavioural, verbal and existential clauses appear as minor clause types. In SH, the behavioural clauses can be realized by many Processes such as bv;x' ‘think’, ~l;x' – ‘dream’, jb;n' ‘look’, [m;v' ‘listen’, rm;v' ‘watch’ (near mental Processes); rBeDI ‘talk, argue’, !Wl or hm'h' ‘murmur, grumble’, hl'a' or rr:a ' ‘curse’, %r:B' ‘bless’ (near verbal Processes); q[;c' or q[;z' ‘cry’, qx;c' ‘laugh’, hg'h' ‘moan, growl’, xp;n' ‘breath’, rr;z' ‘sneeze’, !vey' ‘sleep’, ~d;r' ‘fall in heavy sleep’ (psychological Processes); ~Wq ‘sit up’, ryvi ‘sing’, dq;r' ‘dance’, bk;v' ‘lie down’ (near material). The main Participant of a behavioural clause is the Behaver. Example (201) displays a Behaver as Marked Theme. (200) 2 Kgs 4.35 w~y=zor}r h~N~u~r u~d-v#b~u P=u*m'm sneezed the youngster seven times Process: Behaver Circumstance: Extent behavioural

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

(201) Gen 21.6 K"l-h~V)m}~u everyone who hears Behaver

y]xj~q laughs Process: behavioural

219

- l' with me Circumstance: Accompaniament

In SH verbal clauses, lexical verbs frequently quoted for the realization of verbal Processes are rm;a' ‘say’, rBeDI ‘speak, tell, talk’, dg;n' ‘tell’, hn'[' ‘answer’, hW'c' ‘order, command’, vg;n' ‘require’, [b;v' ‘promise’, la;v' ‘ask’, vr;D' ‘enquire’, hd'y' ‘praise’, %reB' ‘bless’1, @rex' ‘insult’, byrI ‘contend, criticize’, lLeP't.hi ‘pray’. Samuel Meier (1992) and Cynthia Miller (1996) have researched the intricacies of speech in SH non-narrative and narrative texts, respectively. The conclusions of these studies, particularly the latter, provide extensive support to our thesis. The lexical verb rma appears by far the most frequent, both as inflected forms and as rmal (not functioning as a true infinitive), in which case it typically concludes the quotative frame and marks the direct discourse that immediately follows (Miller 1996, 182-5). When inflected, other lexical verbs may accompany rma, not necessarily verba dicendi or metapragmatic verbs (cf. lists A-G in Meier 1992, 99-111). According to Miller, original multiple-verb frames should include only metapragmatic verbs (at least two) that have the same referent (Miller 1996, 149-57). Quotative frames with rmal admit as part of their quotative frame not only metapragmatic verbs, but also metapragmatic phrasal expressions, as well as other categories of verbs (Miller 1996, 192-4, Table 3). Miller also noted that SH uses rmal as a means of differentiating direct discourse in conversational settings from discourse that takes place in non-dialogic settings (Miller 1996, 393-8; cf. Bandstra 1995, 569). Sayer plays the central participant role, with Receiver and Verbage also being present. One of the most popular formulae, marking the direct discourse in SH, follows the pattern seen in #202. (202) Gen 34.8 w~y+d~BB}r j&mor a]TT*m l}am)r said Hamor to them thus: Process: saying Sayer Receiver Verbiage

At times, this verb carries the meaning of strong emotions towards another conscious entity, mainly God (cf. Ps 26.12B). 1

220

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

It should be noted that the infinitive construct form of rm;a' (rmoale), that marks the end of the quotative frame, can be rendered as an adverb replacing a more specific reference to the content. The Sayer features are confined to entities endowed with consciousness, whether human or otherwise. Balaam’s she-ass is able to speak following a miraculous touch from God (#203). (203) Num 22.30 w~TT)am#r h*a*ton a#l-b]lu*m said the she-ass to Balaam Process: Sayer Receiver saying

When quoting the message of a letter, the infinitive construct (rmal) is preferred to the inflected forms of the same lexical verb. (204) 2 Kgs 5.6 w~Y*b}a h~S#p#r #l-m#l#k y]cr*a}l l}am)r [he] delivered the letter to the king of Israel saying Process: Goal Recipient Attribute material

In this case, ‘the letter’ is the Goal of a material clause (Actor understood), and its message does not function as a projection of a Quoting clause but rather as the Attribute of the Goal. Even though multiple verba dicendi Processes may be involved in the quotative frame, each verb maintains a Process function for its respective clause. (205) 2 Kgs 25.24 w~Y]V*b~u l*h#m G=d~ly*hW Wl=a~nv?h#m ¨ w~Y)am#r l*h#m ¨ swore to them Gedaliah and to their men [he] said to them Process: Sayer Process: Receiver saying saying Receiver

SH existential clauses also display the ability to construe existential Processes by means of different lexical verbs: hy'h' ‘be, happen’, hr'P' ‘be fruitful’, hb'r' ‘multiply’, xr;s' ‘hang’, #b;r' ‘lie down’, etc. A large number of such clauses can be found in the creation epic as the following sample shows:

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

221

(206) Gen 1.3 1 2 y=h' aor ¨ w~y=h' - aor ¨ let there be light and it was light Process: Existent Process: Existent existing existing

Alternatively, existential clauses with a q*f~l verbal form of √hyh, can be used when a participant in that clause functions as Marked Theme and precedes the verb. Such a formula, known as w~w-X-q*f~l, is used extensively in biblical narrative allowing the storyteller to focus on details significant to the plot. (207) Gen 41.54 Wb=k"l - a#r#x m]xr~y]m h*y> and in all the land of Egypt there was Circumstance: location Process: existing

l*j#m food Existent

A complex sentence that includes diverse existential clauses with the Existent fronting the verb will allow us to illustrate other aspects of the existential clause. (208) Gen 2.5 w=k)l c'~j h~C*d#h f#r#m y]hy#h B*a*r#x 1¨ and every shrub of the field not yet appeared on the earth Existent Process: existing Circ: location w=k"l-u}c#b h~C*d#h f#r#m y]xm~j 2¨ and every herb of the field not yet sprouted Existent Process: existing K' Oa h]mf'r yhwh a$Oh'm ~l-h*a*r#x 3¨ because did not make to rain Lord God on the earth Process: doing Actor Circ: location w=a*d*m a~y]n l~u&b)d a#t-h~a&d*m> 4¨ and humankind there was not to work the soil. Existent Process: Circ: cause: purpose existing

Of the four clauses in this sentence, only three qualify as existential clauses. Their Processes are realized by the most neutral existential verb (hy'h'

222

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

– clause 1), by a lexical verb that signals an existential meaning (xm;c' – clause 2), and by the denying of existence/possession particle (clause 4). Commonly, existence is indicated by the particle vye, while the denying of existence is rendered by !yIa;. (209) Gen 42.12 y#v - v#b#r there is corn Process: Existent existing

b=m]xr*y]m in Egypt Circ: location

Occasionally, the Existent can be an abstract entity, as in the following example: (210) Gen 20.11 r~q a?n Surely there isn’t Process: existing

y]ra~t a$Oh'm B~M*qom h~ZZ#h fear of God in this place Existent Circumstance: location

Clause 3 in #208 had a lexical verb that can be classified in the meteorological Processes, a category situated at the borderline between material and existential. Unlike in #211, in SH, meteorological events are typically realized as material Processes whose Actor is frequently God himself. (211) 1 Kgs 18.45 w=h~V*m~y]m h]tq~DD+rW u*b'm w=rW~j 1¨ and the sky was darkened by clouds and wind Goal Process: doing Actor w~y=h' it happened Process: existing

g#v#m G*dol 2¨ a heavy rain Existent

Here, the first clause has a material Process of doing, which is realized by a passive verb, therefore the first participant functions as Goal and the second as Actor, as would be expected in a receptive material clause. The second clause, however, has only one participant and uses the verb hy'h' in connection with a meteorological phenomenon. Consequently, the second clause qualifies as an existential clause.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

223

5.3.5. Circumstantial ele m ents Extent refers to the distance in space, or the duration or frequency in time over which the Process unfolds. In SH, Distance is rarely measured in spatial units, but in time units instead. (212) Exod 15.22 w~Y}l=kW v=Ov#t-y*m'm [they] walked for three days CExtent: distance Process: happenning (213) Gen 31.41 u&b~dT'k* [I] served you Process: Goal doing

B~M]dB*r in the wilderness CLocation: place

a~rB~u-u#cr}h v*n> B]vT? b=n)t#yk* fourteen years for your two daughters CExtent: duration CCause: purpose

[...] a¨

w~TT~j&l}p a#t-m~cK%rT' u&c#r#t m)n'm 2¨ and [you] changed my wages ten times CExtent: frequency Process: doing Goal

Location refers to static placement in space/time, but also the source, path and destination of movement in space/time. SH prefers to express location by means of prepositions such as la, (IBHS, 193-4), rx;a; (IBHS, 193), lc,ae (IBHS, 194-5), B. (IBHS, 196; Jenni 1992, 171-328)1, !yBe (IBHS, 199-201), d[;B; (IBHS, 201-2), K. (IBHS, 205), l. (IBHS, 205-7), !mi (IBHS, 2123), d[; (IBHS, 215-6), l[; (IBHS, 216-7), tx;T; (IBHS, 220). (214) Gen 11.31 w~Y}x=aW ]TT*m m}aWr K~cD'm l*l#k#t a~rx> K=n~u~n [they] exited with them from Ur of Chaldeans to go to the land of Canaan. CAccomp CLocation: place CCause: CLocation: place Process: happening purpose

Space references are frequently realized in SH by the support of he locale attached as a suffix on the noun that expresses distance, location or direction (IBHS, 185-6).

1 According to Jenni’s statistics (1992, 69), the preposition b is used in over 70% of all cases to mark circumstances of time or space (temporalisation – 57.9%, lokalisation – 15.9%).

224

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(215) Gen 13.14 Wr=a}h m]n-h~M*qom a&v#r - a~TT> v*m look from the place that you [are] there Process: CLocation: place sensing

x*p)n> w*n#g#B> [...] to the and south, north, CLocation: place

Time/Place references may, at times, be very detailed. (216) Judg 21.12 w~Y*b'aW a ot*m [they] brought them Process: doing Goal

a#l-h~M~j&n#h v]Oh &v#r B+a#r#x K=n*u~n to the camp at Shilo, that [is] in the land of Canaan CLocation: place

Since the best way to substantiate an objective temporal reference in ANE times involved a reference to a historical figure, the Biblical chronicles do exactly this. The place circumstantial element is, again, not material. (217) 2 Kgs 9.29 Wb]vn~ta~j~t #cr}h v*n> l=yor*m [...] m*l~k in the year eleventh year of Joram [...] reigned CLocation: time Process: happening

&j~zy> u~l-y=hWd> Ahaziah over Judah Actor Range

SH exhibits a very unusual construction to realize Location: place, using the preposition !yBe. This appears in pairs, following the phrase model B?n X Wb?n Y, where X and Y are members of two different classes. Its variant, which occurs more rarely, is B?n X l=Y.1 Since in a large number of cases the lexical verbs used to construe the Process are of a type other than material, such clauses employ the preposition in an abstract manner (#218). (218) Gen 16.5 y]vP)f yhwh B?n' Wb?n#k* may judge Lord between me and between you Process: Behaver CLocation: place behaving

Manner refers to the modalities in which the Process is actualized, including Means, Quality, Comparison and Degree, that describe the Process. Several examples from SH will suffice to illustrate the point. Particles B. (IBHS, 197), l. (IBHS, 206-7), K. (IBHS, 202-3), la, (IBHS, 193), d[; (IBHS, 215-6), can contribute to the realization of these functions. 1

IBHS traced at least twenty-four occurrences, p. 200.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW (219) 2 Sam 23.21 w~Y~h~rg}hW [he] killed him Process: Compl doing

225

B~j&n'to with his own spear CManner: means

Comparisons are very frequent in the sapiential texts, though not restricted to them. IBHS (pp. 202-3) distinguishes three domains where comparisons, called ‘agreement’, are active: quantity or measure, kind, and manner or norm. (220) Exod 19.18 w~Y~u~l u&v*no K=u#v#n h~K]bv*n ascended the smoke like the smoke of a kiln CManner: comparison Process: doing Actor

The subcategories of Quality and Degree of the Circumstantial function of Manner employ the usage of adverbs. Although close one to another, these functions are different from the function of an Attribute in a Material clause, used to construe a quality of a Participant (Actor or Goal) after the Process has been completed. (221) Exod 23.15 w=Oa-y}r*aW do not appear Process: happening

[*n~y before me CLocation: place

r?q*m empty-handed CManner: quality

The circumstantial degree can also be realized by means of prepositions, such as B. (IBHS, 197). (222) Deut 28.47 T~j~t a&v#r Oa-u*b~dT* a#t-yhwh a$Oh#yk* B=c]mj> Wb=fWb l=b*b m}rob K{l

because Adjunct

[you] did not serve Process: doing

the Lord your God

joyfully and gladly

wherein plenty of all

Beneficiary

CManner:

CLocation:

time

quality

Some adverbs are appropriate to a more particular usage, namely Degree, provided they encode a sense of intensity, such as high, large, great.1 1 Although one can identifiy multiple instances where the infinitive absolutes are successfully used as Degree, SH uses them also as mood Adjuncts. That is more likely to happen when an infinitive absolute precedes a same-root finite verb. IBHS suggests Jer 51.9 as a good illustration of an exceptional case, where the preposition

226

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(223) Deut 13.15 w=v*u~lT* [you] enquire Process: happening

h?f}b thoroughly CManner: degree

(224) 1 Sam 26.21 w*a#vG#h [I] erred Process: behaving

h~rB}h m=a)d exceedingly CManner: degree

SH uses the preposition d[; by itself, or in association with the noun daom., to express degree. (225) 1 Sam 11.15 w~Y]cm~j v*m v*aWl rejoiced there Saul CLocation: Process: Behaver behaving place

w=k"l-a~nv? y]cr*a}l u~d-m=a)d with all Israel greatly CAccompaniment: CManner: comitative degree

Cause comprises three sub-categories, typically realized by prepositional phrases: Reason, Purpose, and Behalf. The prepositions that SH employs more frequently for the realization of Cause are !mi (IBHS, 213), l[; (IBHS, 217), ![;y: (IBHS, 202), l. (IBHS, 207 – dativus commodi et incommodi), rWb[]B,; or xk;nol.. To distinguish between Reason, Purpose and Behalf is not difficult if the context is taken into consideration. (226) Gen 18.32 Oa a~vj't B~u*bWr h*u&c*r> not [I] destroy on the account of the ten CCause: reason Process: doing (227) Gen 25.21 w~Y#Ut~r supplicated Process: bahaving

y]xj*q Isaac Behaver

(228) 1 Kgs 2.18 a*n)k' a&d~B}r I will speak Sayer Process: saying

l~yhwh l=n)k~j a]vTo to the Lord on behalf of his wife CLocation: Space CCause: behalf

u*l#yk* a#l-h~M#l#k on your behalf to the king CCause: behalf Receiver

la, is used to mark the circumstance of degree.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

227

Contingency has several types: condition, concession, and default. Contingency particles in SH exhibit a variety of other meanings too. Consequently, their specific applications can be determined only in context. The following are among the relevant particles: yKi, B.,i rWb[]B;, l[; or qr;. A Condition circumstance is reiterated several times by God as a response to Abraham’s mediation on behalf of Sodom. (229) Gen 18.28 h&t~vj't B~j&m]V> a#t-K"l-h*u'r will [you] ruin on account of five the whole city? CContingency: Process: doing Goal condition (230) 1 Kgs 15.5 w=Oa-s*r m]K)l a&v#r-x]ww*hW K)l y=m? j~Y*yw r~q B]db~r aWr]Y> h~j]TT'

[he] did not turn aside Process: doing

from everything [[he] commanded him] CLocation: place

all the days of his life CExtent:

duration

except the matter concerning Uriah the Hittite CContingency: default

In #231, the circumstance of condition is interrupted by the main components of the representation structure, being anounced by a demonstrative pronoun, and continued with an infinitive construct. Traditional grammars refer to this situation as a cleft sentence (cf. Gen 34.22). (231) 1 Sam 11.2 B=z)at a#kr)t In this case [I] will make a treaty Process: doing

l*k#m with you Goal

B]nqor l*k#m K"l-u?n y*m'n of picking out to you all the right eye.

Circumstance: Contingency: condition

Accompaniment’s two sub-types – comitative and additive – are expressed by prepositional phrases and have both a positive and a negative aspect. Without being too dogmatic about the use of particles in SH, one can still propose as Accompaniment particles tae (IBHS, 195), !mi (IBHS, 214), ~[i (IBHS, 219), and the like, at the front of prepositional phrases with a comitative function, and ~G', l[; (IBHS, 217), tx;T; (IBHS, 220), and similar particles, to head front prepositional phrases with additive function.

228

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(232) Gen 42.4 w=a#t-B]ny*m'n a&j' yos}p Oa-v*l~j y~u&q)b a#t-a#j*yw but Benjamin the brother of Joseph did not send Jacob with his brothers CAccompaniment: Goal Process: Actor doing comitative (233) Gen 4.25 K' v*t - l' a$Oh'm z#r~j a~j}r T~j~t h#b#l gave me God another seed instead of Abel CAccompaniment: Process: Recipient Actor Goal doing additive

A very unusual situation arises in clauses with multiple Subjects. When the Process is realized by a plural verbal form, the Accompaniment circumstance will be of the additive type, whereas an Accompaniment circumstance in a clause with a Process realized by a singular verbal form will be of the comitative type. (234) Exod 35.22 w~Y*b)aW h*a&n*v'm u~l-h~N*v'm K)l-n=d'b l}b Came (pl) men as well as women, every generous heart CAccompaniment Process: Actor Attribute happening : additive (235) Josh 10.15 w~Y*v*b y=hov%~u ¨ w=k"l-y]cr~a}l u]Mo ¨ returned Joshua and all Israel with him CAccompaniment: Process: Actor Actor happening comitative

In #236, the Accompaniment circumstance is realized without the preposition, preferring a regular conjunction instead. This circumstance is best explained as a sentence with two coordinated clauses, of which the second is without the verb (ellided), but the verb of the previous coordinated clause applies to it too. (236) 1 Sam 17.11 w~Y]vm~u v*aWl heard Saul Process: sensing Senser

w+k"l-y]cr*a}l a#t-D]br? h~PP=l]vT' h*a}L#h with all Israel those words of the Philistine CAccomp: Phenomenon comitative

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

229

Role has sub-types – guise and product. The preposition that displays a preference for Role circumstances in SH is l., labelled by IBHS as ‘l*m#d of purpose’ (p. 209). (237) Judg 14.2 w=u~TT> q=jW -a ot>h now [you] take her Process: Goal doing

L' l=a]V> to me as wife Recipient CRole: Guise

(238) 2 Sam 19.3 w~T=h' h~TT=v%u> B~Yom h~hWa l=a}b#l l=k"l-h*U*m became deliverance in that day into mourning for all the people CLocation: Time CRole: CAccompaniment: Process: Actor doing Product Additive

Matter is a Circumstantial type that describes, is referred to, is narrated about, etc. One formula that appears very frequently in the HB is l[; hZ,h; rb'D'h; or variants (IBHS, 218). (239) Gen 24.9 w~Y]V*b~u lo [he] swore to him Process: saying Receiver

u~l-h~DD*b*r h~ZZ#h concerning this issue CMatter

Angle is a Circumstantial type that describes the origin of an information (source), or the position held by somebody on the matter (viewpoint). The quotation of sources by the biblical documents follows several paradigms, in which the prepositions B., K., l. and la, are given prominence. (240) Exod 6.19 a}L#h m]vP=j)t H~L}w' l=t)l=d)t*m these [are] the clans of Levi according to their genealogies CAngle: source Identified Identifier

Although a different meaning is possible, treating the genealogies as a result or a means to arrange the information, the current interpretation seems more plausible to us. (241) Exod 25.40 w~u&c}h B=t~bn't*m a&v#r-a~TT> m*ra#h B*h*r [you] do according to the model [[which you witnessed on the mountain]] Process: CAngle: source doing

230

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(242) Josh 8.34 q*r*a a#t-K"l-D]br? H~Tor> h~BB=r*k> w=h~Q=l*l> [he] read all the words of the Law, both blessings and curses, Process: Verbiage saying

K=k"l-h~KK*tWb B=s}p#r h~Ttor> according to all written in the Book of the Law CAngle: source

More rarely, the preposition la, can be used to construe source (IBHS, 194). (243) Josh 17.4 w~Y]TT}n l*h#m [he] gave them, Process: Recipient doing

a#l - P' yhwh according to the mouth of the Lord CAngle: source

n~j&l> [...] a property Goal

The other sub-type of Angle circumstantial – the viewpoint – is frequently marked in the HB by means of a prepositional phrase that has at its Head ynEy[eB. (‘in the eyes of ...’). The clauses that host such circumstantial elements are either material clauses (√!tn, √hfh) or attributive clauses realized by stative verbs. The following realization of a material clause expresses the idea that God made sure that his people’s perception of Egypt would be favourable (#244). (244) Exod 11.3 w~Y]TT}n yhwh a#t-j}n h*u*m B+u?n? m]xr*y]m granted Lord favour the people in Egyptians’ view CAngle: viewpoint Process: Actor Goal Client doing

In another context, the circumstantial viewpoint refers rather to the nominalized adjectives as qualifiers. Therefore, the current interpretation is less probable. (245) Deut 12.28 K' t~u&c#h h~Fob w=h~Y*v*r B=u?n? yhwh a$Oh#yk* [you] will do good and righteous in the view of Lord your God CAngle: viewpoint Process: doing Goal

The use of the Semitic phrase ‘in the eyes of ...’ should not be generalised, because it can easily play the role of a Senser in a mental clause, whose Process is realized by a stative verb expressing a moral judgment. Therefore, the second interpretation of #246 is to be preferred.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

231

(246) Gen 34.18 w~Y'f=bW d]br?h#m B=u?n? j&mor Wb=u?n? v=k#m B#n-j&mor seemed good their words in the view of Hamor and to Shechem son of Hamor CAngle: Viewpoint Process: being Carrier Process: sensing

Phenomenon

Senser

5.3.6. Ergativity as an alternative interpretation to transitivity One of the features of English grammar that ergativity enables us to understand is the matter of causality. SFG introduced a new function to take account of the executive role, namely the Initiator in a material clause, the equivalent to Attributor in an attributive clause, and to Assigner in an existential clause (cf. 4.3.7.). This appears to be problematic also for grammarians of SH, since there is no definitive solution to the matter concerning the system of verbal stems. Although IBHS notices the difficulty, its solution remains within the limits of the traditional approach, deriving all stems from the simpler and primitive Qal (p. 358). This view, asserts Siebesma (1991, 6), was shared by all traditional-type grammars, during the last century and a half in relation to the seven main conjugations/families/stems of SH: q~l, n]pu~l, P]u}l, P%u~l, h]pu'l, h"[u~l, h]tP~u}l. By promoting the independence of each stem individually, scholars like Ernst Jenni (1969) offer an alternative interpretation based on Akkadian grammar. Goetze (1942, 5-6) inspired this connection, by noting that, in Akkadian, D-stems correspond to stative verbs. He identified three main groups of stative verbs: (a) durative statives denoting an inherent quality of an entity, (b) perfect statives denoting a condition resulting from the subject’s own action with reference to another entity, (c) and passive statives denoting a state of affairs resulting from another person’s action. Their equivalent D-stem forms add agency to the respective stative verbs: (a) durative D-stem denotes putting an entity in the condition (having the quality) indicated by the stative, (b) perfective D-stem denotes making somebody have something, (c) passive D-stem denotes putting an entity in the state described by the stative. This interpretation is preferred also by von Soden (1969, §88c), who suggested that stative verbs and the D-stem are interrelated, proposing as the main meaning of the D-stem the factitive function of their stative Gstem counterparts. From then on, Siebesma (1991, 7) notes that the idea was adopted by the large majority of Akkadian grammars. Contrary to the

232

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

traditional view of the SH stem system, Jenni (1969, 68-9) proposes a bicameral system. Group I would include stems that focus on the action, therefore called actualis, such as Qal, Ni, Hi, Ho. Group II would consist of stems that focus on the result of the action, therefore called resultative, such as Pi, Pu, Hit. Between the two groups lay verbal forms from the category of the Qal passive participle. Joüon & Muraoka (1991, 124) promotes a stem paradigm in which Ni stays as the reflexive voice counterpart of Qal, a similar role that Hit plays in relation to Pi as part of the ‘intensive’ stem group. The Passive role is played by the Qal passive participle in the ‘simple action’ stem group, as Pu does within the intensive action group and Ho in the causative action group. They admit that Pi’s function is the most elusive of the stems, admitting that they are not at ease with its traditional qualification as ‘intensive’and with the tendency to give its exact meaning (p. 154-5). According to Jenni, Ni focuses on the event with less weight being given to the subject or the agent of the action, but both Qal and Ni give more weight to the action (Jenni 1969, 62-4). A similar interest in the action of the verb is credited to Pi, but only when corresponding to an intransitive Qal. In all the other cases, the Pi focuses on the result as opposed to Qal and Ni, which focus on the action (Jenni 1968, 275). On the basis of the German correspondence with machen/lassen, Jenni (1968, 34-6) proposes that Hi conveys a causative meaning for Pi, having the object as the logical subject of the event described by the verb. Then, Ho functions as passive for Hi, Pu stays as passive for Pi, with Hit as its reflexive. Siebesma (1991) continues in the same vein. Although he applies a similar method to his survey of the Hebrew Ni, Siebesma proposes more of a comparative approach by judging the Ni against other verbal forms with passive meaning such as Pu and Ho. Ni continues to be seen to be in direct opposition to Qal and not to Pi or Hi. Siebesma defers to the theory according to which Ni-y]qf)l and Pu-q*f~l, preferred verbal forms in both cases, evolved from Qal Passive Participle (Siebesma 1991, 171-2). Sinclair’s application of valency to SH verbs focuses on a verb’s valence, a category that ‘refers to the number of noun phrases that are normally constructed with a specific verb in its basic form excluding optional noun phrases such as adverbials which are not required by the verb’ (1991, 64).1 It has been noted that valence increases when verbs become causative (in Pi or Hi), that is the number of objects a given verb He derives his terminology and methodology from D. J. Allerton’s Valency and the English Verb (London/New York: Academic, 1982). 1

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

233

can take increases due to the transformation of the subject into a new object. This subject turned object is agentive, that is, it is ‘the doer’ of the verb’s action, and is cross-linguistically introduced by prepositions or their equivalents (Heb l, b, or !m). Moreover, with the exception of n*t~n, doubly transitive verbs already express a causative notion in Qal (Sinclair 1991, 65). Occasionally, transitive forms of verbs are the causative reflex of their intransitive counterpart (e.g. h[r, Sinclair 1991, 66).1 We suspect that a complementary transitive/ergative interpretation of the verbal forms construed as Processes in clauses would provide a clearer understanding of these phenomena. It is generally accepted that there are few roots that exhibit verbal forms for each of the seven stems such as the paradigmatic √ljq proposes at the theoretical level.2 It is most likely that individual roots prefer a specific set of conjugations, sometimes for cryptic reasons. At times, they complete the stems they lack with the verbal forms offered complementarily by other roots,3 or even by different verbal forms within the limits of the same root.4 Multiplicity of stems does not necessarily imply a variety of distinct meanings. This tendency of the SH 1 Most of the article deals with the matter of object ellipsis, though (pp. 68-74), or with cases of normally transitive verbs occurring without objects (pp. 74-6), creating the illusion that the valence of certain Hebrew verbs is smaller than in reality. 2 Among the exceptions one can mention: √l[p or √har. 3 e.g., the lexical verb ‘to seek’ is realized by two roots in SH: √vrd with forms only for q~l and n]pu~l, and √vqb with forms only for P]u}l and P%u~l. Both roots appear together as pair words in poetry (cf. 1 Ch 16.11). Similarly, ‘to teach’ can be expressed by various roots. √dml and √@la prompt the meaning ‘to learn’ in q~l, but in P]u}l they convey the meaning ‘to teach’. The former has also forms for P%u~l, but otherwise they lack representation across the seven SH stems. There are other roots, nonetheless, which enlarge the verbal paradigm of the lexical verb ‘to teach’ with h]pu'l forms, such as √hry, √rhz, or even √[dy. It is noted that the former has also forms for q~l and n]pu~l (meaning ‘to shoot’) whereas the latter has forms for all stems. The highest represented roots – √hry and √dml appear together in Psalm 25.8-9. The association of these two roots is also noted by other scholars (e.g., Jenni 1968, 119-22). 4 Bergsträsser (1929, §16d) noted this in relation to the complementarity of their respective n]pu~l and h]tP~u}l forms deriving from the following verbal roots: √rrb, √lag, √akd, √amj, √hsk. Jenni (1968, 122-3) also notes the complementarity of the following synonymous verbs: √jlp (qal ‘to escape’, pi. ‘to deliver’), √jlm (ni. and pi. ‘to escape’), √[vy (ni. and hi. ‘to deliver’), √lcn (ni. ‘to deliver oneself’, hi. ‘to deliver’), √rz[ (qal ‘to help’, ni. ‘to be helped’).

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

234

verbal roots may be tentatively explained by the diachronic process of language transformation.1 Take, for example, the root of the denominative verb √vdq (to be consecrated). Although Ho is not accounted for at all in the HB,2 the other stems are represented as follows: Pi (45 occurrences)3, Hit (22 occurrences)4, Hi (22 occurrences)5, Qal (11 occurrences)6, Ni (8 occurrences)7, and Pu (4 occurrences)8. We shall consider several passages that make use of finite verbal forms derived from √vdq. (247) Exod 30.29 w=q]DD~vT* a)t*m [you] consecrate them Theme Rheme Fin/ Mood

Pred

Compl

Residue

Process: doing

¨ [...] ¨ K"l-h~N)G}~u B*h#m y]qD*v ¨ [...] and everything touching them will be holy Marked Theme Rheme Subject Mood

Goal

Carrier

Medium

Medium

Fin/

Pred Res

Process: being

Sample (247) displays two clauses, the former of which is realized by a Pi, and the latter by a Qal. At the thematic level, the structure of the second one is defamiliarized by means of a Marked Theme conflated with the According to one of these theories, the D-stem evolved from a durative paradigm of y]qf)l in q~l, following the paradigm *yaqattalu (Meyer 1972, 40). 2 Another similar example is √$rb. 3 Gen 2.3; Exod 13.2; 19.10, 14, 23; 20.8, 11; 28.41; 29.1, 27, 33, 36, 37, 44 (2x); 30.29, 30; 31.13; 40.9, 10, 11, 13; Lev 8.10, 11, 12, 15, 30; 10.3; 16.19; 20.8; 21.8 (2x); 22.32; 25.10; Num 6.11; 7.1; Deut 5.12; 32.51; 1 Sam 16.5; 2 Ch 7.7; 29.17 (2x); Neh 3.1 (2x); 13.22; Job 1.5; Jer 6.4; 17.22, 24, 27; 22.7; 51.27, 51; Ezek 20.12, 20; 36.23; 37.28; 38.16; 44.19, 24; 46.20; Joel 1.14; 2.16; 4.9; Micah 3.5. 4 Exod 19.22; Lev 11.44; 20.7; Num 11.8; Josh 3.5; 7.13; 1 Sam 16.5; 2 Sam 11.4; 1 Chr 15.12, 14; 2 Chr 5.11; 29.15, 34; 30.3, 15, 17, 24; 31.18; 35.6; Is 30.29; 66.17; Ezek 38.23. 5 Lev 27.14; Num 3.13; 8.17; Josh 20.7; Judg 17.3; 2 Sam 8.11 (2x); 1 Kgs 9.3, 7; 2 Kgs 12.19; 1 Chr 18.11; 26.26, 27; 30.8; 36.14; 2 Chr 7.16, 20; 29.19; Neh 12.47; Jer 1.5; 12.3; Zeph 1.7. 6 Exod 29.21, 37; 30.29; Lev 6.11, 20; Num 17.2, 3; Deut 22.9; 1 Sam 21.6; Is 65.5; Hag 2.12. 7 Exod 29.43; Num 20.13; Is 5.16; Ezek 20.41; 28.22, 25; 36.23; 39.27. 8 2 Chr 26.18; Ezrah 3.5; Is 13.3; Ezek 48.11. 1

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

235

Subject. At the experiential level, other notable dissimilarities become evident. The first one derives from the complementary transitive/ergative interpretation of the clauses. In the case of the first clause, the unfolding Process has an Initiator/Agent, even though it is implicit (you). Therefore, ergatively speaking, it is effective. As for the second clause, the only explicit Participant that contributes to the unfolding of the Process is the Carrier. Its Attribute has been incorporated by the verb stem (q~l in this case). Since it does not have agency, the second clause is understood as middle. Besides, although built on the same root, the two clauses differ in terms of Process types, the first – doing, the second – being, a difference that is promoted by the stem. If, from the Pi stem of √vdq, one assumes a complex ritual, then it is right to accept the verb that realizes the Process of the first clause as material. The distinction between the Actor and Goal is preserved in the case of Hi stem too, as in #248. There, a receptive reconstruction of this clause – ‘every firstborn in Israel is consecrated to me by me’ – allows a more transparent view of the Participants. The person of God retains the functions of Agent and Beneficiary at the same time, whereas ‘every firstborn of Israel’ functions as Medium in an operative clause. (248) Num 3.13 h]qD~vT'

[I] consecrated Theme Rheme Fin/ Mood

Pred

l'

to me

k"l-B=kor B=y]cr*a}l m}a*d*mu~d-B=h}m>

every firstborn in Israel

from human to the beast

Adjunct

Complement

Adjunct

Adjunct

Recipient

Goal

CExtent:

CExtent:

Beneficiary

Medium

Location

Location

Residue

Process: doing

When a Hit is used instead, the distinction between the ergative and transitive interpretation of the experiential structure is completely obstructed by the verbal stem, in that the Actor and Goal completely overlap, and Medium and Agent are superimposed.

236

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(249) Num 11.18 h]tq~DD=vW consecrate yourselves Theme Rheme Finite/

Predicator

a~TT#m w~a&j?k#m you and your relatives Subject

Residue Mood Process: doing

Actor Agent

Since the verb in Hit lacks the Complement, there is a tendency to consider it intransitive. This conclusion is only partially true, because the Complement is identified with the Subject, so that the same entity that performs the action is also its object. At the experiential level, Goal/Medium is implicit but identical in person and number with the Actor/Agent. This clause is different, though, from other operative clauses by promoting Agent and Medium as the same person and by having the Medium understood from the verbal form. Probably the best solution to the problem of reflexive verbal forms is, following Bateman (2000), to add another category to those generated by the system of AGENCY. Thus, we should have middle transitive clauses alongside middle and effective clauses. If all are described by means of types of Process (Bateman 2000), then middle clauses have intransitive verbs, effective clauses have transitive verbs, and middle-transitive clauses have reflexive forms. In order for this new presentation of the system network of AGENCY to do justice to the fact that reflexive verbal forms possess Agency (similar to effective operative clauses), one more thing is needed. Because the Agent/Subject is the same person as Medium/Complement (implicit), we distinguish the Agent of the middle-transitive clause from the Agent of the effective operative clause by renaming the former ‘Reflexor’. Passive stems – Ni, Pu, and Ho – where present, bring the category of voice to the ergative interpretation of the system of transitivity.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

(250) Exod 29.43 w=n]qD~v [it] will be consecrated Theme Rheme Finite/

Predicator

Mood

Residue

237

B]kb)d' by my glory

Process: doing

Adjunct

Initiator Agent

This receptive clause displays a Subject conflated with Medium (implicit – ‘the place of the Tabernacle’). Nonetheless, ‘God’s glory’ does not function as Actor, because it does not suffer any transformation, in particular not the one described by the Process. It rather serves as Initiator. (251) Gen 12.15 w~TT%Q~j h*a]V> was taken the woman Theme Rheme Fin/

Pred

Mo-

Subject

B?t P~ru{h [to] Pharaoh’s house Adjunct

-od Residue

Process: doing

Actor

CLocation:

Place

Medium

Another example (#251) of a receptive clause (Ni verb) has room for an explicit Subject/Medium without specifying who plays the role of the Initiator, since they are mentioned in the context, i.e. Pharaoh’s officials. The clause in #252 makes use of a Ni verbal form, but this time the Subject functions as Beneficiary ergatively, whose Agent is an Adjunct, thus a prepositional phrase.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

238

(252) Gen 9.11 w=Oa-y]KK*r}t K"l-B*v*r uod and will not be cut off all flesh again Theme Rheme Fin/

Pred

Mo-

Subject

m]M? H~M~BWl by the waters of the flood

Adjunct

Adjunct

CExtent:

Initiator

-od Residue

Process: doing

Client Beneficiary

time

Agent

We conclude our section on SH stems with a discussion of the verbal realizations displayed by √$rb. Since in Qal this root does not have other form than that of passive participle (%WrB'), its most regular stem is Pi and it tends to produce effective operative clauses. (253) Gen 1.22 w~y=b*r#k a)t*m blessed them Theme Rheme Fin/ Mo-

Pred

Compl

Residue

Process: saying

a$Oh'm God Subject -od

Target

Sayer

Medium

Agent

Even though the same root is used, the verbal form of the following clause (Ni) does not require the presence of an Agent. Therefore, the clause is middle.1

1 Jenni (1992, 164) considers that this occurrence of the preposition b has the role of formal legitimation, being an abbreviation of the formula ‘in the name of ...’. If that is the case, then we should consider it as a circumstance of Manner: quality. It could also play the function of the Manner: means circumstantial (‘through him’).

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

239

(254) Gen 18.18 w=n]br=kW bo K)l Goy? h*a*r#x will be blessed in him all nations of the earth Theme Rheme Fin/ Pred Adjunct Subject MoResidue -od C Process: behaving Manner Behaver Medium

Next, we have the Hi verbal form in another Genesis context and in an effective operative clause. This time, Agent/Subject is only implicitly present. (255) Gen 24.11 w~Y~br}k h~GG=m~L'm m]jWx l~u'r #l-B=a}r h~M*y]m [...] [he] made kneel the camels outside the city beside a well of water [...] Theme Rheme Fin/ Mood

Pred

Complement

Adjunct

Adjunct

Actor Medium

CLocation:

CLocation:

Residue

Process: material

place

place

Example (256) displays a middle-transitive type of a material clause. As expected in a case like this, the Subject plays the role of the Reflexor from an ergativity perspective. (256) 1 Sam 23.19 h&Oa d*w]d m]sT~TT}r ¿is not David hiding himself Marked Theme Rheme Adj

Subject

Mood

Fin/

u]M*nW among us

Pred

b~M=x*dot B~j{rv> ... in the strongholds of Horesh ... ?

Adjunct

Adjunct

CLocation:

CLocation:

Residue Actor Reflexor

Process: doing

place

place

From the preceding data, it appears to be safer to construe the function and meaning of each verbal form (including the reference to its stem) in the given context of its respective clause. We conclude this section

240

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

with a chart that integrates the networks of the systems for Agency, Ranging, and Voice, insofar as were able to confirm it in the light of the data (cf. next page).

5.4. LOGICAL S TRUCTURE Along the experiential meaning, the ideational metafunction also realizes logical meaning. This refers to the relationship among adjacent clauses. According to SFG, adjacent clauses are organized in clause complexes on the basis of two main systems, namely interdependency and logico-semantic type. Their description, as prompted by the SH prose, and the textual evidence represents the object of study in this chapter. At it has been already discussed (cf. 3.1.1.), the Masoretic tradition marks the end of a clause complex by the accent sop P*sWq, a formal orthographic signal anticipated by P*sWq, a m#t#g-type accent placed under the last accented syllable of the verset.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

241

5.4.1. Interdependency (taxis) In Biblical studies, text linguists are particularly fond of the double function an individual clause has both at the level of sentence and text. For example, Niccacci proposed a general rule for distinguishing primary clauses from secondary ones, on the basis of the place of verb forms in the clause: if the verb occupies the most salient position, the clause is primary (Niccacci’s ‘main’ sentence); second position verb forms indicate secondary clauses (Niccacci’s ‘dependent’ sentence). He proposes a further distinction, noting that those clauses that can stand alone in the text are also ‘independent’ (Niccacci 1988, 14). Although Niccacci has some good insights into SH syntax, to which we will refer constantly, we prefer to remain within the limits of SFG, in terms of method and terminology. In the SFG, the clause complex is the realization of meaning at the ideational: logical level. According to the Systemic Theory, the relation between verb form and the genre of the text as promoted by text-linguistics is the legitimate expression of the realization principle that links context with content (cf. Stratification in 4.1.). Thus, culture is realized in meaning (semantics) and meaning is realized in wording (lexico-grammar). 5.4.1.1. Parataxis SH displays chains of clauses connected by parataxis in several textual environments. According to text linguists such as Longacre, the verbal form that supports paratactic relations between adjacent clauses varies according to the type of text: w~Y]qf)l for narrative discourse (Longacre 1989, 81; cf. Niccacci 1990, 29-32), w=q*f~l for prescriptive and predictive discourse (Longacre 1989, 107), volitives for hortatory discourse (Longacre 1989, 121). Similarly, q)f}l appears to be the main verb form used to relate background activities in all these types of discourse.1 Terminologically, Niccacci prefers to speak of ‘narrative’ and ‘discourse’ as the two main types of prose texts, as opposed to poetry.2 ‘Narrative’ stands for third-person stories, having w~Y]qf)l as the backbone verb form, whereas ‘discourse’ represents direct or indirect addresses with y]qf)l as the main verb form (Niccacci 1988, 10; 1 Such a shift from one verb form to a similar verb form in the course of developing a text is called by Niccacci, following Weinrich, ‘a homogenous tense shift’ (1990, 111-2). 2 Although Watson translated Niccacci’s ‘discorso’as ‘discourse’, other Biblical scholars prefer the term ‘comment’ (e.g., Talstra 1991; 1995).

242

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

cf. Niccacci 1989, 13-17 and 29).1 To these, he also adds ‘narrative discourse’ a first-person narrative, i.e., an autobiographical approach to history, whose foreground is realized again by w~Y]qf)l verb forms (Niccacci 1990, 102-9). Narrative is, by far, the most thoroughly studied text type. Typically, a chain of w~Y]qf)l verb forms constrains the reader to see clauses tied together by parataxis. Since w~Y]qf)l is an initial verb form in narrative, i.e., it always appears at the beginning of a clause (Niccacci 1990, 35), the division of a clause nexus into its respective clauses is relatively simple; all that is necessary is to follow the w~Y]qf)l verb forms in the sequence.2 (257) Exod 34.5 1 w~Y}r}d yhwh B#u*n*n a¨ descended Lord in the cloud 2

w~Y]ty*X}b u]Mo [he] stood with him

3

w~Y]qr*a B=v}m yhwh c¨| [he] called upon the name of the Lord.

v*m b¨ there

In Hatav’s sample narrative text, approximately 39 per cent of all verb forms identified are of the w~Y]qf)l type (2445 occurrences), with a lower 1 Although Niccacci’s studies on the Hebrew verbal system antedate Longacre, the latter further detailed text typology. For Longacre, discourse is a general term that encompasses narrative, predictive and hortatory texts. Dawson (1994, 94-9) narrows down these categories and produces four large types of discourse, each having two subcategories: narrative (prediction, story), procedural (how-to-do-it, how-it-was-done), behavioural (exhortation, eulogy), and expository (futuristic essay, scientific paper). Even though not all of these are represented in the HB, the distinction is welcome. These eight categories are distinguished by the three parameters that define them, namely Agent Orientation, Projection, and Contingent Temporal Succession. Unfortunately, none of the previous theories makes room for poetic texts as a distinct category. 2 We endorse Cook’s idea that w~Y]qf)l is not necessarily connected to temporal succession, although it expresses it with priority (Cook 2004, 261-4). Cook prefers the term ‘foreground’ to explain the ordo naturalis characteristic of narration, that implies a temporally successive interpretation, a default (unmarked) mode of relating events in narrative, expressed by w~Y]qf)l verb forms. It should be noted that his understanding of foreground differs from that of Heimerdinger’s (1990), viz, as special salience (cf. Cook 2004, n. 27). Therefore, we consider that a hypotactic relation describes, with more neutrality, the ability of w~Y]qf)l verb forms to link in narrative chains.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

243

distribution for q*f~l (19 per cent) and y]qf)l (18 per cent) verb forms (cf. Hatav 1997, 31). Most w~Y]qf)l verb forms are part of a narrative sequence (94 per cent) and sequentiality is interpreted by Hatav (1997, 56f) on the time-line. Niccacci proposes a further distinction, that between the narrative w~Y]qf)l and the continuation w~Y]qf)l. Although they both build up the foreground of a text, the former appears right from the beginning of a narrative text, whereas the latter is preceded by other verb forms or constructions and continues whatever the antecedent construction prompted (Niccacci 1990, 177-80). In contrast with the narrative sequence, smooth-running long prescriptive chains are hard to find, other clause types frequently fragmenting it. When encountered in prescriptive texts, w=q*f~l comes first in the sentence and builds up the backbone of such texts (foreground), thus being the equivalent of w~Y]qf)l in narrative (Niccacci 1990, 183; Hatav 1997, 56f). Most w=q*f~l verb forms (over 93 per cent) are part of a non-narrative sequence (Hatav 1997, 56f). When present, a regular clause nexus with w=q*f~l verbal forms will look like the prescriptive text in #258. (258) Exod 28.9 1 w=l*q~jt* a#t-v=T? a~b=n?-v)j~m a¨ [you] take two stones of onyx 2

Wp]T~jT* u*l?h#m [you] engrave upon them

v=mot B=n? y]cr*a}l b¨ the names of the sons of Israel

When w=q*f~l appears in other text types, it is always preceded by another verb form of construction that imposes on it its linguistic attitude, perspective and prominence. For that reason, Niccacci labelled such a w=q*f~l a ‘continuation form’ (Niccacci 1990, 82f, cf. 182-6).1 The correspondence between w~Y]qf)l and w=q*f~l is best seen at work when prescriptive texts are retold as third person narratives. This is exactly the case with the long passage that spells out the divine regulations Cook (2004, 267) agrees that w=q*f~l verb forms appear to built up the foreground of procedural and instructional discourse, such as Leviticus 1-5 and Exodus 25-30, respectively, as quoted by Longacre. In both discourse types, temporal succession is implied. The only difference between the two discourse types resided with the verb form that introduced the discourse, i.e., a protasisapodosis construction, and an imperative construction, respectively. This distinction may not always be followed. 1

244

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

concerning the building of the Tabernacle (Exod 25-30), then recounted for later (Exod 36-39), of which account Niccacci analysed an excerpt (1990, 85). Such transformations from ‘instruction’ to ‘execution’ or from ‘command’ to ‘fulfilment’ are frequent in prophetic narratives (Niccacci 1990, 86-8). Similarly, w=q*f~l verbal forms may be seen in paratactic relation as part of larger predictive discourses, again as part of the ‘foreground material’ (Cook 2004, 266). (259) 1 Sam 10.3 1 w=j*l~pT* m]V*m w*h*l=a> [you] will pass from there and onwards

a¨

2

Wb*at* u~d-a}lon T*bor [you] will come to the oak of Tabor

3

Wm=x*aWn V*m v=Ov> a&n*v'm c¨ [you] will find there three men

b¨

Exodus 34.10-26 is an exceptional prescriptive text in which the main line is built from (Waw)-X-y]qf)l constructions, even though they were initiated by imperatives like v=m*r (v. 11) and h]V*m#r (v. 12). A sample will illustrate the general outlook of the text. (260) Exod 34.13 a¨ 1 K' a#t-m]zB=j)t*m T]T)xWn their altars [you] pull down 2

w=a#t-m~X}b)t*m T=v~B}rWn b¨ and their pillars [you] break

3

w=a#t-a&v}r*yw T]kr)tWn c¨| and his Ashera [you] cut off

A descriptive clause nexus prefers nominal clauses instead. (261) 1 Sam 14.5 1 h~v}n h*a#j*d m*xWq m]X*pon mWl m]km*v a¨ one cliff [was] narrow to the north in front of Mikmash 2

w=h*a#j*d m]N#g#d mWl G*b~u b¨| and the other to the south in front of Geba.

Typically, volitive verb forms can link together as the backbone of a hortatory discourse. Other verb forms, such as w=q*f~l and w=y]qf)l, can interfere in a hortatory chain (Niccacci 1990, 88-95, 187). The short

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

245

sequence that follows is built up by an imperative, a hortative and a negated y]qf)l.1 (262) Exod 9.28 1 h~uT'rW a#l-yhwh [...] a¨ [you] supplicate to the Lord 2

w~a&v~L+j> a#tk#m b¨ [I] release you

3

w=Oa t)s]pWn l~u&m)d c¨ and not [you] continue to stay

Currently known as conjunctives, the most common SH linker is w*w. We have already noticed it with this function, but connected to the verb form it serves as consecutive w*w, fronting both q*f~l and y]qf)l verbal forms. There are also occasions when the simple conjunction is employed to render the paratactic connection. Since the traditional category of ‘nominal clauses’ has a variety of realizations in SH, one would expect other Process types to be present. The following example displays an imperative (clause a) and three y]qf)l verb forms (clauses b, c, d). Although the last three clauses stay together, having the same Subject (we = I and the boy), the primary clause of this nexus also connects with the first clause of this verse. Therefore, one can say that the second nexus is nested within the first. (263) Gen 22.5 1 v=bW - l*k#m P)h u]m- h~j&mor a¨ remain you here with the donkey 2

1

w~a&n' w=h~N~u~r n}l=k> u~d-K)h b¨ but I and the boy will go up to there

2

w=n]vT~j&w#h and[we] will worship

3

w=n*vWb> a&l?k#m d¨| and [we] will return to you

c¨

1 With the exception of imperatives which are negated by la;, all the other volitives cannot be negated; in order to produce a negated volitive (other than imperative), SH prefers to use an y]qf)l with aOl. Niccacci (1990, 78-81) even recommends as the distinctive sign of a volitive y]qf)l, its use in a volitive chain opened by a volitive (e.g., Gen 43.10-14; Exod 10.24; 19.20-22; 25.2-3).

246

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

5.4.1.2. Hypotaxis Many particles play an important subordination role in SH. Alongside B., ~ai, and yKi with various roles, one can also find K., ![;y:, etc. Other prepositions commonly involved in opening circumstantial clauses are rva-d[, ynpl, ~rjb, hnavrb for anteriority, and yrxa, rva-yrxa, etc. for posteriority (cf. Joüon 1991, 625). hT'[;(w>) can also be added, a particle introducing result or conclusion (Niccacci 1990, 101). Several instances of hypotaxis call for our attention in SH. Secondary clauses in a hypotactic clause nexus can function as temporal, conditional, causal, reason/purpose or result clauses. Temporal clauses can be realized by various means, but frequently they are introduced by prepositions such as ~ai, yKi, ~r,j,, B., dA[, etc. Among the possibilities for producing a temporal clause, Niccacci also added nominal clauses (1990, 138).1 When it precedes a clause, not necessarily a verbal one, w~y=h' proves to be more than the main indicator of narrative, marking a dependent circumstantial clause (#264, cf. Niccacci 1990, 49-50).2 Waw-X-q*f~l constructions may also prove useful to subordinate clauses (#265, cf. also Joüon 1991, 624). (264) Judg 11.5 w~y=h' K~a&v#r-n]lj&mW b=n?-u~Mon u]m-y]cr*a}l a¨ β when fought the Ammonites with Israel w~Y}l=kW z]qn? g]lu*d l*q~j~t a#t-y]pT> m}a#r#x fob α went the elders of Gilead to bring Jephta from the land of Tob

b¨|

(265) Num 17.15 w~Y*v*b a~h&r)n a#l-m)v#h a#l-P#t~j a)h#l mou}d a¨ α returned Aaron to Moses to the entry of the Tabernacle w=h~M~G}p> n#u$x*r> b¨| β [when] the pestilence restrained 1 This is a generous category that includes in Niccacci’s own terminology ‘single noun clauses’(SNC), ‘compound noun clauses’(CNS) and ‘infinitive clauses’. SNC include what other grammars would consider participial clauses and existential clauses. CNS seems the most confusing term of all, since its composite character is suggested by the association of a finite verb with a prepositional phrase. 2 Niccacci (1990, 48-54) looks at temporal circumstance as if it is a coherent syntactical incident, although his examples indicate that he means both clause constituents and temporal clauses. For this reason, he is not able to see the relevance of (Waw)-X-q*f~l or (Waw)-X-y]qf)l as syntactic devices in clauses with temporal constituents.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

247

The protasis of a conditional sentence is another category of hypotactic secondary clause, frequently found in the Law books of the HB.1 (266) Exod 21.30 a]m-K)p#r yWv~t u*l*yw a¨ β if a ransom [he] set on him w=n*t~n P]dyon n~pvo K=k)l a&v#r-yWv~t u*l*yw b¨| then [he] will give ransom of his life according to all which [is] set on him

α

Reason and purpose clauses are also frequently met as dependent clauses. (267) Num 19.3 1 β K' m? n]DD> Oa-z{r~q u*l*yw a¨ because the water of cleansing was not sprinkled on him α 2

f*m}a y]hy#h unclean [he] will be,

b¨

c¨| uod f%ma*to bo still his uncleaness [is] on him

Result may describe the dependent clause of a clause nexus, mediated by the particle hT'[;(w>), sometimes in association with hNehi, described by Niccacci ‘as an adverbial expression of time with logical force’ (1990, 101). (268) Gen 31.16 K' k"l-h*u)v#r Æa&v#r h]x'l a$Oh'm m}a*b'nW Ø l*nW hWa Wl=b*n?nW a¨ α since all riches Æthat has taken God from our fatherØ ours [is] it and our sons β

w=u~TT>

ÆK)l

a&v#r a*m~r a$Oh'm a?l#yk* bØ u&c}h c¨|

therefore

Æeverything that

said

God

to you Ø

do

Nevertheless, it seems that its logical, or rather, resultative force is most frequently used to conclude an argument, by introducing a judgment or decision. When it serves this function, hT'[;(w>) appears at the head of a 1 Niccacci’s suggestion that relative clauses can function as protasis in a conditional sentence is inaccurate (1990, 131). Since the relative clause displaced a regular noun, it is better interpreted as an embedded clause. The only example of SNC that functions as conditional clause (Judg 6.13) is built upon two independent clauses connected paratactically, of which the first is an existential clause and the second an interrogative WH- one. Their conditional relationship is implied but not explicitly expressed. Cook also defends a modal function of weqatal in conditional sentences (2001, 135: example 20).

248

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

clause nexus, connecting adjacent clause nexuses. In such a case, the preferred meaning is ‘consequently’ (e.g., Gen 27.3, 8, 43; 31.30, 44; etc.). The purpose/result hypotactic clause can be realized also by the modal usage of w=q*f~l (Cook 2001, 134: sample 19a). (269) Gen 7.4 a*n)k' m~mf'r u~l-h*a*r#x a~rB*u'm yom w=a~rB*u'm l*yl> a¨ α I am going to cause it to rain upon the earth for forty days and forty nights β

Wm*j't' a#t-K"l-h~y=qWm [...] b¨|

so that I wipe out every existing thing

Temporal hypotactic clauses can be realized in SH by means of temporal particles and diverse verb forms, including infinitives, as #270 shows (cf. also Exod 34.34). (270) 2 Sam 19.31 G~m a#t-h~K)l y]Q*j α also everything [he may] take β

a¨

a~j&r? a~v#r- B*a a&d)n' h~M#l#k B=v*lom a#l-B?to b¨| after which came my lord the king in peace to his house

A similar example from SH of a temporal clause interrupting the primary clause follows. (271) Gen 13.10 K' k%L*h m~vq#h K=g~n-yhwh K=a#r#x m]xr~y]m B)a&k> x)u~r a¨| α the whole of it was watered like Lord’s garden like Egypt as you go to Zoar. β

>> b¨ before ruined the Lord Sodom and Gomorrah

The elision of Subject is frequently employed in biblical texts, being implicit from the verb form. This is why we preferred to place the proper pronoun in square brackets in translation. (272) Gen 23.7 1 w~Y*q*m a~br*h*m a¨ rose Abraham 2

w~Y]vT~jW l=u~m-h*a*r#x l]bn?-j}t b¨| and [he] bowed down to the people of the land, [that is] to the Hittites.

The whole Mood element can be missing in the secondary clause of a hypotactic nexus. In SH, especially when infinitive verb forms preceded by

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

249

l*m}d are used, but they are rather to be interpreted as clause constituents (Cause Circumstantials) rather than clauses by themselves. There are occasions, though, when the infinitive can play more of a verbal role in a clause, without any interference from a preposition. 5.4.2. Logico-semantic type: expansion Of the two logico-semantic types, we will deal first with expansion and all its options of realization, namely elaborating, extending and enhancing. The same system of symbols is maintained: = for elaborating, + for extending, and × for enhancing. 5.4.2.1. Expansion through elaborating Paratactic elaboration describes different types of relation between clauses in SH too. In the case of paratactic clause nexuses, one would expect similar verb forms in both clauses. Especially in the case of the elaboration paratactic relation, since the secondary clause comments on something already known, it may prefer a negative realization. In such cases, since w~Y]qf{l verbal forms cannot be negated (they are not to be fronted by anything), a q*f~l takes its place (#273). (273) Judg 8.28 1 w~Y]K*n~u m]dy*n l]pn? B=n? y]cr*a}l a¨ was subdued Madian before the Israelites =2

w=Oa y*s=pW l*c}t r)av*m b¨| and not repeated to raise their head.

Typically, when the author presents the same event from two different perspectives, either from two different planes (a material and a philosophical one), or from two perspectives of the same plane (e.g. two characters), a paratactic elaboration by exposition strategy is followed. (274) Judg 9.23 w~Y]vl~j a$Oh'm rW~j r*u> B?n a&b'm#l#k Wb?n B~u&l? v=k#m a¨ 1 sent God a bad spirit between Abi-Melek and the masters of Shechem =2

w~Y]bG=dW b~u&l?-v=k#m

betrayed

B~a&b'm#l#k b¨|

the masters of Shechem Abi-Melek.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

250

(275) Judg 9.40 1 w~Y]rD=p}hW a&b'm#l#k a¨ [they] chased him Abi-Melek =2

w~Y*n*s m]P*n*yw b¨| [he] fled from him

Paratactic elaboration by exemplification in narrative discourse links again two w~Y]qf)l verb forms. (276) 1 Kgs 1.3 1 w~Y=b~qvW n~u&r> y*p> [they] sought a young woman beautiful =2

B=K)l G=bWl y]cr*a}l a¨ amidst all borders of Israel

w~Y]mx=aW a#t-a&b'v~g h~VWn~M'm b¨| [they] found Abishag from Shunem

Clarification as a means to express paratactic elaboration implies a comment of some sort in the secondary clause on a constituent of the primary clause, or on it as a whole. (277) Judg 11.38 1 w~TT}l#k h'a w=r}uot#yh* a¨ went she and her friends =2

w~TT}b=K= u~l-B=tWl#yh* u~l-h#h*r'm [she] bewailed over her virginity on the mountains.

b¨|

An evaluative comment has a similar purpose to a clarification comment. The following example displays such a case with a supplementary hypotactic relation. (278) Gen 48.17 1 w~Y~r=a yos}p a¨ α saw Joseph β =2

K' - y*v't a*b'w y~d-y=m'no u~l-r)av a#pr~y]m b¨ that put his father his right hand on the head of Ephraim w~Y}r~u B=u?n*yw c¨| and [it] displeased him

Clauses can be associated into complex structures, where diverse types of relation can be evidenced. Clearly, the relation between clauses is part of an author’s interest. Clause (a) of #279 stands as the primary clause whose

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

251

secondary clause (b) reiterates the same information from a different angle (1 ˆ = 2). The third and last clause (c) of this nexus becomes more specific about the deity adopted by Israel (2 1 ˆ = 2). Numerous similar examples can be found in the biblical narratives. (279) Judg 8.33 1 w~Y*vWbW B=n? y]cr*a}l a¨ turned the Israelites [∅: from God] =2

w~Y]znW a~j&r? h~BB+u*l'm b¨ [i.e. they] prostituted to the Baals

=3

w~Y*c'mW l*h#m B~u~l B=r't l}aOh'm c¨| [e.g. they] set up for them Baal-Berit as divinity.

Lastly, SH exhibits various techniques for expressing hypotactic elaboration, of which clauses with participles and/or ~v' rv,a] are of main interest for us. Typically, the so-called non-defining relative clause, introduced by rv,a], picks up some element of the dominant clause and gives more details about it, as if a descriptive gloss (#280). When there is no direct relationship between the relative clause and its antecedent clause, we have a defining relative clause, meeting the qualification of embedded clauses (see #281 as well as 5.4.2.4.). (280) Gen 2.11 hWa h~S)b}b a}t K"l-a#r#x h~j&w'l> a¨ α it flows around all the country of Havila =β

a&v#r-v*m h~ZZ*h*b b¨| [is] there gold

(281) Judg 8.34 w+Oa z*k=rW B=n? y]cr*a}l a#t-yhwh a$Oh?nW a¨ α not remembered the Israelites the Lord our God =β

h~M~X'l aot*m m]Y*d K"l-a)y=b?h#m m]S*b'b b¨| the one who delivered them from the hand of all their enemies around

Frequently, the dependent relative clause interrupts the dominant clause, as clause β in #282 shows:

252

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(282) Exod 9.26 r~q B=a#r#x G)v}n > Oa h*y> B*r*d a¨| α only in the land of Goshen there was no hail =β

> b¨ [were] there the Israelites

5.4.2.2. Expansion through extending SH possesses both paratactic and hypotactic extension clauses, with the specific requirement of conjunctions and dependent type of clause met for the latter. First, we will offer several examples of paratactic extension clauses. In #283, the two clauses complement each other, revealing informations about Gilead’s parents, prompting the reader to infer the confused pedigree of the story’s hero. Even though the two clauses do not have similar Processes (the first is attributive, the second material), they have the same rank, thus both being independent clauses. (283) Judg 11.1 1 w+hWa B#n-a]V> zon> a¨ he [was] the son of a prostitute +2

w~Yol#d G]lu*d a#t-y]pT*j b¨| and begot Gilead Jephta.

A secondary paratactic extending clause may continue the idea of the primary clause by replacing or subtracting some information. The following clause nexus, clearly paratactic, may be understood as he killed them all except Jotham. (284) Judg 9.5 w~Y~h&r)g a#t-a#j*yw B=n? - y=r%B~U~l v]bu'ma'v u~l-a#b#n a#j*t a¨ 1 [he] killed his brothers sons of Jeru-Baal seventy men on a single rock +2

w~Y]ww*t}r yot*m B#n - y=rWB~u~l h~Q*f)n b¨|

and remained

Jotham the son of Jeru-Baal the youngest

Other techniques in SH to produce a secondary clause by variation employ a w=q*f~l verb form (cf. Lev 25.43; 1 Kgs 3.11; 2 Kgs 8.10), a waw-X-q*f~l (cf. Gen 42.10), or a particle such as ~l'Wa (cf. Gen 48.19) and (~ai) yKi (cf. Gen 15.4; 19.2; 1 Sam 8.19). Traditional grammars call this clause type ‘adversative’ (Joüon 1991, 641-2).

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

253

(285) Gen 17.5 1 w=Oa-y]Q*r}a uod a#t-v]mk* a~br*m a¨ and not will call yet your name Abram, +2

w=h*y> v]mk* a~br*h*m b¨| but will be your name Abraham.

A primary negative clause may be emptied of most of its constituents, except for the negation itself, as the following example with a waw-X-q*f~l construction shows. (286) Gen 42.10 1 Oa a&d)n' a¨ not, my lord! +2

w~u&b*d#yk* B*aW l]vB*r-a)k#l b¨| but your servants came to buy food.

As an illustration for the paratactic extending clause by alternation, consider #287, where the secondary clause presents an alternative to the primary one. Note that each of the clauses in paratactic bond has an extra clause that elaborates on the meaning of its primary counterpart, or extends on it, respectively. (287) 1 Sam 26.10 1 1 ao - yomo y*boa =2 either his day will come +2

1 +2

a¨

b¨ w*m}t and [he] will die

c¨ d¨| ao b~m]lj*m> y}r}d w=n]sP> or in the battle [he] will descend and [he] will be snatched away

Hypotactic extending clauses are not frequent in SH. One encounters, though, a significant number of secondary clauses realized by alternation. Typically, such a clause complex gives prominence to dependent clauses, which are much truncated. (288) Gen 18.21 +α w+a]m - Oa a¨ and if not β

a}d*u> b¨| [then I] will know

Two more constructions render variation in SH; traditional grammars call this type of clauses ‘exceptive’ (Joüon 1991, 642-3). First, yKi sp,a, appears

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

254

in the dependent clause following the positive statement of the dominant one as in #289.1 Second, ~ai yKi introduces the dependent clause, if the dominant one is realized as a negative statement, as in #290.2 In addition, yTil.Bi can appear with both affirmative and negative dominant clauses (IBHS, 643). (289) Judg 4.9 h*l{k a}l}k u]M*k a¨ α [I] will come with you +β

a#p#s K' Oa t]hy#h T]pa~rT=k* U~l-h~D#r#k a&v#r a~TT> hol}k except that will not be your honour on the way which you walk

b

¨|

(290) 1 Sam 21.5 a?n l#j#m j{l a#l-T~j~t y*d' a¨ α there is no ordinary bread under my hand, +β

K' a]m-l#j#m q{d#v y}v b¨| except holy bread there is.

A more complex structure is involved in Moses’ intercession for Israel (#291). Note how various strategies of hypotactic extension are used. Clause (b) is dependent on (a) and in a hypotactic addition relation, having both the Process and the Goal elided. The same clause (b) is dependent on (c), as part of a nexus where alternation defines the relation between (b) and (c). Overall, we have only extension and hypotaxis. (291) Exod 32.32 w=u~TT> a]m T]cc*a h~F*at*m a¨ α either [you] will take their sin +β

+β

w=a]m-a~y]n b¨ or nought [∅: take their sin]

α

c¨| m=j}n' n*a m]S]pr=k* Æa&v#r K*t*bT*Ø [then] blot me, please, from the book Æwhich you have writtenØ

Expansion by variation by means of hT[;;(w>) as in #292 should also be considered.3

Cf. also Gen 43.3; Num 13.28; Deut 15.4; Judg 4.9; Amos 3.4. Cf. also Gen 32.27; Lev 22.6. 3 See also Num 22.34; Deut 10.22; Josh 1.2; Judg 6.13, etc. 1 2

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

255

(292) Gen 32.10 K' b=m~ql' u*b~rT' a#t-h~Y~rD}n h~ZZ#h ¨ α surely with my staff [I] crossed this Jordan +β

w=u~TT> h*y't' l]vn? m~j¬ ¨| but now [I] become two camps

5.4.2.3. Expansion through enhancing Enhancement is realized in the biblical texts with the implicit or explicit support of prepositions, conjunctions and other such particles. As is the case in English, in SH too, precise identification of the type of enhancing that best describes the secondary clause may be difficult at times. Temporal clauses, to start with, can express succession, simultaneity and anteriority in relation to their primary clause (cf. Joüon 1991, 621-7). Temporal sequence of paratactic enhanced clauses is found frequently in same type verb chains, as seen in the narrative (#293), or in verbal forms common to a given text type, such as the hortatory discourse in #294. (293) Gen 12.15 1 w~Y]raW a)t*h c*r? p~ru)h a¨ saw her the officials of Pharaoh, ×

2

w~y+h~l=lW a)t*h a#l-P~ru)h b¨ then [they] praised her to Pharaoh,

×

3

c¨| w~TT%Q~j h*a]V> B?t P~ru{h then was taken the woman to Pharaoh’s palace.

(294) Exod 8.16 1 h~vK}m B~BB)q#r a¨ rise up in the morning, ×

2

w=h]ty~X}b l]pn? p~ru)h ... b¨ then stand in front of Pharaoh

×

3

w=a*m~rT* a}l*yw c¨| then said to him

Its hypotactic counterpart makes use of various means in order to suggest simultaneity and anteriority. The most common formulae are (waw)-X-q*f~l, w~y+h', and various conjunctions. Altering the narrative sequence halts the action in order to give more attention to the weaving of

256

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

other incidents. Note how in both clauses of the following clause nexus the author made use of a (waw)-X-q*f~l structure in order to achieve simultaneous circumstance. (295) 1 Sam 9.17 × Wv=mWa}l r*a> a#t-v~aWl a¨ β when Samuel saw Saul w~yhwh u*n*hW b¨| then the Lord answered him

α

The situations where a (waw)-X-q*f~l construction expresses a circumstance for the following finite verb, in a w~Y]qf)l chain, that was just severed are less complex.1 A temporal enhancement of the dependent clause with the result of expressing simultaneity, can also be accomplished by a yhiy>w: clause.2 (296) Gen 15.12 × w~y=h' h~V#m#v l*boa a¨ β while the sun was setting w=t~rD}m> n*p=l> u~l-a~br*m b¨| a deep sleep fell upon Abram

α

Occasionally, yhiy>w: is associated with various conjunctions of anteriority, together marking the dependent clause. Only those syntactic units that qualify themselves as clauses, having a (non-)finite verb (#297) are considered here.3 (297) 1 Sam 15.9 × w~y=h' a~j&r? h}s~BBW a)to a¨ β After [they] returned it w~TT=h' y~d - yhwh B*u'r m=hWm> g=dol> m=a)d b¨| was the hand of the Lord against the town a panic very great

α

Cf. 2 Sam 10.14; 1 Kgs 11.21; Exod 11.10 in Niccacci (1990, 66). This seems to be the case with yhyw followed by temporal adjuncts in 1 Sam 16.6; 7.2, 10; 8.1; 18.6, 19; 23.26; 24.1; 25.2, 37; 30.1 (van der Merwe 1999b, 105) or with preposition k as in 1 Sam 4.18; 5.10; 9.26; 13.10; 18.1; 24.17 (van der Merwe 1999b, 106). 3 Cf. also 2 Sam 17.21; 1 Kgs 13.23, 31; 2 Chr 25.14 or 1 Sam 1.20; 5.9; 7.20; 24.6; 30.25 (van der Merwe 1999b, 106). 1 2

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

257

Alternatively, temporal paratactic clauses introduced by yhiy>w: can be continued by an adverb of time or a preposition + noun, by a preposition + an infinitive, by yKi/rv,a] + a finite verb/participle/relative clause or by a nominal clause (Niccacci 1990, 51ff).1 At times, temporal conjunctions suffice to initiate a dependent clause (cf. Josh 7.8; 24.20; Judg 11.36; 19.23). Location: time circumstantials may precede the temporal conjunction, but they should be taken as part of the dependent clause (cf. Josh 9.16; 23.1) and not as separate units (cf. Niccacci’s ‘compound noun clause’). (298) 2 Sam 19.31 G~m a#t-h~KK)l y]Q*j a¨ α even everything [he] can take ×

a~j&r? a&v#r - B*a a&d)n' h~M#l#k b~v*lom a#l-B?to b¨| after that return my lord the king in peace to his home

β

Spatial enhancement coupled with taxis produces clauses that are associated together by means of spatial reference. Paratactic secondary clauses of this type give details about other Processes that took place in the very location of the primary clause, and, in order to accomplish this, the adverb ~v' is used (#299).2 (299) Gen 2.8 1 w~Y]F~u yhwh a$Oh'm G~n - B=u}d#n a¨ planted the Lord God a garden in Eden ×

w~Y*c#m v*m a#t-h*a*d*m b¨| he placed there the man

2

A hypotactic rendering of this option in SH would make use of the relative pronoun rv,a] in association with other prepositions. We find examples with a progressive sequence (α ˆ ×β) as in #300, or with a regressive sequence (×β ˆ α) as in #301. (300) Gen 28.15 a¨ Wv=m~rT'k* α [I] will protect you ×

B=k)l a&v#r - T}l}k b¨ wherever [you] will go

β

1 2

Also 1 Sam 7.2; 8.1; 24.1 (van der Merwe 1999b, 106). Cf. also Gen 11.2; 12.8; 21.33 inter alia.

258

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(301) 2 Sam 7.7 × B=k)l a&v#r-h]th~L~kT' B=k"l - B=n? y]cr*a}l a¨ β wherever [I] walked with all the Israelites h&d*b*r D]B~rT' a#t-a~j~d v]bf? y]cr*a}l b¨ ¿a word [I] said to one of the rulers of Israel?

α

In order to avoid confusion with hypotactic elaborating clauses, it should be considered whether a&v#r has a connective role or not, tieing the dependent clause to a specific constituent from the previous clause (subject or object), which it describes. If that is not the case, but the clause rather expands a Location: place Circumstantial, we have a genuine hypotactic dependent clause, enhanced spatially, such as the one in #302.1 A paratactic clause enhanced as manner connects to its primary clause by the traditional connectives, with the distinction that it is interpreted as manner. Of the three types of manner enhancement – means, quality, comparison – probably the most frequent in SH is comparison. Joüon (1991, 643-4) treats it as a distinct clause type altogether (#303).2 (302) Gen 21.17 K'-v*m~u a$Oh'm a#l-qol h~N~u~r a¨ α heard God the voice of the youngster ×

B~a&v#r hWa - v*m b¨ where he [is] there

β

(303) Exod 26.4 (cf. 36.11) w=u*c't* l%la)t T=k}l#t u~l c=p~t h~yr'u> h*a#j*t m]Q*x> B~j)b*r#t a¨ 1 make loops of blue at the edge of the first curtain from the end to the joint ×

2

w=k}n

T~u&c#h B]cp~t h~yr'u> h~Q'xon> B~M~jB#r#t h~V}n't b¨

and likewise make

at the edge of the last curtain

at the second joint.

Means can be implicitly present in a simple coordination of clauses: (304) Deut 4.16 1 P#n - T~vj]tWn lest [you] act corruptly

a¨

Cf. also Lev 21.11; Judg 5.27. Paratactic enhancement by comparison is more frequent in HPy, where it does not require the use of prepositions due to brevity of expression (e.g. Prov 17.3 cf. Joüon 1991, 644). 1 2

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW ×

2

259

w~u&c't#m l*k#m P#s#l T=mWn~t K"l-s*m#l z*k*r ao n+q}b> b¨ by making for yourselves idols in the likeness of all images, male or female

At times, the secondary clause precedes the primary clause: (305) 1 Sam 20.33 1 w~Y*f#l v*aWl a#t-h~j&n't u*l*yw l=h*KK)to a¨ cast Saul the spear towards him to smite him ×

2

w*Y}d~u y=hon*t*n ... b¨ [whereby] learned Jonathan

Hypotactic clauses of the manner enhanced type are realized in SH by the presence of such particles as rv,a]K;, K., B., provided that the dependent clause is an extension of a Manner Circumstantial. The dependent clause of the following clause nexus is a variety of manner enhancement, namely quality, a clause that replaces ‘precisely’ its equivalent Circumstance. (306) Gen 7.16 w=h~B*a'm z*k*r Wn=q}b> m]K"l - B*c*r B*aW a¨ α those entering, male and female, from all creatures entered ×

β

K~a&v#r x]ww> a)to a$Oh'm b¨ as commanded him God

Exactly the same prepositional group may suggest comparison, whenever an element of the dominant clause is in the situation described by the dependent clause. (307) 2 Sam 16.23 w~u&x~t a&j't)p#l Æa&v#r y*u~x B~Y*m'm h*h}m b Ø a¨ α the counsel of Ahitophel Æwhich he gave in those daysØ ×

β

K~a&v#r y]va~l a'v B]db~r h*a$Oh'm c¨ as if inquire one for the word of God

We conclude this section on the expansion of the nexus clause through enhancement with a discussion on the cause-condition type. As expected, paratactic enhancement makes use of simple coordination, the relation between clauses being derived from the relation between verbs (cf. Joüon 1991, 627-9, 634, 636).

260

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

[cause ˆ effect] (308) Num 32.15 1 w=y*s~p uod l=h~N'jo B+M]dB*r a¨ [he] will continue yet to let you in the wilderness ×

2

w=v]j~TT#m l=k"l-h*u*m h~ZZ#h b¨ and so [you] will destroy all this people

[effect ˆ cause] (309) Num 36.7 1 w=Oa-t]S)l n~j&l> l]bn? y]cr*a}l m]M~F#h a#l-m~F#h a¨ not [you] remove the inheritance of the Israelites from tribe to tribe ×

2

K' a'v B=n~h*l~t m~F}h a&B)t*yw y]dB=qW B=n? y]cr*a}l b¨ for each in the inheritance of his father’s tribe shall cleave the Israelites

[condition: positive] (310) Deut 20.13 1 Wn=t*n*h yhwh a$Oh#yk* B=y*d#k* a¨ will give the Lord your God in your hands ×

2

w=h]K't* a#t-K"l-z=kWr*h l=p'-j*r#b b¨ and then smite all males by the sword

Concession can be rendered also by juxtaposition (cf. Gen 18.27; 48.14, etc. – Joüon 1991, 641). (311) Gen 18.27 1 h]N}h-n*a hoa~lT' l=d~B}r a#l-a&d{n*y a¨ look [I] was willing to speak to the Lord ×

2

w=a*n{k' u*p*r w*a}p#r b¨ [though] I [am] dust and ashes

As for hypotactic cause-condition enhancement in SH, more variations are available. Reason, purpose or result are signalled by particles such as yk, rb[b, rva, ![ml, ytlbl, bq[, or they can be realized by non-finite clauses with infinitives (Joüon 1991, 634-5, 636-40; Niccacci 1990, 132-3). In traditional grammars, these particles produce various clause types: conditional (IBHS, 636-8), final and result (IBHS, 638-40), or causal (IBHS, 640-41).

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

261

[reason] (312) Gen 29.34 u~TT> h~P~u~m y]L*w#h a'v' a}l~y a¨ α now this time will be joined my husband unto me ×

K' - y*l~d=TT' lo v=l)v> b*n'm b¨ because [I] begot him three sons

β

[result] (313) Josh 8.22 w~Y~KKW aot*m a¨ α [they] smote them ×

u~d-B]lT' h]va'r-lo c*r'd Wp*l'f b¨ until none was left neither survivor nor fugitive

β

[purpose] (314) Gen 27.10 1 w=h}b}at* l=a*b'k* a¨ bring it to your father ×

2

α ×

β

w=a*k*l b¨ [so that he] will eat B~u&b%r a&v#r y=b*r#k=k* l]pn? moto c¨ so that [he] will bless you before his death

Whereas concession owes much to conjunctions yKi, ~ai, l[; (mostly in (Joüon 1991, 641), condition is preferably accomplished by means of ~ai, ~ai yKi, or yleWl. POETRY)

(315) Exod 13.17 w=Oa-n*j*m a$Oh'm D#r#k a#r#x P=l]vT'm a¨ α not led God the way of the land of Philistines ×

K' q*rob hWa b¨ although near [was] he

β

Of course, the most common conditional clause consists of a protasis and an apodosis.1 The former represents the dependent clause and is typically introduced by ~ai/aOl ~ai, or yKi; the latter is also used by the Though, occasionally, the apodosis can be wanting altogether (cf. Gen 38.17b; Num 5.20; Exod 32.32 – Joüon 1991, 632). 1

262

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

dominant clause and is occasionally introduced by za'. As for the verb forms used in the protasis-apodosis relation, it seems that the preferred formula is y]qf)l/w=q*f~l (IBHS, 526-9), although, more frequently than might be expected from this verb form, q*f~l can also take over the function of the protasis verb (Joüon 1991, 630). [condition: positive] (316) Gen 13.16 a]m-yWk~l a'v l]mnot a#t-u&p~r h*a*r#x a¨ α if could someone to count the dust of the earth β

G~m - z~ru&k* y]M*n#h b¨ then also your offspring [he] counts.

[condition: negative] (317) Lev 22.6 w=Oa y)ak~l m]n-h~Q(d*v'm a¨ α not eat from the holy things ×

β

K' a]m-r*j~x B=c*ro B~M*y]m b¨ unless [he] washed his flesh in water

5.4.2.4. Embedded expanded clauses SH identifying embedded clauses can interfere with the Process of pragmatic mapping and seems to be of great importance. SFG understanding of embedding does not accommodate C.L. Miller’s understanding of a Process with a similar name, which is better described by multiple level projection (see infra 5.4.3). Typically, relative clauses introduced by rv,a] or hz< (mostly in HPy, cf. IBHS, 336-7) can be considered definite relative clauses provided that they relate directly to the antecedent clause, elaborating upon one of its nominal or adverbial constituents. Note how, in the following example, each of the two clauses of the clause nexus ends up with an embedded clause, here marked with Æ...Ø, introduced by rv,a] and having a finite verb. The two main clauses are in a paratactic extending relationship, though.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

263

(318) Gen 2.2 1 w~y+k~l a$Oh'm B~Yom h~V=b'u' m=l~akTo Æa&v#r u*c> bØ a¨ finished God in the seventh day the work Æ= which he did Ø +2

w~Y]vB)t B~Yom h~V=b'u' m]KK"l-m=l~akTo Æa&v#r u*c> dØ c¨| [he] rested in the seventh day from all the work Æ= which he didØ

Sometimes the embedded clause can have a non-finite verb or no verb at all, as happens in some existential clauses. (319) Gen 1.7 1 w~Y~u~c a$Oh'm a#t-h*r*q'~u a¨ made God the firmament +2

w~Y~bD}l B?n h~M~y]mÆa&v#r m]T~j~t l*r*q'~u cØ b¨ [he] divided between the water Æ= which [is] under the firmamentØ WB?n h~M~y]m Æa&v#r m}u~l l*r*q'~u eØ and the water Æ= which [is] above the firmamentØ

d¨|

Example (320) displays embedded expansion by elaboration, using hz y#a$h*b~n' a'v' that now will love me my husband

Other contexts may seem closer to the interpretation of K' recitativum (cf. Esh 1958). (336) Gen 20.11 w~Y)am#r a~br*h*m a¨ α said Abraham b¨ “β K' a*m~rT' α I said [to myself?] “β

r~q a?n - y]ra~t a$Oh'm B~M*qom h~ZZ#h c¨ surely there is no fear of God in this place

Sometimes the indirect speech is more evident: (337) Gen 29.12 w~Y~G}d y~u&q)b l=r*j}l a¨ α told Jacob to Rachel “β

K' a&j' a*b'h* hWa b¨ that the relative of his father [is] he

“γ

w=K' b#n r]bq> hWa c¨ and that the son of Rebekah [is] he

270

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Instead of yKi, one also finds the relative pronoun rv,a] (cf. Joüon 1991, 590). (338) 1 Kgs 22.16 u~d-K~M#h p=u*m'm a&n' m~vB]u#k* a¨ α how many times I adjure you “β

a&v#r Oa-t=d~B}r a}l~y r~q-a$m#t B=v}m yhwh b¨ that not [you] speak to me but the truth on the name of the Lord

Complementizers frequently co-occur with indirect interrogative clauses (Miller 1996, 116-9). Among the examples there is much irrealia material related to possible future scenarios, wherein the irrealis is quoted vividly as if it takes place in objective reality (cf. Lev 25.20; Josh 4.6, 21; 22.24; 1 Kgs 14.3). It should be noted that the verb form used as Process in the projecting clause is y]qf)l. (339) Gen 29.15 m*j*r y)am=rW b=n?k#m l=b*n?nW l}am)r a¨ α in time to come will say your descendents to our descendents, saying: “β

m~h-L*k#m w=l~yhwh a$Oj? y]cr*a}l b¨ what [have] you [to do with] the Lord, God of Israel

Others, though, are more integrated into the narrative they are part of, mainly as projected proposals, that is, the verb of the projecting clause is a volitive (Num 15.34; Josh 7.19; 1 Sam 10.15; 14.43; 2 Kgs 2.9).1 (340) Gen 29.15 h~GG'd> L' a¨ α tell me “β

m~h-M~cK%rT#k* b¨ what [is] your wage?

Indirect speech can also be realized by means of infinitival complements, where l. serves as a complementizer, with the possibility that the subject of the infinitival complement may be co-referential with either the speaker or addressee of the projecting clause (Miller 1996, 123-9).

1 There are exceptions to the previous rule, by which a q*f~l takes the place of the projecting clause verb, and a complementizer introduces a declarative projected clause instead (e.g. Neh 2.12).

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

271

(341) Num 9.4 w~y=d~BB}r m)v#h a#l-B=n? y]cr*a}l a¨ α told Moses to the Israelites: “β

l~u&c)t “Prepare

h~PP*s~j b¨ the Passover”

Although, in many modern languages, the main difference between direct speech and indirect speech lies in the use of the conjunction ‘that’ to introduce content-clauses of the latter type, in SH, this differentiation lies in the verbum dicendi selected (Goldenberg 1991, 85). Thus, dyGIhi always presents indirect speech, whereas rm;a' always presents direct speech. When rm;a ' appears to introduce indirect speech, it corresponds to the meaning ‘command’ or ‘think’. Goldenberg (1991, 88) agrees with others that the use of deictic markers (mainly pronouns) represents the most consistent marker of indirect speech, but in SH, ‘the combination of persons involved would not enable us to distinguish direct and indirect personal markers’ (Goldenberg 1991, 92). One of the most frequent formulae of reported thought in SH conveys the apologetic recognition of God’s pantocratic authority as ‘to recognize that I am the Lord’. Here the lexical verb √[dy realizes the Process of the projecting clause.1 (342) Joel 4.17 a¨ l=m~u~n T}d=uW α so that [you] will acknowledge ‘β

K' a&n' yhwh a$Oh?k#m b¨ that I [am] the Lord your god

Other verbs may also appear as Processes of the projecting clause, such as √!ma. (343) Exod 4.5 l=m~u~n y~a&m'nW α so that [they] will believe

a¨

1 Out of its 86 occurrences in HB, over 75% appear in Ezekiel, over 10% in Exodus, with the rest distributed across other books (Deut 29.5; 1 Kgs 20.13, 28; Is 45.3; 49.23; 60.16; Jer 9.23; 24.7; Joel 4.17).

272

‘β

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

K' - n]ra> a}l#yk* yhwh a$Oh? a&b)t*m b¨ that revealed to you the Lord God of their ancestors

Occasionally yKi may be replaced by rv,a] at the head of the projected clause: (344) Exod 11.7 l=m~u~n T}d=uWn a¨ α so that [you] will acknowledge ‘β

a&v#r y~pl#h yhwh B?n m]xr~y]m Wb?n y]vr*a}l b¨ that discriminates the Lord between Egypt and Israel

A rather rare situation appears in #345, where the dependent projected clause precedes the dominant one, and the common lexical verb √rma has the meaning ‘think’ or ‘mean’. (345) Exod 2.14 ‘β h~l=h*r=g}n' a¨ ¿killing me? α

a~TT> a)m}r b¨ you think

For Sternberg, free indirect speech is admittedly a typical literary device of HB that allows the author to mould his example as he pleases, serving his purpose of textual ambiguity (Sternberg 1985, 52-3). For him, the marker of free indirect speeh is hNEhiw>, operating as a clue to subjective opinions (p. 243), or to insights into one’s mind (p. 398). Obviously, hNEhiw> does not always function as a marker of free indirect speech (cf. IBHS, 6748). At a strict syntactic level, Kogut (1986) revives the idea that hNEhiw> introduces a content clause or a content object, without any evidence of being aware of Sternberg’s theory. Linguistically speaking, it is rather a deictic particle, indexing focus and viewpoint (Alonso-Schökel 1956). After evaluating the examples offered by Sternberg, Miller concludes that, although similar to modern western literature, SH free indirect speech displays several peculiarities.1 Even though hNEhiw> can be used to render a personal viewpoint, this has to be a visual perception, in which case the projecting clause refers to the narrator’s perspective, but the projected 1 Goldenberg (1991, 82-3) proposes more examples of free indirect speech: Gen 12.13; 41.15; Neh 6.6; 2 Ch 15.19; Hos 7.2; Ez 33.13-14; Job 35.14; Ps 9.21; 10.11, 13.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

273

clause with hNEhiw> refers to a character perspective. Besides, clauses with hNEhiw> are never projected quotes. Provided that their deictic centre coincides with the deictic centre of the projected clause, hNEhiw> can introduce direct quotations (Miller 1996, 85-9). (346) 1 Kgs 19.6 ‘β w~Y~BB}f a¨ [he] looked α

w=h]N}h m=r~a*v)t*yw u%G~t r=x*p'm w=x~PP~j~t m*y]m b¨ and look at his head a cake baked on stone and a jug of water

A large number of similar examples seem to suggest that such projected clauses are of the exclamative existential type following a behavioural clause. To summarise, SH makes use of several connectives such as yKi, rva],] and various WH- particles. Free indirect speech is a hybrid, reporting information on a paratactic structure. Frequently, hNEhiw> accompanies the projected clause. 5.4.3.3. Embedded projected clauses Even though it is not as frequent as in some modern languages, embedded projection appears in SH too. Some of the nouns that project embedded clauses are h['muv. ‘report’, rs'Wm ‘teaching’, lqo ‘report, utterance’ and rb'D' ‘report, claim, promise’.1 (347) 1 Sam 4.19 w~TT]vm~u a#t-h~V=m%~u> Æa#l - h]L*q~j a&ron h*a$Oh'm bØ [she] heard the report Æthat was taken the ark of God Ø ÆWm}t j*m'h* w=a'v*h cØ Æand that died her father-in-law and her husbandØ

a¨

Of all these projecting nouns, the most frequent one is obviously rb'D', which is usually accompanied by the complementizer rmll (cf. Num 36.6; Josh 1.13; 2 Kgs 19.9/Is 37.9; Neh 1.8-9).

1 In HPy, the embedded clause may stay as the secondary line of a couplet whose first line ends with the projecting noun. Even-Shoshan identifies no fewer than fifty occurrences of rs'Wm (e.g. Prov 19.27AB), but only two projected embedded clauses (cf. also Jer 35.13).

274

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(348) 2 Sam 13.33 w=u~TT> a~l-y*c}m a&d)n' h~M#l#k a#l-l]Bbo D*b*r l}am)r now should not put my lord the king to heart the report saying ÆK"l-B=n? h~M#l#k m}tWب Æall the sons of the king diedØ

The presence of the complementizer allows various constituents of the main clause to interpose between the projecting noun and its embedded clause. (349) Gen 45.16 w=h~Qol n]vm~u B?t P~ru)h l}am)r ÆB*aW a&j? yos}pØ ¨ the report was heard in Pharaoh’s palace, saying Æhave come the brothers of JosephØ

These occurrences are different from those presented with the introductory oracular formula, which make use of embedded clauses expanded through enhancement (clause (b) in #350). (350) Num 22.20 w=a~k a#t-h~DD*b*r Æa&v#r a&d~B}r a}l#yk* bØ a¨ a{to t~u&c#h c¨ surely the word Æ× which [I] said to youØ that [you]should do

Projection by fact nouns are rarer in SH than in English. One may include here the embedded clause projected by the noun tAa (‘sign’) and some non-finite clauses with infinitive construct. (351) Judg 6.17 w=u*c't* L' aot Æv*a~TT> m=d~BB}r u]M'Ø ¨ give me a sign Æthat you are speaking with meØ (352) 1 Sam 24.12 ÆK' B=k"rt' a#t-K=n~p m=u'l=k* aØ Æw=Oa h&r~gT'k* bØ ¨ [the fact that] ÆI cut off the skirt of your robeØ Æand not [I] killed you Ø c¨ Wr=a}h d¨ ÆK' a?n D~u B=y*d' r*u> w*p#v~u eØ ¨ [you] know and [you] perceiveÆthat there isn’t on my hand evil and transgressionØ

Example (352) displays a double fact, realized by two embedded clauses as nominalizations (clauses (a) and (b)), functioning as Identifiers in an identifying clause, whose Identified element is realized by another embedded clause standing as nominalization (clause (e)), introduced by yKi also. Two imperative verbs, iterated as oath formula, function as the identifying Process (clause (c) and (d)). Unfortunately, similar examples are rather difficult to find.

A S YSTEMIC FUNCTIONAL GRAMMAR OF STANDARD HEBREW

275

To summarise, the two systems of networks that are complexing clauses in English – taxis and logico-semantic type – can be found in SH too. Parataxis tends to be construed by same verbal forms linked in chains: w~Y]qf{l’s in narrative texts, w=q*f~l’s in prescriptive texts, volitives in hortatory texts and nominal predicators in descriptive clauses. Hypotaxis makes more use of conjunctions, w=q*f~l’s, w~y=h'’s, and w~w-X-q*f~l verbal forms, as well as w=u~TT> and w=h]N#h particles. Relationships between adjacent clauses are further complicated by the existence of a second system of networks, that of the logico-semantic type. With its two sub-types, expansion and projection, and the possibility of embedding, the logico-semantic type allows for a finer complexing of clauses. Paratactic expansion, whether through elaboration, extending or enhancing, prefers to put together clauses that have an identical or equivalent verbal realization, and employs with some success the w=q*f~l’s, w~y=h'’s, and w~w-X-q*f~l constructions. Hypotactic expansion makes particular use of conjunctions and particles, such as b, k, rvak, rvab, ~a yk, ylwl and yk spa. Paratactic projection employs the use of verbal Processes for projected quotation and mental Processes for mental projection. Hypotactic projection gives preference to verbal Processes realized by lexical verbs such as [dy, dygh, !ma and even rma ‘to think’, but conjunctions such as rva, yk appear as well. Projected embedded clauses are prompted mainly by nouns such as h[mv, rfwm, lq, twa but mainly rbd. The following chapter will focus on the application of these SFG principles to HPy. Several Psalms displaying the QYYQ verbal sequence have been selected for SFG analysis, only to allow minimum observations of the phenomenon in the context of longer texts. This leads to a careful consideration of all Psalm couplets in our database and to an interpretation of the results.

6 PSALM COUPLETS WITH QYYQ VERBAL SEQUENCE: A SFG INVESTIGATION Mukařosky identified POETRY as the most defamiliarised form of language (cf. Rosenbaum 2000, 150). Among the familiar patterns to defamiliarise (foreground) language in HPy, Rosenbaum included: inversion, insertion, parallelism, gapping, repetition, word-pairs, chiasmus, swapping, pivot, terrace and staircase. As may be inferred, these are structural deviations from the regular a-b-c structure of the basic pattern. After establishing the SFG principles on SH prose texts (Chapter 5), we will now investigate their applicability to HPy. We will, therefore, analyse a different Psalm at each of the three levels of meaning. The results are then interpreted so as to open the way for the same type of analysis to be applied to Psalm couplets with QYYQ verbal sequence. Each poetic verse (be it a couplet or a tercet), is displayed on verselines with their translation below and the SFG analysis underneath. Versets are marked with exponential Arabic numerals (1, 2, 3, etc), verse-lines are marked with capitals (A, B, C, etc.) and clauses are marked with exponential minuscules (a, b, c, etc). Verbal forms are not translated in detail, but their root meaning (√) is given and their verbal form is specified in the margin.

6.1. THEMATIC S TRUCTURE OF PSALM COUPLETS WITH A QYYQ VERBAL SE QUENCE 6.1.1. Psalm 38 Since it contains the superscript, which is irrelevant for our discussion of the verbal sequence in HPy, the first verse of this Psalm is ignored. From an information structure viewpoint, line B usually builds on the meaning set by line A. Therefore, there is a greater chance to find a New element functioning as focus in the primary verse-line than in the secondary one. In this regard, Lunn’s thesis on the primacy of line A over line B is correct. 277

278

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

References to the information unit are made only when it does not follow the expected information structure. Psalm 38 2A yhwh a~l-B+q#xP=k* Lord, not in your wrath Interp Topical Marked Theme Given New

tok'j}n' a¨ √rebuke me

vol // vol

Rheme

B Wb~j&m"tk* t=y~S=r}n' b¨| or in your rage √discipline me Text Top MTheme Rheme Given New

The Psalm starts with a couplet that has both its volitive clauses displaying a marked Theme. The Circumstantial occupies the Predicator slot. Parallelismus membrorum may explain the necessity for a Marked Theme in line B/clause (b) following the pattern of the previous one. The information unit in line A can be considered New in its entirety with the exception of the Vocative that refers to an entity easy to reconstruct. ‘Or in your rage’ in line B is not new any more, after being already mentioned in line A. A K' - j]X#yk* n]j&tW b' a¨ because your arrows √penetrate me Text Topical Rheme Marked Theme New (focus) Given 3

qtl // wayqtl

B w~T]nj~t u*l~y y*d#k* b¨| √descend upon me your hand Theme Rheme Given New

Given the reason for the previous poetic unit, the couplet in verset 3 consists of two declarative clauses whose Predicator is realized by two verbal forms from the same root, a q*f~l and a w~Y]qf)l, respectively. Whereas line B, coextensive with clause (b), is regular in terms of thematic structure, line A/clause (a) has the Subject ‘your arrows’ in the most salient

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

279

slot. The appearance of ‘your arrows’ in this couplet is unexpected both thematically and informationally. Here the New element functions as focus. A a?n - m=t)m B]bc~r' m]PP=n? z~um#k* a¨ there isn’t soundness in my flesh because your indignation Theme Rheme New (focus) Given 4

exist // exist

B a?n - v*lom B~u&x*m~y m]PP=n? j~F*at' b¨| there isn’t peace in my bones because my sin Theme Rheme Given New

Two declarative clauses compose the verse-lines of the couplet in verset 4 and, in both cases, the Predicator is realized by existence particles fronting the respective clauses. Lines A and B are characterized by close grammatical parallelism. Since ‘your indignation’ in line A is known to the reader from the very beginning of the poem, the New element here is ‘there is no soundness in my flesh’. In line B, the New element stays at the end of the clause where it is expected. A K' u&w)n)t~y for my guilt Text Topical Marked Theme Given 5

u*b=rW r)av' a¨ √pass over my head

qtl // yqtl

Rheme New

B K=m~C*a k*b}d y]kB=dW m]M#N' b¨| like a heavy burden √be heavy to me Text Topical Rheme Marked Theme Given New

Here we have the first of the three QYYQ verbal sequence examples in this Psalm. Everything but the grammatical parallelism of the previous couplet can be found here. Both verse-lines/clauses have a Marked Theme, but the Circumstantial ‘like a heavy burden’ replaces the Subject ‘my guilt’, which is elided in line B. The two clauses are complementary to each other. ‘My guilt’ echoes ‘my sin’ from the previous couplet. Varying the verbal form is complementary to the double Marked Theme.

280

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

A

h]ba'vW a¨ n*m~QW j~BBWr)t*y [they] √stink √fester my wounds Theme Rheme Theme Rheme Given New 6

qtl / qtl // ø

B m]PP=n? a]ww~lT' b¨| because my folly Rheme Given New

Verset 6 consists of a couplet with a complicated syntax but regular thematic structure. The only two verbs present, both q*f~l, stand together in line A. Clause (b) stretches over two verse-lines. The Subject ‘my wounds’ explicit in clause (b) is only implicitly present in clause (a). The information unit of line A includes more than one clause, but, in line B, the clause is elliptical, without altering the expected information structure. This situation is repeated in verse 7. A n~u&w?t' a¨ v~j)t' u~d-m=a)d b¨ [I] √bend [I]√bow down very much Theme Rheme Theme Rheme Given New 7

qtl / qtl // qtl

B K"l-h~Yom q)d}r h]L*k=T' c¨| all day long mourning [I] √walk Topical MTheme Rheme Given New

Verset 7 consists of a couplet with three clauses again, this time all declarative with an explicit finite Predicator (q*f~l). The only exceptional situation is evidenced in the last clause, whose topical Theme is realized by the Circumstantial ‘all day long’. If the sixth verse of this poem is a variant of the seventh verse, then the Marked Theme seems to be triggered by repetition of similar verbal forms.

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

A K'-k=s*l~y for my loins Text Top MTheme Given 8

m*laW n]ql#h a¨ √be filled [with] burning

281

qtl // exist

Rheme New

B w=a?n m=t)m B]bc*r' b¨| and there isn’t soundness in my flesh Theme Rheme New (focus) Given

Two declarative clauses, of which the first has a Predicator realized by a finite verb (q*f~l) and the second by an existence particle. Whereas the second clause has a regular thematic structure, clause (a) has the Subject ‘my loins’ as topical Theme. The information structure takes a twist in line B, where the known reference ‘my flesh’ is placed at the end of the clause. a¨ w=n]dK?t' A n=pWgot' u~d-m=a)d b¨ [I] √grow dumb [I] √be crushed very much Theme Rheme Theme Rheme Given New 9

qtl / wqtl // qtl

B

v*a~gT' m]N~h&m~t l]BB' c¨| [I] √groan on account of the anguish of my heart Theme Rheme Given New

Regular thematic structure is noted in the case of the three clauses (q*f~l/w=q*f~l//q*f~l) that compose the ninth verse, each being more complex syntactically than the previous one. Following up on the type identified in verse 6 and 7, verse 9 alters the second of the verbal forms and, as a result, line B does not have a Marked Theme.

282

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

A a&d)n*y n#gDD+k* k"l-T~a&w*t' a¨ Lord, in front of you [are] all my desires Interpersonal Topical Rheme Theme Given New 10

nom // qtl

B w=a~nj*t' m]M=k* Oa-n]sT*r> b¨| and my sighing from you not √be hidden Text Topical Rheme Marked Theme New (focus) Given

Unlike clause (a), clause (b) exhibits a Marked Theme realized by the Subject ‘my sighing’ in a declarative with a finite verb (q*f~l). ‘My sighing’ stands as the New element, functioning as focus and Marked Theme. Another example of couplets with three q*f~l verbal forms follows. A l]B' my heart Topical MTheme Given 11

s=j~rj~r a¨ √palpitate Rheme

u&z*b~n' k)j' b¨ √leave me my strength Theme New

qtl / qtl // qtl

Rheme

B w=aor - u?n~y G~m-h}m a?n a]T' c¨| and the light of my eyes, too they, isn’t with me Text Topical Rheme Marked Theme New (focus) Given

Unlike verses 6, 7 and 9, clauses (a) and (c) in verse 11 display Subject ‘my heart’ and ‘the light of my eyes’, respectively, as Marked Theme although their quality as declarative clauses is slightly different: the former has a finite verb (q*f~l) and the latter is an existence verbless clause. The New element swapped places with the Given in line B, allowing ‘the light of my eyes’ to be in focus.

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

A a)h&b~y w=r}u~y m]N#g#d n]gu' y~u&m)dW a¨ my friends and my companions from aloof my plague √stand Topical Rheme Marked Theme New (focus) Given 12

B Wq=rob~y and my neighbours Text Topical Marked Theme Given

283

yqtl // qtl

m}r*j)q u*m*dW b¨| at distance √stay Rheme New

Verset 12 consists of two declarative clauses with finite clauses, and a QYYQ verbal sequence examples. The two clauses develop themselves according to a similar paradigm, with the finite verb from the same root pushed backwards by its Subject (‘my friends and my companions’ and ‘my neighbours’, respectively). Grammatical parallelism, intersected with Subject markedness, produces a symmetrical structure and an unfamiliar one too. New information (‘my friends and my companions’) is introduced in line A of this verse as focus. A w~yn~qvW m=b~qv? n~pv' a¨ √lay snares the seekers of my life Theme Rheme Given New 13

wyqtl // qtl // yqtl

B w=d)r=v? r*u*t' D]BB=rW h~wwot b¨ and the seekers of my ruin √speak of chasms Text Topical Rheme Marked Theme Given New C Wm]rmot K"l-h~Yom y#hGW c¨| and treacheries all day long [they]√plot Text Topical Rheme Marked Theme New (focus) Given

284

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

The only tercet of this Psalm has three declarative clauses with the Predicator realized by a finite verb, one clause per verse-line. In the last two clauses one of the less expected constituents occupies the most salient slot, the Subject ‘the seekers of my ruin’ and the Complement ‘treacheries’, respectively. The latter functions also as focus at the information structure level. This tercet may well be a variant of a couplet with QYYQ verbal sequence. A w~a&n' but I Text Top MTheme Given 14

k=j}r}v Oa a#vm*u a¨ like a deaf not √hear

yqtl // yqtl

Rheme New

B Wk=a]L}m Oa y]pT~j - P'w b¨| and like a dumb not √open my mouth Text Top MTheme Rheme Given New

Apart from the ellision of Subject in line B/clause (b), the two verselines/clauses are symmetrical. In both verse-lines, a marked Theme replaces the regular one, i.e. the finite verb (y]qf{l), the Subject ‘I’ in line A and the Circumstantial ‘like a dumb’ in line B. A w*a$h' K=a'v Æa&v#r Oa-v)m}~u b Ø a¨ [I] √become like a man Æwhich not √hearØ Theme Rheme Given New B w=a?n B=p'w Tok*jot c¨| and there isn’t in his mouth arguments Rheme Given New 15

wayqtl // exist // exist

This is probably the most unusual construction of this Psalm. The couplet has three clauses, but the last two qualify the subject of the first clause (embedded clauses). All three clauses display regular thematic structure, including clause (b), whose subject elided and which is an embedded clause.

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

A K' - l=k* because for you, Text Top Marked Theme Given 16

yhwh joj*l=T' a¨ Lord, [I] √await

285

qtl // yqtl

Rheme New

B a~TT> t~u&n#h a&d)n*y a$Oh*y b¨| you √answer, Lord my God Top MTheme Rheme Given New

This is the last of the three QYYQ verbal sequence examples in Psalm 38. God’s person appears in both verse-lines of this couplet, first simply as a reference in the Circumstantial ‘for you’, then as the Subject ‘you’. In both cases, this reference fronts the Predicator functioning as marked Theme. These verse-lines are overloaded with references to ‘the Lord’, one of the characters of this poem, thus part of the known information. A K' - a*m~rT' because [I] √say: Theme Rheme Given 17

a¨

P#n - y]cm=jW “only not [they] √rejoice Theme Rheme New

l' b¨ over me

qtl / yqtl // qtl

c¨| B B=mof r~gl' u*l~y h]gD'lW while slipping my foot against me [they] √exalt themselves” Topical MTheme Rheme Given New

Line A of this couplet has two clauses, line B being subordinate to the first. If the projecting clause is left aside as anacrusis, the remaining clauses may produce another QYYQ verbal sequence example. While clause (b), like its preceding clause, follows the pattern of a regular thematic structure, clause (c) has the Circumstantial ‘while slipping my foot’ as topical Theme instead.

286

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

A K' -a&n' l=x#l~u n*kon a¨ because I to fall am ready Theme Rheme Given New 18

B Wm~kaob' and my pain Text Top MTheme Given

ptcp // exist

n#gD' t*m'd b¨| [is] before me always Rheme New

Verset 18 consists of a couplet with two verse-lines coextensive with two clauses, neither of which has a finite verb. Since clause (a) is a declarative: non-finite: participial clause, its thematic structure is regular. Not so clause (b); as a classification verbless declarative clause, the Predicator would have been expected first, not the Subject ‘my pain’, which here functions as Marked Theme. A K'-u&w)n' a~GG'd a¨ for my guilt [I] √confess Text Top MTheme Rheme Given New 19

yqtl // yqtl

B a#da~g m}j~F*at' b¨ [I] √be troubled by my sin Theme Rheme Given New

Two y]qf{l verbal forms realize the Predicator of each of the two clauses that compose this couplet, one for each verse-line. The regular thematic structure of the second clause is not matched by the first, whose Complement ‘my guilt’ is Marked Theme. A similar structure appears in the next couplet, this time with q*f~l verbal forms. From their close placement, it would appear that q*f~l and y]qf{l verbal forms are interchangeable in HPy.

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

A w=a)y=b~y j~Y'm u*x}mW a¨ and my vigorous enemies √be numerous Text Topical Rheme Marked Theme Given New 20

287

qtl // qtl

B w=r~BW c)n=a~y v*q#r b¨| √be many the ones hating wrongly Theme Rheme Given New

Another innovative couplet appears in verset 21, where the Theme occupies line A of the couplet and the Rheme comes next in line B. 21

A Wm=v~L=m?

r*u> T~j~t fob>

ø // yqtl

a¨|

and the ones rewarding evil Text Topical Marked Theme Given

for

good

B y]cf=nWn' T~j~t r*d=p' - fob b¨| √slander me for pursuing good Rheme New

This couplet has only one verb, a y]qf{l, and that appears at the head of line B. All authorities consulted, both ancient and modern, agree on this division of the couplet into these two verse-lines. Here the information unit covers the two verse-lines of the verse and the New element comes at the end of the verse. The non-recoverable information in this verse refers to that quality of the poet of ‘pursuing good’. A a~l - T~u~zb}n' yhwh a¨| not √forsake me, Lord Theme Rheme New (focus) Given 22

vol // vol

288

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

B a$Oh~y a~l God, not Interp Text Theme Given

T]rj~q m]M#N' b¨| √be far from me Top Rheme New

This couplet has two verbs, one on each verse-line, both negative volitives. A jWv> l=u#zr*t' a¨ √hurry to my help Theme Rheme Given New 23

vol // ident

B a&d)n*y T=vWu*t' b¨| the Lord [is] my salvation Theme Rheme Given New The last couplet of the Psalm displays regular thematic structure given the fact that the first clause is volitive and the second one is a declarative:verbless:identification. Surveying the Psalm as a whole, there are twenty-two verses, of which only one (v. 13) is a tercet, the rest being couplets. Syntactically, fourteen of the twenty-one couplets can be considered regular having one clause per verse-line. Five couplets have three clauses (vv. 7, 9, 11, 15, 17), one couplet continues one of the two clauses in the primary verse-line on to the secondary one (v. 6), and another couplet has only one clause extended over both its verse-lines (v. 21). The analyses of the information structure of the information units this poem consists of reveals that in most cases the Given + New order of elements is followed (14/22 ~ 64 per cent).1 When the regular information structure is altered, a New element is given the focus function only in one of the verse-lines of the verse. Therefore, two patterns of tonic contour emerge: a descending-ascending trajectory (↓↑, cf. vv. 3, 4, 12, 22) and an ascending-descending trajectory (↑↓, cf. 8, 10, 11). The only tercet in this Here are included the information units of the following verses: 2, 5, 6, 7, 9, 14-20, 21 and 23. 1

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

289

poem has the latter pattern in its last verse-lines. It appears that the information structure of a given poetic unit is employed as a cohesive factor. The Psalm displays forty-nine clauses, extended over forty-four verselines, of which twenty-three front the constituent that regularly comes first with other material functioning as Marked Theme. There are no interrogative clauses in this Psalm nor any exclamative clauses either. The remaining mood types are distributed as follows: declarative:finite clauses (34), declarative:non-finite:participle clauses (2), declarative: nonfinite:identification clause (1), declarative:non-finite:classification clauses (2), declarative:non-finite:existence clauses (3), and volitive clauses (5). It is very important to note that, of the ten occurrences of y]qf{l, eight are in the context of Marked Theme (80 per cent). Nineteen q*f~l verbal forms occur of which 11 are in the context of Marked Theme (57 per cent). None of the w=q*f~l (9A, 20B) and w~Y]qf)l (3B, 13A, 15A) verbal forms appear with Marked Theme. All three instances where the QYYQ verbal sequence appears show Marked Theme in both verse-lines of the couplet. It appears that y]qf{l carries the idea of Markedness with it in the clause where it appears. Also relevant to this conclusion is verse 21, with only one clause, the verb being fronted by a topical Theme as long as a verse-line. 6.1.2. Evaluation of the database Upon surveying our database of Psalms after verse-line delimitation was applied to it, one hundred and two couplets were included in our list of examples that display a QYYQ verbal sequence (cf. Appendices 1 and 2). After surveying the couplets in our data, the following results emerge. MTheme//Theme

Theme//MTheme MTheme// MTheme Theme//Theme Total

q*f~l//y]qf{l 15 18 19 8 60

y]qf{l//q*f~l 11 4 14 13 42

Total 26 22 33 21 102

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

290

In more detail the above statistics look as follows: MTheme//

Theme

Theme//MTheme

MTheme// MTheme

Theme//Theme

q*f~l//y]qf{l 13.6a; 30.7; 60.12; 62.5a; 66.18; 67.7; 68.11; 74.14; 78.20b; 101.3b; 111.5; 118.10; 119.11; 119.101; 139.13 2.1; 6.10; 18.5; 26.4; 26.5; 63.7; 63.8; 65.12; 71.17; 73.3; 73.9; 81.7; 83.6; 92.5; 102.15; 116.2; 119.69; 140.3 7.14; 17.11; 18.41; 26.12; 27.4a; 27.10; 37.23; 38.5; 38.16; 49.15a; 50.19; 51.8; 65.4; 78.64; 88.14; 89.13; 89.36; 104.6; 119.95 73.6; 74.1b; 95.11; 104.5; 108.12; 119.82; 119.121; 132.11

y]qf{l//q*f~l 9.8; 23.6; 31.6; 37.11; 73.18; 73.27; 119.104; 127.1b; 127.1c; 132.17; 142.4b 8.7; 44.11; 44.12; 47.5

10.14b; 11.3; 33.21; 38.12; 41.4; 50.3b; 55.5; 58.9; 83.3; 85.12; 87.6; 104.26; 119.93; 119.100 3.7; 5.6; 18.37; 28.1c; 44.13; 49.9; 60.11; 89.44; 104.16; 107.2; 108.11; 126.2b; 139.20

Table 6.1: Thematic f-structure of Psalm couplets with QYYQ verbal sequence

Several conclusions can be drawn with some degree of certainty from the above tables. (1) Examples with a q*f~l//y]qf{l verbal sequence are more frequent than those with a y]qf{l//q*f~l verbal sequence. The largest number of verses in the Psalter is produced by couplets with the same verbal form in each verse-line, most likely driven by grammatical parallelism. In this context, couplets with QYYQ verbal sequence stand as rather rare exceptions.1 Conserving the thematic structure in 21 examples could well indicate that couplets with QYYQ verbal sequence are exceptional cases of couplets with the same verbal forms. (2) There is a large majority of cases where the thematic f-structure is clearly altered in at least one of the two verse-lines of the couplet (81/102). Out of these cases of couplets with unfamiliar structure, the y]qf{l verselines have a Marked Theme in 29 of the cases, and the q*f~l verse-lines in 19 of the cases, the other 33 displaying a Marked Theme in both verse-lines. 1 To these, one could have added, without any hindrance to the paradigmatic couplet, other unorthodox examples, such as couplets with more than one Predicate on at least one of the verse-lines.

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

291

This may indicate that couplets with QYYQ verbal sequence are very unstable in terms of thematic structure. In such a couplet, at least one of the verbs has to be accompanied by a Marked Theme. (3) This data also means that a q*f~l verse-line with a regular Theme is more likely to trigger a Marked Theme in the parallel y]qf{l verse-line than vice versa. Therefore, one can safely say that there is a tendency in HPy verse to have a Marked Theme when a y]qf{l verb is used in one of the verse-lines of a couplet, with a q*f~l verb in the parallel verse-line. Thus, a q*f~l verse-line is more stable than a y]qf{l one. The QYYQ verbal sequence functions as a cohesion factor at the level of couplets that contain it.

6.2. MOOD STRUCTURE OF PSALM COUPLETS WITH QYYQ VERBAL SEQUENCE

6.2.1. Psalm 63 This Psalm is a relatively short one with most verse-lines built up of couplets, one of which various verbal sequences are met. Since it contains the superscript, irrelevant for our discussion of the verbal sequence in HPy, the first verse of this Psalm is ignored. Psalm 63 A a$Oh'm a}l' a~TT> a¨ a&v~j&r#K* b¨ O God, [are] my God you, [I] √seek for you Vocative Pred Subject Fin/ Pred Compl Residue Mood Mood Residue 2

B x*m=a> l=k* √be thirsty for you Fin/ Pred Compl MoResidue

n~pv' c¨ my soul Subject -od

C K*m~H l=k* √faint after you Fin/ Pred Compl MoResidue

b=c*r' d¨ my flesh Subject -od

nom / yqtl // qtl // qtl // ø

292

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

D B=a#r#x-x]Y> w=u*y}p B=l'-m*y]m e¨| in a dry and weary land without water Adjunct Residue

The unity of this verse (a quatrain) is supported by the strict parallelism of the middle verse-lines and the fluidity of the marginal verselines. Although elliptical, one could see clause (e) as the continuation of clause (b). Whenever an explicit Subject appears (clauses a, c, d), it is fronted by the Complement of the respective clause. All the clauses are declarative. A K}n B~Q)d#v j&z't'k* a¨ thus in the sanctuary [I] √see you Compl Adjunct Fin/ Pred ResMood -idue 3

qtl // ø

B l]raot u%ZZ=k* Wk=bod#k* b¨| to √behold your power and your glory Adjunct Complement Residue

Elliptical declarative clause (b) completes the idea of another declarative clause (a) by means of an infinitive construct and two complements. The Finite verb operator of the first clause is again a q*f~l fronted by a Marked Theme. A K' fob j~sD=k* because [is] better your kindness Adjunct Pred Subject ResMood 4

B c=p*t~y my lips Subject Mood

m}j~Y'm a¨ than life Adjunct -idue

nom // yqtl

y=v~B=jWn=k* b¨| √praise you Fin/ Pred Compl Residue

Both clauses are declarative and have explicit Subjects. Clause (a) is a classification clause and displays the expected regular order. Clause (b) is a declarative:finite clause with its Finite verbal operator – a y]qf{l – fronted by its Subject.

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE 5A

K}n So Adj Res-

a&b*r#kk* b=j~Y*y a¨ [I] √bless you in my life Fin/ Pred Compl Adjunct Mood -idue

293

yqtl // yqtl

B B=v]mk* a#C*a k~P*y b¨| in your name [I] √raise my palms Adjunct Fin/ Pred Compl ResMood -idue

This couplet has each of its verse-lines co-extensive with a declarative clause in which the Finite verbal operator is an y]qf{l. In clause (b), ‘in your name’ functions as Adjunct in the proposition and Marked Theme in the message most likely in order to realize a closer bond between this verse-line and the previous one, namely, a syntactical chiasmus. A K]mo j}l}b w*d}v}n T]cB~u n~pv' a¨ like marrow and fatness √be sated my soul Adjunct Fin/ Pred Subject ResMood -idue 6

B w=c]pt? r=n*not and [with] lips of joy Adjunct Res-

yqtl // yqtl

b¨| y=h~L#l - P' √praise my mouth Fin/ Pred Subject Mo-idue -od

Verse 6 displays two verse-lines co-extensive with two declarative clauses and both have their Finite verbal operator realized by y]qf{l verbal forms. In this case, the poet preferred to set the syntactic constituents in a more rigid parallelism net. This may be the reason for having an unfamiliar constituent at the head of both clauses functioning as Marked Theme at the textual level and as Adjuncts at the proposition level. We have assumed that the preposition is missing in the second clause, and have supplied one in order to give some sense to the clause. A a]m z=k~rT'k* u~l - y=xWu*y a¨ when [I] √remember you on my couch Adj Fin/ Pred Compl Adjunct ResMood -idue 7

qtl // yqtl

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

294

B B=a~vm%rot in the may watches Adjunct Res-

a#hG#h [I] √muse Fin/ Pred Mood -idue

B*k b ¨| at you Compl

Here we find the first of the two examples of QYYQ verbal sequences this Psalm offers. We note especially the fact that the verse-line whose Finite verbal operator is realized by a y]qf{l has a constituent functioning both as Adjunct and Marked Theme. A K' because Adjunct Res8

h*y't* u#zr*t> [you] √be a help Fin/ Pred Compl Mood -idue

L' a¨ to me Adj

qtl // yqtl

B Wb=x}l K=n*p#yk* a&r~N}n b¨| and in the shadow of your wings [I] √sing Adjunct Fin/ Pred ResMood -idue

Nothing changes in terms of clause mood, but the alternation of Finite verbal operator from q*f~l to y]qf{l is accompanied by a change in thematic structure. The second clause has the verb preceded by a constituent whose thematic function as Marked Theme is conflated with Mood function of Adjunct. A D*b=q> √cling Fin/ Pred MoRes9

B B' on me Adj Res-

n~pv' a~j&r#yk* a¨ my soul to you Subject Adjunct -od -idue

T*m=k> √support Fin/ Pred Mo-idue

qtl // qtl

y+m'n#k* b¨| your right hand Subject -od

Both clauses in verse 9 are declarative and have their Finite verbal operators realized by q*f~l verbal forms, but while the first clause follow a regular structure, the second one has the Adjunct functioning also as Marked theme at the textual level.

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

A w=h}M> l=voa> and they to destroy Subject Adjunct MoRes10

y+b~qvW √seek Fin/ Pred -od -idue

n~pv' a¨ my soul Compl

295

yqtl // yqtl

B y*b)aW B=t~jT]Yot h*a*r#x b¨| [they] √go down in the lowest places of the earth Fin/ Pred Adjunct Mood Residue

Again, both Finite verbal operators of the two clauses, which are coextensive with the two verse-lines of verset 10, are realized by y]qf{l verbal operators. It is only the first of the two that is fronted by an explicit Subject, functioning also as Marked theme at the textual level. In the second clause/verse-line the Subject is elided. A y~G'r%hW u~l-y+d?-j*r#b a¨ [they] √be deliver over to the sword Fin/ Predicator Adjunct Mood Residue 11

yqtl // yqtl

b¨| B m=n*t v%u*l'm y]hyW the portion of jackals [they] √be Complement Fin/ Pred ResMood -idue

The couplet in verset 11 displays similar characteristics to the one in the previous verset. This time there is no explicit Subject in either of the clauses. Clause (b), which is co-extensive with line B, has a Complement functioning as Marked Theme at the textual level. It is likely that structural considerations constrained the placement of y]qf{l verbal forms towards the extremities of the couplet. A w=h~M#l#k y]cm~j b}aOh'm a¨ but the king √rejoice in God Subject Fin/ Pred Adjunct Mood Residue 12

B y]th~L}l √praise Fin/ Pred MoRes

K"l-h~N]vB*u Bo b¨ all who swear by him Subject -od

yqtl // yqtl // yqtl

296

C K' because Adj Res

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

y]S*k}r P' dob=r? v*q#r c¨| √be shut up the mouth of those speaking falsehood Fin/ Pred Subject Mood

A tercet concludes the poem. All its three verse-lines have y]qf{l Finite verbal operators, but only the first one is preceded by an unexpected constituent. In conclusion, we can survey the Mood f-structure of this poem from several perspectives: Finite verbal operator, Subject, and relation to thematic f-structure. Of the 26 clauses this poem consists of, in one the verb is elided (2e), two have a non-finite verb form (3b and 4a), leaving the large majority to comply with the presence of finite verbal forms: sixteen y]qf{l and seven q*f~l. There is a (2:1) ratio in favour of the y]qf{l. With the exception of two clauses (2e, 3b) where the Subject is elided, all the other clauses display a Subject either explicit (14 occurrences) or implicit (10 occurrences). Occasionally, an explicit Subject can function also as Marked theme (4b, 10a, 12a). When Thematic and Mood clause structures are seen together, one observation seems inescapable: y]qf{l commonly functions as Finite verbal operator in clauses co-extensive with the second verse-line of a couplet, where another constituent (mostly Complement or Adjunct) functions also as Marked Theme. Of the 16 cases where a y]qf{l appears, there are nine such cases: 4b, 5b, 6a, 6b, 7b, 8b, 10a, 11b, 12a. Whenever present, except for 6a and 6b, the explicit Subject functions as Marked Theme. Only two q*f~l verbal forms appear in such cases (3a, 9b). If y]qf{l, as Finite verbal operator, triggers a Marked Theme, at least at the theoretical level for now, the six exceptions that fail to comply have to find another explanation. The examples in 2b, 12b and 12c seem to be justified by the fact that they are part of larger poetic units – a quatrain and a tercet. That leaves the examples in 5a, 10b, and 11a unexplained.

2a 2b 2c 2d 2e 3a 3b 4a 4b 5a 5b 6a 6b 7a 7b 8a 8b 9a 9b 10a 10b 11a 11b 12a 12b 12c

A A B C D A B A B A B A B A B A B A B A B A B A B C

              

          

   -

 -

       -

297

Vocative

MTheme/

Adjunct

MTheme/

Compl

MTheme/

Subject

MTheme/

Explicit Subject

YIQTOL

Poetic line

Reference

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

 -

Table 6.2: Distribution of Marked Theme in relation to Mood constituents in Ps 63

It seems that in 10b the explicit Subject/Marked Theme is elided, but in 5a and 11a structural reasons such as chiasmus may have been responsible for leaving the y]qf{l verbal form in a position familiar to the Mood f-structure (verb first). Apparently an explicit Subject in a verse-line whose Predicator is realized by a y]qf{l functions as Marked Theme, provided that the

298

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

verse-line is co-extensive with a clause and part of a couplet (4b, 10a). Exceptional patterns are noted in the case of verse-lines 6a and 6b, which have Adjuncts as Marked Theme. Nine of the fifteen y]qf{l verbal forms have a Marked Theme in their clause. Of the eleven explicit Subjects, only three function as Marked Theme; in the remaining cases other constituents are preferred: a Vocative (2a) or Adjuncts (6a, 6b, 9b), the others not being charged with a significant thematic function (4a, 9a, 12b, 12c). These preliminary conclusions need further confirmation from other poems and especially from the QYYQ verbal sequence couplets that our database consists of. 6.2.2. Evaluation of the database The surveyed database includes 60 examples of q*f~l//y]qf{l verbal sequence couplets and 42 more examples of y]qf{l//q*f~l couplets, totalling 204 verse-lines and as many clauses. (1) With the exception of 14 verse-lines,1 the corpus contains only declarative clauses. The exceptions are either Wh- interrogative clauses2 or Yes/No interrogative clauses3. (2) In terms of polarity, one can notice a weak representation of the Marked (negative) variant.4 (3) Moodwise, there is only one couplet where mood announced in line A is altered in line B, an interrogative clause being preferred instead (11.3). Most likely, under the constraints of parallelism, the Psalm couplets under observation maintain a similar mood in both verse-lines. (4) One can also note that the Subject is not always explicitly mentioned in the text. When Subject is missing, the reference to it is derived from the verbal form. This is particularly the case with affirmative declarative clauses in SH. Therefore, the use of Subject charged with thematic markedness gives to such clauses an air of defamiliarisation that qualifies them for POETRY. HPy makes special use of personal pronouns as Subject/Marked Theme (cf. Table 6.3). Since most variations concerning the Mood structure are noted here, a table may prove useful. Bold fonts are used for y]qf{l//q*f~l couplets. 2.1A, 2.1B, 11.3B, 60.11A, 60.11B, 60.12A, 60.12B, 74.1A, 74.1B, 78.20bA, 108.11A, 108.11B, 108.12A, 108.12B. 2 2.1A, 2.1B, 11.3B, 60.11A, 60.11B, 74.1A, 74.1B, 108.11A, 108.11B. 3 60.12A, 60.12B, 78.20bA, 108.12A, 108.12B. 4 There are 23 examples: 3.7A, 5.6A, 7.14A, 18.37B, 26.5B, 30.7B, 44.13B, 58.9B, 60.12A, 60.12B, 66.18B, 78.64B, 89.44B, 95.11B, 101.3B, 104.5B, 108.12A, 108.12B, 119.11B, 126.2bB, 127.1bA, 127.1cA, 132.11aB. 1

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

299

As becomes clear from the Table 6.3 (y]qf{l//q*f~l couplets are in bold), an explicit Subject appears in 91/204 clauses. Given the number of explicit Subjects that function as Marked Theme (50/91, ~ 55 per cent), in relation to situations where there is no Marked Theme to accompany an explicit Subject (30/91) or where other constituents are the Marked Theme (11/91), it appears that Subject tends to conflate with the Marked Theme (Appendix 4). Although there are a few instances where, in both verse-lines, Subject is conflated with Marked Theme,1 Hebrew poets regularly avoided explicit Subjects conflated with Marked Theme in both verse-lines where they appear.2 Such ‘heavy’ constructions are completely avoided when a verse-line that is parallel with a verse-line, in which the Subject is conflated with Marked Theme, does not receive an explicit Subject.3 Subject person speaker poet (1st person)

God

1

Reference in Psalms explicit implicit 13.6aA; 30.7A; 18.41B; 26.4A; 26.4B; 88.14A; 119.69B 26.5A; 26.5B; 26.12B; 27.4aA; 27.4aB; 30.7B; 38.16A; 63.7A; 63.7B; 63.8B; 66.18A; 71.17B; 73.3A; 73.3B; 92.5B; 101.3bA; 116.2B; 118.10B; 119.11A; 119.11B; 119.95B; 119.101A; 119.101B; 119.121A; 3.7A; 23.6B; 28.1cB; 31.6A; 119.93A; 119.100A; 119.100B; 119.104A; 119.104B wanting 81.7A; 89.36A; 89.36B; 95.11A; 132.17A; 132.17B

q*f~l//y]qf{l: 27.10, 78.64, y]qf{l//q*f~l: 11.3, 38.12, 55.5, 83.3, 85.12,

87.6. 2 There are eleven such examples: 2.1, 6.10, 13.6a, 18.5, 65.4, 67.7, 88.14, 89.13, 104.26, 127.1b, 127.1c. 3 There are twenty four such examples: q*f~l//y]qf{l -18.41, 26.12, 30.7, 37.23, 38.5, 38.16, 49.15a, 50.19, 60.12, 65.12, 68.11, 73.9, 74.14, 81.7, 118.10, 119.69, 139.13, y]qf{l//q*f~l - 9.8, 23.6, 37.11, 41.4, 44.11, 50.3, 73.27.

300

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Subject person addressee poet (2nd pers sg) enemy God

addressee+ (2nd person pl) noninteractant (3rd person sg)

enemy God

Israel

extension of the poet other

Reference in Psalms explicit implicit wanting wanting wanting 50.19A 38.16B; 60.12A; 65.4B; 51.8A; 51.8B; 60.12B; 74.14A; 89.13A; 63.8A; 65.12A; 68.11B; 139.13A; 10.14bB 71.17A; 74.1bA; 74.14B; 92.5A; 104.6A; 108.12A; 108.12B; 119.82B; 119.121B; 139.13B; 5.6B; 8.7A; 8.7B; 18.37A; 28.1cA; 31.6B; 41.4B; 44.11A; 44.12A; 44.12B; 44.13A; 44.13B; 73.18A; 73.18B; 73.27B; 89.44A; 89.44B; 119.93B wanting 33.21B wanting 6.10A; 6.10B; 27.10B; 66.18B; 67.7B; 132.11aA; 9.8A; 41.4A; 87.6A; 126.2bB; 127.1bA; 127.1cA 37.23A; 68.11A; 10.14bA; 11.3B; 60.11A; 60.11B; 108.11A; 108.11B 13.6aB; 26.12A; 88.14B; 119.82A; 55.5A 49.15aB; 67.7A; 73.6A; 73.6B; 74.1bB; 33.21A; 49.9A; 50.3bA; 85.12A; 85.12B; 87.6B; 104.26B; 127.1bB; 127.1cB; 142.4bA

58.9A 7.14A; 7.14B; 37.23B; 78.20bA; 78.20bB; 104.5A; 111.5A; 111.5B; 116.2A; 132.11aB; 9.8B; 47.5A; 47.5B wanting

wanting 101.3bB; 104.5B; 49.9B; 50.3bB

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

Subject person enemies

noninteractant (3rd person pl)

other

301

Reference in Psalms explicit implicit 2.1A; 2.1B; 18.41A; 17.11A; 17.11B; 50.19B; 73.9B; 118.10A; 49.15aA; 62.5aA; 119.95A; 140.3A; 3.7B; 62.5aB; 73.9A; 83.6A; 5.6A; 44.11B; 83.3A; 83.6B; 119.69A; 140.3B; 83.3B; 139.20A; 139.20B 58.9B; 142.4bB 18.5A; 18.5B; 27.10A; 38.5B; 95.11B; 102.15B; 38.5A; 65.4A; 65.12B; 37.11B; 126.2bA 78.64A; 78.64B; 81.7B; 89.13B; 102.15A; 104.6B; 11.3A; 18.37B; 23.6A; 37.11A; 38.12A; 38.12B; 55.5B; 73.27A; 104.16A; 104.16B; 104.26A; 107.2A; 107.2B

Table 6.3: Distribution of Subject in couplets with QYYQ verbal sequence

(5) A Subject conflated with a Marked Theme appears more frequently with a y]qf{l verse-line. A Marked Theme with a y]qf{l in a primary verseline is more likely to trigger an explicit Subject becoming Marked Theme in line B than vice versa. With only one exception (Ps. 104.26), Marked Theme/Subject appears again in the secondary (q*f~l) verse-line after being present in the primary (y]qf{l) verse-line.1 (6) When Subject is implicit, Adjunct is more likely to function as Marked Theme preferably after a y]qf{l (23/53), as opposed to those that follow a q*f~l (10/53). In eighteen other cases a Complement functions as Marked Theme (cf. Appendix 4).

6.3. TRANSITIVITY STRUCTURE OF PSALM COUPLETS WITH QYYQ VERBAL SEQUENCE 6.3.1. Psalm 26 For the study of transitivity/ergativity structure of verses in Psalm couplets, another Psalm which contains couplets with QYYQ verbal sequence was selected. 1

Cf. Ps. 11.3; 38.12; 41.4; 55.5; 83.3; 85.12 and 87.6.

302

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Psalm 26 1A v*pf}n' yhwh a¨ [you] √vindicate me, Lord, Process: doing Goal Medium B K' - a&n' B=t%M' because I in integrity CManner: Actor Medium quality C Wb~yhwh in the Lord CLocation: place

vol / qtl // qtl / yqtl

h*l~kT' b¨ √walk Process: happening

B*f~jT' c¨ Oa a#mu*d d¨| [I] √trust do not [I] √slip Process: Process: sensing happening

This tercet has one finite verb on each verse-line, with the exception of the last. A Process of happening follows a different Process type each time. Whereas lines A and B have room for a constituent with the role of Medium, line C lacks it completely. Conjunctive and disjunctive forces are at work in all the verse-lines of this verse. Note that the first constituent functioning as Medium is the Goal in clause (a), but the Medium of clause (b) functions as Actor. A B=j*n}n' [you] √examine me, Process: Goal doing Medium 2

B

x*r=p> [you] √test Process: doing

yhwh a¨ w=n~S}n' b¨ Lord, and [you] √try me Process: Goal doing Medium

vol / vol // vol

k]lyot~y w=l]B' c¨| my inner me and my mind Goal Medium

The second couplet contains only volitives, two in line A and one in line B. With the exception of the vocative that appears in clause (a), the three clauses that of this verse are similar in their Mood structure. 3

A K'-j~sD+k* l=n#g#d u?n*y a¨ [is] your kindness before my eyes CLocation: place Existent Medium

nom // qtl

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

303

B w=h]th~L~kT' B~a&m]T#k* b¨| [I]√walk within your truth Process: happening CLocation: place

A simple experiential structure describes the third couplet of this poem, an existential clause (a) preceding clause (b) with a finite verb. The reason for its atypical structure may be its syntactical subordination to the previous couplet by means of K'. A Oa-y*v~bT' u]m-m=t}-v*w=a a¨ not [I] √seat with deceitful man CAccompaniment: Process: happening comitative 4

qtl // yqtl

B w=u]m n~u&l*m'm Oa a*boa b¨| and with hypocrites do not [I] √go CAccompaniment: Process: happening comitative

This is the first example of a couplet with a QYYQ verbal sequence in this poem, both verbal forms being negated. The experiential structure is similar but reversed in line B compared to line A (chiasmus). A c*n}at' [I] √hate Process: sensing 5

q=h~l m=r}u'm a¨ the assembly of evildoers, Phenomenon Range

qtl // yqtl

B w=u]m n~u&l*m'm Oa a*boa b¨| and with the wicked do not [I] √seat CAccompaniment: Process: sensing comitative

This new example of QYYQ verbal sequence couplet follows the paradigm of the previous example by having a similar Process in both verse-lines (this time a sensing Process), and by reversing the order of its constituents in line B. Here, the poet preferred not to have line B mirroring line A, so that the negated verb and the constituent that fronted the Process is a Circumstantial, not its Phenomenon.

304

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

A a#rj~x B=n]Q*yon K~P*y a¨ [I] √wash in innocence my palms CManner: Process: Goal doing means Medium 6

yqtl // yqtl

B w~a&s)b=b> a#t-m]zB~j&k* yhwh b¨| [I] √go around your altar, Lord CLocation: Process: happening place

Except for the similarity in verbal forms (both are y]qf{l), the two verse-lines of the couplet in verset 6 are completely different in terms of their experiential structure and Process choice. Comparing this couplet with the previous two, one cannot avoid noticing that dissimilarity of verbal form allows for similarity of experiential structure. A l~vm]~u to √make heard Process: saying 7

B=qol Tod> a¨ in sound of thanksgiving CMatter

inf c // inf c

B Wl=s~PP}r K"l-n]pl=aot#yk* b¨| and to √tell all your wonders Process: Verbiage saying Range

Again, identity in verbal forms (this time infinitive construct) is linked to diversity in the experiential structure. It is tempting to consider the two Processes as behavioural, but given their ability to project, they qualify as verbal Processes. 8

A yhwh a*h~bT’ m=uon B?t#k* a¨ Lord, [I]√love the dwelling of your house Process: Phenomenon sensing Range

qtl // Ø

B Wm=qom m]vK~n K=bod#K* ¨| and the place of the dwelling of your glory Phenomenon Range

This couplet is fragmentary verbwise. Its sensing Process (a q*f~l) of line A functions as a double duty verb, being elided in line B, which consists

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

305

only of an extra-long Phenomenon. Things do not change much in terms of verbal usage in the next couplet, except that the only verb is a y]qf{l this time. A a~l-T#a$s)p u]m-j~F*a'm not [you] remove with the sinners CAccompaniment: Process: doing comitative 9

n~pv' a¨ my soul Goal Medium

yqtl // Ø

B w+u]m - a~nv? d*m'm j~Y*y b¨| and with the men of blood my life CAccompaniment: Goal comitative Medium

Otherwise, both verse-lines of the couplet in verset 9 are perfectly symmetrical. Apparently, perfect symmetry of the verse-lines was not a temptation to Hebrew poets. A a&v#r-B'd?h#m z]M> a¨ in whose hands [is] evil devices CLocation: place Existent Medium 10

B w'm'n*m and their right hand Scope Range

m*l=a> √be full Process: doing

nom // qtl

V)j~d b¨| bribes Actor Medium

The relative particle subordinates this couplet to the previous one and, again, as in the case of the couplet in verset 3, an atypical structure of its verse-lines follows. It is to be noted that whilst line A is coextensive with a verbless existential clause, line B is developed as a material clause. A w~a&n' B=t%M' a}l}k a¨ but I in my integrity [I] √walk CManner: Actor Process: quality happening Medium 11

yqtl // vol / vol

b¨ c¨| B P=d}n' w=j*N}n' [you] √redeem me and [you] √save me Process: Goal Process: doing Goal doing Medium Medium

306

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Before concluding the poem, and as he did in the case of the second couplet, the poet prefers again a denser verbal sequence, with two similar verbal forms in line B (volitives). The experiential structure of the two verse-lines is completely different, as well as their choice of Processes. A r~gl' my foot Senser Medium 12

u*m=d> √stand Process: sensing

qtl // yqtl

b=m'vor a¨ on level ground CLocation: place

B B=m~qh}l'm a&b*r}k yhwh b¨| in congregations [I] √bless the Lord CLocation: Process: Phenom place sensing Range

The conclusion belongs to a QYYQ verbal sequence, the third in this poem. Although similar Processes (sensing) are used, the order of the similar constituents is again altered. Additional constituents are placed at the extremities of this couplet. According to the MT, this poem consists of 12 versets. In our opinion, they correspond to as many poetic units. All the verses are regular couplets with one verb per verse-line, except lines 1A, 1B, 2A, 11B with two verbs each, and 8B, 9B without any verbs. That means that there are 28 clauses in this poem, of which two have elided verbs. The author preferred, for the verse-lines of five verses, to have lexical verbs realizing the same Process type (2, 4, 5, 6, 7 ~40 per cent), and the verse-lines of five verses to share a particular voice (2, 4, 5, 7, 12 ~40 per cent). Consequently, the most regular verses are 2, 4, 5, and 7, representing 32 per cent of the total. Such preferences could owe much to parallelism as a poetical device in HPy. material 1a, 1b, 1d, 2a, 2b, 2c, 3b, 4a, 4b, 6a, 6b, 9a, 10b, 11a, 11b, 11c

mental 1c, 5a, 5b, 8a, 12a, 12b

verbal 7a, 7b

identifying 3a, 10a

Table 6.4: Distribution of Process type in Psalm 26 middle 1b, 1c, 1d, 3a, 4a, 4b, 5a, 5b, 6b, 7a, 7b, 8a, 10a, 11a, 12a, 12b

middletransitive 3b

Table 6.5: Distribution of voice in Psalm 26

operative 1a, 2a, 2b, 2c, 6a, 9a, 11b, 11c

medioreceptive 10b

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

307

If the observation of verbal sequence is added to this information, we notice that q*f~l//y]qf{l (verses 4, 5, 12) is among the most uniform and well represented verbal sequences of this poem. In fact, the majority of verbal forms is divided between q*f~l (8/24) and y]qf{l (8/24), with volitives having a good representation (6/24). Probably the most important conclusions that can be drawn from the analysis of such a poem can be synthesised in two principles. (a) The poet does not seek identical experiential structures for the clauses of the poem, nor identical constituent order for both verse-lines of the couplets. Identical experiential structure in both verse-lines triggers a change in verb usage, either by eliding the finite verb (8, 9) or by a preference for verbless clauses (3, 10). (b) When Processes realized by finite verbs in couplets with one verb per verse-line are similar, their verbal forms are not identical and the order of similar constituents is altered (4, 5, 12). The couplet in verset 7 is an exception, but its verbs are not finite (infinitive constructs). Finite verbs with identical forms do not admit similar experiential structure in both verse-lines (verse 6). Of course, these principles need to be verified by analysing other couplets in the Psalter. For now, we will check the validity of the second principle, since it refers directly to couplets with QYYQ verbal sequence. 6.3.2. Evaluation of the database Upon applying this analysis to our 102 verses database, several conclusions can be drawn from the transitivity and ergativity viewpoint. (1) Among the 204 lexical verbs used as Processes for the verse-lines that our database consists of, there are 113 material ones.1 As expected in the case of such texts, a large number of lexical verbs realize mental clauses (62)2. With the exception of behavioural clauses (41)1, the other types appear rather incidentally: verbal (4)2, existential (3)3, relational (3)4. 1 qfl//yqfl – 7.14; 17.11; 18.5, 41; 37.23A; 38.5; 49.15a; 50.19; 65.4, 12; 67.7; 68.11; 71.17A; 73.6, 9, 14; 78.20, 64A; 81.7; 88.14B; 89.13A; 95.11B; 101.3bB; 104.5, 6; 111.5A; 118.10; 119.69A, 121; 139.13; yqfl//qfl –8.7; 9.8; 11.3; 18.37; 23.6; 31.6; 37.11A; 41.4; 44.11, 12, 13; 50.3; 58.9A; 60.11; 73.18, 27; 89.44; 104.16B, 26; 107.2B; 108.11; 119.93B; 126.2bB; 127.1b, 1c; 132.17; 142.4. 2 qfl//yqfl – 2.1; 6.10; 13.6a; 26.5; 27.4a; 30.7; 37.23B; 51.8; 60.12; 62.5a; 63.7; 66.18; 73.3; 101.3bA; 102.15; 111.5B; 119.11, 69B, 82, 95, 101; 140.3; yqfl//qfl – 3.7; 5.6B; 10.14A; 28.1c; 33.21; 37.11B; 47.5; 55.5; 58.9B; 85.12B; 87.6A; 119.93A, 100, 104.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

308

(2) From the ergativity viewpoint, poets prefer middle clauses (no Agent, not even implicit). With its 129 examples, middle clause stands as a majority option.5 Clauses with Processes of saying, sensing, or behaving prefer this option, but there are also some doing Processes without Agent. The operative clause follows with its 72 examples.6 Very few exceptions are noted: one medio-receptive clause (30.7B) and two effective: receptive clauses (q*f~l//y]qf{l – 37.23A; y]qf{l//q*f~l – 87.6B). (3) On a closer look, the couplets that swap voice in their respective verse-lines allow one to notice a preference for y]qf)l verbal forms for a middle realization. verbal sequence qfl//yqfl

yqfl//qfl

line B: middle 37.23; 38.5; 71.17; 81.7; 89.13; 92.5; 104.5; 104.6; 116.2; 119.69 23.6; 37.11; 44.11; 44.13; 104.16

line A: middle 18.41; 30.7; 49.15a; 51.8; 88.14; 119.82 73.27; 83.3; 87.6; 107.2; 119.93; 142.4b

Table 6.6: Distribution of middle voice in couplets with QYYQ verbal sequence

(4) Couplets with QYYQ verbal sequence prefer lexical verbs that realize material and mental Processes, manifesting a strong tendency towards having the same Process type and voice in both verse-lines. When qfl//yqfl – 26.4, 12; 27.10; 38.16; 63.8B; 74.1b; 78.64B; 83.6; 88.14A; 89.13B, 36; 92.5; 108.12; 116.2; 132.11a; yqfl//qfl – 5.6A; 38.12; 83.3; 85.12A; 139.20. 2 qfl//yqfl – 71.17B; 95.11A; yqfl//qfl – 107.2A; 126.2bA. 3 yqfl//qfl – 10.14B; 49.9B; 87.6B. 4 qfl//yqfl – 63.8A; yqfl//qfl – 49.9A; 104.16A. 5 qfl//yqfl – 2.1; 6.10; 13.6a; 18.41A; 26.4, 5, 12; 27.4a, 10; 30.7A; 37.23B; 38.5aB, 16; 49.15aA; 51.8A; 60.12; 62.5a; 63.7, 8; 66.18; 67.7; 68.11; 71.17B; 73.9; 74.1b; 78.64; 81.7B; 88.14A; 89.13B, 36; 92.5B; 95.11; 101.3b; 102.15; 104.5B; 104.6B; 108.12; 111.5; 116.2B; 118.10; 119.11, 69B, 82A, 95, 101; 132.11a; 140.3; yqfl//qfl – 3.7; 5.6; 9.8; 10.14b; 11.3; 18.37; 23.6B; 28.1c; 33.21; 37.11B; 38.12; 44.11B, 13B; 47.5; 49.9; 50.3b; 55.5; 58.9; 73.27A; 83.3A; 85.12; 87.6A; 104.16B, 26; 107.2A; 119.93A, 100, 104; 126.2b; 139.20; 142.4bA. 6 qfl//yqfl – 7.14; 17.11; 18.5, 41B; 38.5aA; 49.15aB; 50.19; 51.8B; 65.4, 12; 71.17A; 73.3, 6; 74.14; 78.20b; 81.7A; 83.6; 88.14B; 89.13A; 92.5A; 104.5A, 6A; 116.2A; 119.69A, 82B, 121; 139.13; yqfl//qfl – 8.7; 23.6; 31.6; 37.11A; 41.4; 44.11A, 12, 13A; 60.11; 73.18, 27B; 83.3B; 89.44; 104.16A; 107.2B; 108.11; 119.93B; 127.1b; 127.1c; 132.17; 142.4bB. 1

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

309

this rule is not followed, verse-lines with y]qf{l verbs are more likely to accompany a middle clause, as pairs for verse-lines realized by operative clauses. (5) Seventy-two of these couplets (~70 per cent) display the preference for using a similar Process in both verse-lines, with a slightly higher average among the couplets with the q*f~l//y]qf{l verbal sequence.1 This may well be a reflection of parallelismus membrorum, but it also shows that finite verbs prefer similar Process realization in couplets with different verbal forms (such as q*f~l//y]qf{l or y]qf{l//q*f~l). Whether verbs in these circumstances trigger a change in the order of constituents in the second verse-line or not is another matter. Of the seventy-two examples described by the above qualities, only in six cases do both verse-lines share a similar order of constituents (27.10; 28.1c; 38.12; 60.11; 85.12; and 108.11). At the other extreme, one finds fourteen examples of couplets with similar constituents but with their order altered (q*f~l//y]qf{l - 6.10; 13.6a; 18.5; 65.4; 67.7; 73.9; 119.82; y]qf{l//q*f~l - 44.12; ), or even reversed (q*f~l//y]qf{l - 26.4; 73.3; 81.7; 101.3b; y]qf{l//q*f~l - 104.16; 127.1c). Alternatively, an additional constituent appears preferably in the first verse-line of the couplet (q*f~l//y]qf{l - 7.14; 92.5; 102.15b; 119.11, 95; 140.3; y]qf{l//q*f~l 41.4; 47.5; 73.18; 139.20).2 A variant to this type of structure is the one where both verse-lines are loaded with an extra different constituent (q*f~l//y]qf{l - 2.1; 26.5, 12; 51.8; 63.7; 89.36; 139.13; y]qf{l//q*f~l 8.7; 119.104).3 Extra constituents can be added even in couplets where the constituent order does not change, as if to alter the impression that there is a similar order of constituents. Again, the prefered verse-line for such loading is line A (q*f~l//y]qf{l - 27.4a; 30.7; 78.64; 119.101;

1 There are thirty examples of couplets with different processes: qfl//yqfl – 37.23; 38.5; 49.15a; 63.8; 65.4; 71.17; 78.64; 88.14; 89.13; 95.11; 104.5; 111.5; 119.69; yqfl//qfl – 5.6; 9.8; 10.14b; 11.3; 18.37; 23.6; 37.11; 44.11; 49.9; 58.9; 73.27; 87.6; 104.26; 107.2; 119.93; 126.2b; 142.4b. 2 Exceptions to this situation are only 73.6; 78.20; 83.6; 127.1b, where the added constituent appears in the second line. 3 The heaviest structure from this perspective are in 74.1b, 14 (q*f~l//y]qf{l) and 3.7 (y]qf{l//q*f~l) with two different constituents added to each verse-line, and in 66.18; 68.11 (q*f~l//y]qf{l), which have both a different constituent and an extra one.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

310

y]qf{l//q*f~l - 33.21; 44.13; 89.44; 132.17).1 Some couplets, although they do not alter the order of constituents, change the constituents themselves (q*f~l//y]qf{l - 38.16; 50.19; 65.12; 118.10; y]qf{l//q*f~l - 31.6; 50.3b; 83.3; 119.100). The most loaded couplets are probably 62.5a (q*f~l//y]qf{l), 18.41 and 104.6 (y]qf{l//q*f~l), where the author preferred both to change constituents and to add extra constituents in line A, without altering the order of constituents as it appears in line B.

6.4. LOGICAL STRUCTURE OF PSALM COUPLETS WITH QYYQ VERBAL SEQUENCE

In order to illustrate the workings of the analysis of a clause complex, we have chosen two poems, The Song of the Sea as a representative for Archaic HPy and Psalm 7 for the Psalter. 6.4.1. Psalm 7 In the case of Psalm 7, the great majority of verbal forms belong to the y]qf{l type – of the 27 occurrences, 16 are volitives and 11 are indicative y]qf{l; only 8 are of the q*f~l type. We should bear in mind the fact that, in non-poetic texts, volitive y]qf)l occupies the first position only when it builds up the main verse-line of a ‘speech’ (Niccacci 1990, 76). In contrast to this, indicative y]qf)l, the verb form used as background in ‘discourse’, is always in second position (Niccacci 1990, 77f). Since it does not display poetical qualities, the superscript is not included in the following analysis. Psalm 7 1

2

A yhwh a$Oh~y B=k*

j*s't'

a¨

qtl//vol/ vol

O Lord, my God, in you [I] √seek refuge ×

2 1 =2

B hov'u}n' m]K"l-r)d=p~y b¨ w=h~X'l}n' c¨| [you] √deliver me from those chasing me and [you] √rescue me

This clause complex consists of two clause nexuses of which the second is nested. As can be seen, while line A is coextensive with a clause (q*f~l), line B includes two clauses, whose Processes are realized by volitives. Since all the clauses continue the idea of their precedent, they are all in a paratactic relationship. Nonetheless, clause (b) enhances the meaning Exceptions to this situation are found only in 17.11 and 119.121 (q*f~l//y]qf{l), which prefer to load the new constituent in line B. 1

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

311

of clause (a) by presenting the expected consequence, namely salvation. Clause (c) just elaborates on the meaning of clause (b). 1

3

=2 ×β

B

α

A P#n-y]fr)p K=a~ry}h n~pv' a¨ lest [he] √tear like a lion my soul

yqtl // nom / nom

b¨ w=a?n m~X'l c¨| P)r}q [while] √tear apart there isn’t rescuer

Verset 3 offers a second example of a nested clause that binds together two clauses realized by same verse-line predicators, in this case a participle and an existential particle. Should this syntactic device account for verselines with two Predicators? So far, all the previous examples investigated indicate that such an explanation is more than plausible. Strictly speaking, these examples do not necessarily fit the characteristics of verbal patterns of internal parallelism suggested by Watson (1985a, 366-9). Since the nested clause qualifies its precedent, probably the definition of internal parallelism needs further refining. 1

4

A yhwh a$Oh~y a]m - y*c't' z)at a¨ O Lord, my God, if [I] √do this

=2

B a]m- y#v - u*w#l if there is guilt

qtl // nom

B=k~P*y b¨| in my hands

A long protasis is launched here, is continued through the following verse, and its apodosis is given in a tercet in v. 6. The general, undefined meaning of the initiating primary clause is spelled out in the following clauses (4b, 5a, 5b), each of them expanding the idea of the preceding clause. 1

5

A a]m - G*m~lT' vol=m' r*u a¨ if [I] √repay my ally with evil

+2

B w*a&j~L=x> xor=r' r?q*m b¨| [I] √plunder my foe for nothing

qtl // wayqtl

Verset 5 offers the first of the two instances where the q*f~l//y]qf{l verb sequence appears in this poem. The q*f~l is fronted by the conditional particle and y]qf{l has a conjunctive w~w. Although it is elided in clause (b), the conditional particle is implicitly understood there too. Therefore, the two clauses are in a paratactic relationship of the extending type.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

312

1

1

6

A y]r~D)p aoy}b

n~pv' a¨ w=y~C}g

b¨

×2

let √pursue my enemy my soul +2

1 =2

vol / vol // vol // vol

and let [he] √reach

B w=y]rm)s l*a*r#x j~Y*y c¨ let [he] √trample to the ground my life C Wk#bod' l#u*p*r y~vK}n s#l> d¨| and my glory to dust let [he] √lay

The word-pairs n~pv'//j~Y*y//Wk#bod' support the cohesion of these three verse-lines. Lacking a closer synonymous meaning of √smr, we propose the following arrangement: a clause complex consisting of three clause nexuses, of which two are nested. All the verbs are volitives and are connected to one another paratactically. Clause (b) enhances the idea of pursuit, describing its result: reaching the pursued. 1

7

A qWm>

vol // vol // vol

yhwh B=a~PP#K* a¨

[you] √arise, O Lord, in your anger =2

B h]N*c}a B=u~brot xor=r*y [you] √rise up against the rage of my enemies

=3

C w=uWr> a}l~y m]vP*f Æx]ww't* cØ d¨| [you] √awake for me judgement Æ× [you]√commandØ

b¨

This tercet is a cohesive unit thanks to two sets of word-pairs, one of verbs (all volitives), the other of nouns: √~wq//√afn//√rw[, and @a//hrb[//jpvm respectively. Given that they belong to the same semantic field, it seems that they then fulfil all the requirements to qualify as wordpairs (cf. Watson 1984c, 128). With the exception of an enhanced embedded clause (c), the clauses that belong to this clause complex develop naturally by paratactic extension. 1

8

A w+u&d~t l+a%M'm T=sob=b#K* a¨ the congregation of nations √surround you

×2

b¨| B w=u*l#yh* l~M*rom vWb> and over them on heigh [you]√return

vol // vol

In verset 8, verbal forms of the same type (volitives) again realize the Processes of two clauses coextensive with their respective verse-lines,

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

313

connected by a paratactically enhanced relationship with the primary clause, producing a positive condition. 1 +2

9

1 =2

A yhwh y*d'n u~M'm a¨ O Lord, [you] √judge the nations

vol // vol // nom

b¨ B v*pf}n' yhwh K=x]dq' [you] √judge me, O Lord, according to my righteousness cØ d¨| C Wk=t%M' Æ u*l*y and according to my integrity Æ [that] is in me cØ

Verset 9 displays a tercet, whose verse-lines are coextensive with as many clauses. Alternative renderings are possible though. GHB prefers a couplet (5+3) split at the traditional disjunctive mark, A^tn*j. A more balanced division would preserve the ternary structure and allow the two circumstantials of manner to stay together in a third verse-line (3+3+2). In such a case, the principles of half-line parallelism are followed somewhat closely (Watson 1985b, 349-51), but only at the expense of syntactic parallelism. The current division has the advantage of construing clause (d) as an elliptical counterpart of clause (b), even though supplemented by an embedded clause (c). Note that the clauses of the first clause nexus are realized by same type verbs (volitives). 1

10 aA

+2

b ¨| B Wt=kon}n x~D'q but [you]√make firm the righteous

y]gm~r-n*a r~u r=v*u'm a¨ [you]√complete the evil of the wicked

vol // yqtl

Here a volitive and a Po y]qf{l contribute to the realization of a couplet where the clauses, coextensive with the verse-lines, are connected paratactically, the expanded clause having an opposite meaning to that of its primary counterpart. 1

bA

Wb)j}n

l]Bot Wk=l*yot c¨

[he]√examine hearts ×2

nom // nom

and kidneys

B a$Oh'm x~D'q d¨| God [is] fair

This verset continues the idea of the previous one with a more general thought. Again, the two clauses are in a paratactic enhanced [reason] relationship although, this time, the Predicators are different: a participle

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

314

and an adjective. Verset 11 (below) displays the reverse situation, but there the Predicator of the first clause is realized by a prepositional phrase. 1

11

A m~g]N' u~l-a$Oh'm a¨ my shield

×2

B mov'~u

nom // nom

[is] with God

y]vr? - l}b

b¨|

[he]√save the upright heart

A couplet with attributive clauses follows in verset 12, the secondary clause enhancing [cause: result] the meaning of its primary counterpart, to which it is connected paratactically. 1

12

A a$Oh'm v)p}f x~d'q a¨ God [is] a fair judge

×2

B w=a}l z)u}r B=k"l-yom b¨| a God [that is] indignant every day

nom // nom

Verset 13 consists of a couplet with double clauses on each verse-line, all but one being supported by y]qf{l verbal forms. Clause (a) is the protasis of a conditional clause whose apodosis – clause (b) – is continued by extended parataxis with two more clauses, (c) and (d). β α 1

×

+2 +3

13

A a]m-Oa y*vWb a¨ j~rBo y]lfov b¨ if not √repent his sword √sharpen

yqtl / yqtl // qtl / yqtl

B q~vTo d*r~k c¨ w~y=kon=n#h* d¨| his bow [he]√bend and [he] √string it

The second instance of the q*f~l//y]qf{l verbal sequence appears in verset 14, with the specification that the verbs are fronted by Marked Themes.

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

1

14

A w=Oa h}k'n K=l?-m*w#t to him he √prepare weapons of death

=2

B j]X*yw l=d)l=q'm y]pu*l his arrows for burning [he]√make

315

qtl // yqtl

a¨

b¨|

Verset 15 distributes its three clauses over two verse-lines. Clause (c) elaborates on the meaning of clause (b), to which is connected paratactically. Their Predicators are realized by q*f~l verbal forms. The primary clause, however, whose meaning clause (b) builds up by enhancement [reason], has a y]qf{l, and states the cause of the effect represented by clause (a). 1 ×2

15

1 =2

A h]N}h y=j~B#l - a*w#n a¨ Alas! [he]√pledge evil

yqtl // qtl / qtl

B w=h*r> u*m*l b¨ w=y*l~d v*q#r c¨| [he]√conceive trouble [he]√beget falsehood

Nesting the secondary clause of a double clause verse-line appears again in verset 16. Clause (b) extends paratactically the meaning of clause (a). The main clause nexus includes clauses (a) and (d) linked by enhanced parataxis [reason]: cause^effect. 1

1

16

A Bor

K*r> a¨ w~Y~jP=r}hW

b¨

+2

a cistern [he]√dig ×2

qtl / wayqtl // wayqtl / yqtl

and [he]√hollow it

B w~Y]P)l B=v~j~t Æ y]pu*l c Ø d¨| [he]√fall in the pit Æ×[he] √makeØ

Paratactic elaboration describes the relationship between the clauses that compose verse 17AB, a couplet with the same verb form (i.e., y]qf{l). The principle of correspondence upon which elaboration builds, is evident from the word-pairs present in the couplet: √bwv–√dry, var–dqdq, and lm[– smx. 1

17

A y*vWb u&m*lo b=r)avo a¨ √return his mischief on his head

=2

B w=u~l q"dq(do j&m*so y}r}d b¨| and upon his skull the violence √descend

yqtl // yqtl

316

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

The poem closes with a dedication couplet, expressed by two volitives, realizing the Process of the two respective clauses. Again, their paratactic relationship is strongly supported by the word-pair hdy–rmz. 1

18

a¨ A aod#h yhwh K=x]dqo [I] √praise the Lord for his righteousness

vol // vol

=2

B w~a&z~M=r> v}m - yhwh u#lyon b¨| [I]√sing the name of the Lord, Most High

Of the 26 clause nexuses, eight are nested; three more embedded clauses have to be added to their number. In terms of types of relationship between clauses, elaboration and enhancement are the preferred paratactic systems, being represented by ten and eight examples respectively. Another six clause nexuses display paratactic extension. Only two examples illustrate the hypotaxis relationship, both being developed through enhancement. In this context, vv. 5 and 14, which display the QYYQ verbal sequence prefer paratactic extension and enhancement, respectively. 6.4.2. Evaluation of the database Now we turn to our database of couplets with QYYQ verbal sequence and survey the 102 clause nexuses in it. They display an evident preference for paratactic relations, a tendency noted also in the case of the two poems previously analysed (cf. Table 6.7). taxis 1^2

logicosemantic =

+

× “ ‘

q*f~l//y]qf{l

y]qf{l//q*f~l

2.1; 6.10; 7.14; 13.6a; 17.11; 18.5; 27.4a; 38.5, 16; 49.15a; 50.19; 51.8; 60.12; 62.5a; 63.7; 65.12; 67.7; 74.1b; 81.7; 83.6; 89.13, 36; 92.5; 102.15; 104.6; 108.12; 111.5; 132.11a; 139.13; 140.3 18.41; 26.4, 5, 12; 27.10; 65.4; 68.11; 73.6, 9; 74.14; 78.20b; 88.14; 101.3b; 116.2; 118.10; 119.69, 95 63.8; 71.17; 73.3; 78.64; 104.5; 119.121 95.11; 119.82 30.7

8.7; 9.8; 10.14b; 18.37; 23.6; 33.21; 37.11; 38.12; 41.4; 44.12, 13; 47.5; 49.9; 55.5; 58.9; 60.11; 73.18, 27; 83.3; 89.44; 104.16; 108.11; 132.17; 139.20 31.6; 44.11; 50.3b; 85.12; 104.26 5.6 126.2b 87.6

PSALM C OUPLETS WITH QYYQ VERBAL SEQUENCE

taxis α^β

logicosemantic = + × “ ‘ embedded clauses

q*f~l//y]qf{l

317

y]qf{l//q*f~l

wanting wanting 66.18; 119.11, 101

wanting wanting 11.3; 28.1c; 119.93, 100, 104; 127.1b, 1c; 142.4b wanting wanting 3.7; 107.2

wanting wanting 37.23

Table 6.7: Distribution of clause nexuses in our database according to taxis and logico-semantic type

Translated into numerical values, the distribution of clause relationship types produces the following result: taxis 1^2

α^β

logicosemantic = + × “ ‘ = + × “ ‘ embedded clauses TOTAL

q*f~l//y]qf{l

y]qf{l//q*f~l

30 17 6 2 1 nil nil 3 nil nil 1

24 5 1 1 1 nil nil 8 nil nil 2

60

42

TOTAL 54 22 7 3 2 nil nil 11 nil nil 3

88

11

3 102

(1) It can be noted that the majority of clause nexuses prefers the system of parataxis (88/102). As far as the logico-semantic type is concerned, elaborating is clearly the most attractive solution (54/102), followed by extending (22/102) and enhancement (18/102). (2) In the case of hypotaxis clause complexes, the dependent clause precedes its respective dominant clause, with the exception of the examples from Psalm 119 (vv. 11, 93, 100, 101, 104). This draws attention to the author’s style in Psalm 119.

318

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

(3) When the systems of taxis and logico-semantic type intersect, the most common expression is paratactic elaboration (54/100), which is coherent with synonymous parallelism promoted by traditional biblical criticism. (4) Assessing the relevance of these conclusions, it can be seen that they are in line with the tendency displayed by the two poems previously scanned. Whether this tendency is generalized to all SH poems is still to be determined. However, if paratactic elaboration is cogently related to synonymous parallelism in Hebrew Psalms, then there are reasons to believe that this particular solution is the one most commonly represented in the Psalter.

7 THE CASE OF UGARITIC POETRY The previous SFG study of BH couplets with QYYQ verbal sequence revealed several important characteristics.1 Most likely under the pressure of parallelism as the main driving force of literary creativeness, lyric texts such as those found in the Psalter differ from their epic counterparts at various levels (cf. Chapter 1). After describing our method with samples from SH prose texts (Genesis to Kings) and approximate with confidence the rules that describe a SFG for SH (Chapter 5), we traced the specificity of f-structure of clauses in HPy in relation to QYYQ couplets (Chapter 6). Overall, HPy displays a tendency towards restricting to several basic options the variety of the phenomena noticed in non-poetic texts. Thematically, markedness appears to be involved more frequently in HPy than in prose.2 A large majority of the couplets with QYYQ verbal sequence in our database have in at least one of the lines a marked Theme. Such a solution is to be expected, since POETRY is in itself a defamiliarized type of discourse. As to the mood structure of clauses, one can notice that the couplets in our database prefer declarative clauses and a striking binding to same mood type in both lines. Unlike prose, adjacent clauses in HPy do not preserve the same Subject, so much so that literary critics of HPy were determined to speak of Personenwechsel (change of person).3 A swap from implicit to explicit Subject and vice versa can also be noticed.

1

This section was previously published in“He Unforrowed his Brow and Laughed ”: Essays in Honour of Professor Nicholas Wyatt, ed. by Wilfred G. E. Watson (AOAT 299, Münster: Ugarit-Verlag, 2007), pp. 267-302. 2 A study developed on the Isaac Cycle (Gen 25.19-28.9) indicated that there are twenty-nine marked Themes, and the poem in Gen 25.23 is highly marked (each of the four lines of the ternary structure, cf. Tatu 2007b). 3 J. Sperber, Der Personenwechsel in der Bibel, ZA 32 (1918): 23-33. Cf. Gevirtz (1973b, 170-1). 319

320

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

The transitivity structure of BH couplets reveals a preference for material Processes followed by mental and behavioural Processes and, unlike prose, a very low representation of verbal, existential and relational Processes. The ergativity structure indicates a preference for a middle realization, whereas prose is more likely to prefer a transitive one. Among the many possibilities available in terms of clause complex structure, parataxis is by far the commonest, and elaboration is the logicosemantic type it intersects with in most cases. Since parallelism is justified grammatically, including the verbal form choice, one may denote the couplets with QYYQ verbal sequence as the marked variation of couplets with same verbal form.1 It could well mean that the QYYQ verbal sequence stands as a marked means towards accomplishing the coherence of a couplet, producing unity through diversity.2 In terms of the verbal form that hosts the unfamiliar structure, y]qf{l displays strong preference for markedness, explicit Subjects, middle realization and logico-semantic type. This last section of my work will provide some data from the Ugaritic literature, prompting a comparative study of the phenomena in HPy and UPy, and will suggest further developments in the context of our contributions. In the field of Ugaritic studies and their bearings on the Hebrew grammar and literature, one name needs to be singled out, that of C.H. Gordon. He addressed his concern for the verbal sequence in his Ugaritic grammar increasingly from one edition to the next. In his most recent work, he engages with the topic of the Ugaritic verb in the context of the Semitic verbal system (Gordon 1965, 67-91). Grammatically, he investigates the ‘tense’ value of y]qf{l and q*f~l. Epistemologically, he looks for the difference between the verbal usage in PROSE and POETRY. Terminologically, he considers what y]qf{l and q*f~l stand for (Gordon 1965, 67-9).

1 This implies more than a variation of antithetical grammatical parallelism (Buth 1986). 2 Wendland (1995, 89ff) promotes the retention of same tense/aspect over adjacent lines as an expression of grammatical cohesion. If that is the case then couplets with QYYQ verbal sequence function rather disjunctively, marking different stages of a poem’s argument, a suggestion that cannot be substantiated either by our studies in the psalms selected for SFG analysis or with a careful study of the book of Amos. In Amos, the only genuine couplets with QYYQ verbal sequence we could find appear together as a unit (3.7AB; 3.8aAB; 3.8bAB).

THE CASE OF UGARITIC POETRY

321

Concerning the ‘tense’ value of y]qf{l, it can not be proved with precision whether this form stands for both future (y~q~ff~l) and past (y~qf%l). The data is insufficient both in the Ugaritic and in the Qûmran literature. On the difference between the verbal usage in PROSE and POETRY, he noticed that in the administrative texts past action (q*f~l) is sharply distinguished from future action (y]qf{l). Nevertheless, the bulk of the data comes from poetic literature, therefore matters related to syntax have regard mainly to the poetic usage of verbs.1 He noted that terms like y]qf{l or q*f~l are not used in relation to time or aspect categories on the basis of the following reasons: (a) y]qf{l usage ranges from present to future – therefore it may be called ‘the universal tense’; (b) the y]qf{l//q*f~l (and vice versa) linkage with a great deal of interchangeablity; (c) conversive w does not appear in POETRY. C.H. Gordon also attempted to make several points concerning the verbal sequence of UPy. First, he noticed that there is no single paradigm for the structure of the syntactical components in a clause. Second, the QYYQ verbal sequence appears, in his view, randomly. Third, there is a tendency to use the same verbal forms for the corresponding lines (sometimes changed for grammatical reasons, such as avoiding Gt verbal constructions) (Gordon 1965, 122-3). The poetic forms of Ugaritic literature include parallelism of thought (climactic verse as well), formulaic language (themes, introductory formula), metre, strophic structure and fluent word order (pp. 131-44). Beside the rare use of w~Y]qf{l verbal forms, Ugaritic poems display only a few rigid principles, such as: (a) subordinating conjunction, interrogatives, interjections, and negative particles front the other clause components; (b) the construct must precede its genitive; (c) when suffixed by emphatic K, the verb stands at the end of the clause. Although the clause structure paradigm is very diverse, it can be noted that by far, the great majority of clauses follow the ‘Verb-initial (verb before the subject)’ type.2 As can be seen, Gordon is extremely cautious about 1 This situation is no longer exact. See KTU. Most Ugaritic texts published so far are administrative and other lists, royal grants and letters. Gordon warns against avoiding rigid evaluation of the non-poetical texts, which, by their allocated function, have a high level of formulaic structures and less pure narrative passages. 2 Out of the 52 examples given, 13 follow the paradigm ‘verb-before-subject’ (we include here also the cases of ‘verb-before-subject’ with X material fronted, because the relation between Verb and Subject is not altered): 13.120-13.129, 13.135, 13.140, 13.162. As opposed to these there are only seven cases where

322

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

pronouncing a more elaborate assessment of Ugaritic word order because, from his strict grammatical viewpoint, there was no other logical conclusion he could reach. Although he promotes the idea of verse-lines arranged in couplets, as more or less strict arrangements of three stressed words described by parallelism of thought, Gordon (1965, 131-2) does not look for the consequence of such a situation for the verbal sequence in view. In one of his early articles on Biblical literature, Cassuto (1942-43) investigated the connections Biblical and Ugaritic literatures have with the older Canaanite literature. Making exclusive use of rhetorical criteria, Cassuto proved that there are various poetic devices, which both literatures use successfully with similar function.1 Consequently, there is a strong case for Ugaritic and Hebrew literatures to have sprung from a common stem, that is Canaanite, and for the originality of Biblical literature to reside only in ‘its content and spirit’ but not in its form (Cassuto 1942-43, 59). Fifteen years later, John Gray (1957) re-examines the Ugaritic documents in order to assess their importance to the Old Testament literature. This time, the Scottish scholar goes beyond rhetorical devices and makes use of the mythological, heroic, religious and social connections. He itemizes many literary and linguistic connections between Ugaritic and Biblical literature. If we agree with Craigie (1981, 100) that during the fifty years following the discoveries at Ras Shamra a ‘mass of publication’ on the comparative Hebrew-Ugaritic studies has been printed, then there is no need to go over old ground again. What we hope to find by investigating the Ugaritic literature is evidence of couplets with QYYQ verbal sequence being older than that provided by HPy, that will allow us to connect the phenomenon in HPy to a literary tradition.

7.1. THE UGARITIC TEXT As opposed to the Masoretic text of the HB, which represents the climax of a well- established scribal tradition, the Ugaritic and Akkadian texts are not as standardized as one would like them to be. Nevertheless, they are Subject fronts the Verb (X-subject-before-verb included): 13.146-13.150, 13.153, 13.159. 1 Expressions for introducing another character, or introducing an important quotation, when transiting from one subject to another, repetition, rhythmic lines, word-pairs are all remnants of epic poetry left in the narratives.

THE CASE OF UGARITIC POETRY

323

important for the study of poetic devices in Semitic POETRY because similar means were in use by the Canaanite poets as well as by their Hebrew colleagues. Display and usage, standing for form and meaning, are two different things, and the editors/authors of the ANE texts seem to have known the distinction between them. In the case of Ugaritic texts, we cannot escape the unfortunate position of possessing a single copy manuscript. Its textual analysis is restricted to comparing various readings of the same original and/or comparing them to similar documents from cognate literatures. The primary text reproduced in this work comes from the printed reproduction of Herdner (1963) compared with its transliteration and the more recent work of Dietrich, Loretz and Sanmartín (KTU2). Uncertain verse-line delimitation, obscure vocabulary, lack of vocalization, scribal mistakes and altered syntax further complicate the study of Ugaritic narrative poems (Watson 1999, 166-7). Reconstructed characters (degree of certainty varies) will appear in our text in square brackets. Original word dividers were transliterated by a dot. We believe that dividing the blocks of letters into their respective poetic components, by providing the necessary word-dividers where they are suspected to be missing, is part of the interpretation Process. Therefore, the original layout of the text on the tablets is altered due to the rendering of the original into poetical lines (based on parallelism and rhythmical constraints). Translations of the Ugaritic texts are provided by Gordon (1949, 1965), J. Gray (1964), G.R. Driver (1956), Cassuto (1975), Caquot-SznyczerHerdner (1974), Gibson (1977), del Olmo Lete (1981), Pardee (1988, 1997), Margalit (1989), Aitken (1990), Smith (1994), Parker (1997), and Wyatt (2002) inter alia.

7.2. THE POETIC QUALITY OF UGARITIC NARRATIVE POEMS Scholars agree that UPy does not contain any traceable internal markers, other than the word divisor and the end of a line. The interpretation of the vertical wedge’s role in the Ugaritic texts is not monolithic. Sivan (2001, 11-12) argues that the Ugaritian scribes were inconsistent in both dividing words and in marking the end of lines. Tropper agrees with him, concluding that scribes were negligent upon noting a frequent trespassing into the right hand side margin of the column and bearing in mind that, in economy texts and letters, line dividers coincide with word

324

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

dividers (Tropper 2000, 70). Korpel (2005) too qualifies as careless the use of word divisors in Ugaritic cultic texts. None of the aforementioned scholars, though, would go as far as Young in seeing this apparent careless use of word divisors as an expression of an illusory regular metre in UPy (Young 1950, 132). Segert (1987) is of a more balanced opinion. Upon surveying the use of word divisors in the Ugaritic tablets, he concludes that they were employed for various reasons in accordance with the literary genre of the text contained by each tablet (Segert 1987, 284-6). The use of word divisors in poetic texts seems to follow prosodic constraints, exemplary in this regard being the hymn to ‘Anat (KTU 1.13), where the graphic lines marked at the end by a word divisor coincide with verse-lines. Apart from this poem, the other poems (e.g., KTU 1.100, 1.20, 1.21, 1.17-19, 1.1-6, 1.24) make use of the word divisor less rigidly, allowing Segert to assume that practical reasons were in place, as it were the scribes tried to use the writing space more economically (cf. Segert 1987, 286-8). Watson could not find much regularity in terms of lineation either. Too few texts have good lineation (Watson 1982, 311),1 some employing horizontal lines for delimitating strophes and stanzas (p. 312).2 Mabie (2004) identifies seven lineation types and attributes to each of them various structural functions.3 Horwitz represents the optimistic camp. He showed statistically ‘that in more than 95 per cent (1898/1997) of the instances in mythological texts, the end of a line on a tablet marks word division’ (1973, 169). In the nonmythological texts, the percentage is slightly lower (Horwitz 1977, 127). Of the two markers, Horwitz credits the latter to be the oldest, noticing that the word divisor is missing in many texts (Horwitz 1977, 128). If Horwitz’s theory is correct, then all the anomalous cases can be explained as results of a late introduction of the vertical wedge as a literary innovation with various and unstable functions. Occasionally, the scribe 1 Watson included the following texts: 10, 47, 65, 71, 72, 81, 85, 96, 97, 102, 118, 123, 140, 154 and 161, as well as the ‘lists’. 2 KTU 1.13, 43, 86, 101, 103, 104, 105, 111 are given as examples for texts with ruled strophes, and KTU 1.23 (obv), 40, 42, 43 and 100 for texts with ruled stanzas. 3 Most of them are related to the ideational structure of the texts, not to their syntactic structure as it is too strongly assumed in the title of his article. Even in the case the double horizontal line in the divination texts where a syntactical significance is attributed to it (Mabie 2004, 298), the function is more structural than anything.

THE CASE OF UGARITIC POETRY

325

decided to split a word over two adjacent lines, seemingly struggling to accommodate rhythmic, grammatical and aesthetic criteria within the limits of the given scribal tradition (Horwitz 1979, 393). Horwitz entertains the idea that the vertical wedge was employed occasionally as a metrical device as well (Horwitz 1973). In his study, Horwitz shows statistically that there is a high probability for the word divider to function as a metrical device. Given the particularities of his study (sample size of 25 and a computed probability of +0.63), in order for the coefficient to equal zero there is a chance of 1 to 1000. Horwitz (1973, 171) proves the applicability of this study to the understanding of the metrical structure of text KTU 1.3 iii 38-45, where each primary line has three vertical wedges and the secondary line has four. Another sample could be KTU 1.100 (Horwitz 1979, 394). Without being dogmatic about his discovery, following Greenberg, Horwitz qualifies the linguistic forces at work behind Ugaritic texts as being folk linguistics (Horwitz 1973, 172). Thus, lacking the rigidity of academic linguistics, Horwitz is right in concluding that vertical wedges could have been employed as metrical markers despite the fact that occasionally they altered the expected morphology and/or syntax. Speaking of rhythm in HPy and UPy, there is little certainty on the actual method known and employed by the ancient bards. Nevertheless, modern scholars have advocated different ways of calculating the metre. Whilst surveying the modern studies engaged in the debate over Ugaritic metre (and early HPy), Parker (1974, 283-7) identified three main positions: Albright, and counting syllables; Young, and the elusive metre; and Dussaud-Segert-Kosmala, and counting words. Pardee (1980, 117-24) surveys the more recent theories of Ugaritic metre. According to him, D.K. Stuart continues the Albright tradition of syllable count and B. Margalit proposes a theory of counting words that resembles the Dussaud-Segert-Kosmala approach and owes much to LeySievers metrical theory. In a later contribution, Pardee (1988, 4 n. 8) gives credit to Loretz for his theory that proposes counting consonants and to Freedman for counting vocables (p. 5 n. 9). At a closer look, Pardee dislikes all of them and notices that none of them could be used with the certainty of classical metrical verse in terms of regularity and predictability. Instead, a more valid approach would be to promote the musical interpretation of the original poems (chanting), using the quantitative measurements of Ugaritic verse-lines only for approximation. Then parallelism would be the main structural element in

326

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

UPy (Pardee 1980, 125-9). Parker too suggested that parallelism should be charged with metrical significance (1974, 287-90), although later he admitted that ‘while it cannot be demonstrated or recognized that metre is an essential feature of Ugaritic poetry, it must be granted that the length and rhythm of cola may be occasionally significant. (Parker 1989, 10).

Since 1981, Pardee defended strongly the thesis that the main poetic device UPy and HPy have in common is parallelism, only to support it with even more boldness in his 1988 work by comparing one poem from Ba‘al Cycle (‘Anat I) to a poem from Proverbs (chapter 2). Similarly, Segert (1979, 731) proposes parallelism as ‘the most prominent feature of Ugaritic poetry’.1 Greenstein (1974, 90-91) holds the opinion that in Ugaritic, unlike biblical Hebrew, the prevalent poetic form is the couplet with syntactically parallel lines and the main verb elliptical in the secondary line.2 According to Parker (1989, 10), parallelism is ‘universally admitted’ as the main structural principle of Ugaritic verse, whose most common manifestation is the couplet with its most common alternative, the tercet (pp. 13-16).3 Pardee (2000, 61b) also militates for the stability of the word pairs tradition in UPy as Watson did (1999, 181-3), and advocates the preference Ugaritic literature displays for narrative poetry, incantations and ritualistic texts, as opposed to liturgical, lyrical, sapiential and oracular poetry preferred by the HB (Pardee 2000, 62a). Whether word order qualifies as a distinctive characteristic of UPy is not yet certain. Nonetheless, ‘despite the enormous variety of word order in KRT’, Wilson (1982, 30-31) sees ‘a degree of predictable order’. Finite verbal forms in Ugaritic usually fill in the initial slot of a clause (pp 23-24). Exceptional behaviour is noticed in some of q*f~l cases surveyed (5/45) and some y]qf{l cases (14/45) when the verb comes last. y]qf{l verbal forms can occupy even medial positions (11/45 cases). Imperatives come 1 Other characteristics of UPy may include the organization of clauses in verses of two or more lines, the grouping of verses in larger sections (strophe and stanza), the use of repetition, word pairs, formulae and formulaic patterns, sound patterns, and figurative language (Watson 1999, 173-87). For a survey of similar studies see Watson (1999, 165 n. 2). 2 The other variation of Canaanite poetry Greenberg is talking about is climactic (staircase) parallelism (Loewenstamm’s expanded colon). 3 Obviously, as in HPy, in UPy parallelism has many forms depending on the level the cohesion between the component lines appears: semantic or syntactical (cf. Watson 1999, 169-73).

THE CASE OF UGARITIC POETRY

327

without exception at the head of the clause. Wilson could trace a certain ‘interchangeability of verb forms with no effect on word order or sentence structure’ most likely due to a ‘frozen’ state some poetic passages reached (p. 31). As rightfully noted by Watson (1989, 442), Wilson’s research takes no account of parallelism between verbal forms and the peculiarities of various text types. In conclusion, parallelism has a constrictive force on the poet allowing him to accommodate in the secondary line only those units that will fit the standard imposed by the primary line and achieve the desired effect. Rhythmic quantitative measurement of the lines should be allowed only to approximate the number of units per line (seemingly mostly three units, but occasionally four or even two units). As always in literary works, one should be aware that texts are the product of two opposite forces: standardization and conformity on one hand, originality and nonconformity on the other.1 Therefore, whilst looking for standardized structures, we need to try not to impose a priori rigid paradigms on the poems.

7.3. THE UGARITIC DATABASE Several authors contributed suggestions of samples with QYYQ verbal sequence from the Ugaritic narrative verse, most of them having same root verbs. Our database will consider only those verses whose lines coexist as a literary entity (couplet), having only one finite verb per line functioning as Predicate for a clause coextensive with the verse-line. Therefore, all examples suggested should be tested in terms of poetic structure prompted by their inner parallelism forces. The grammatical analysis of their verbal forms can be checked against Tropper’s grammatical analysis of verbal forms (1992) and his Ugaritic Grammar (2000). Cassuto identifies several samples both in Ba‘al Cycle and in Aqht’s Legend. Cassuto

(1937, 149)

(1938, 128) (1938, 128 n. 56)

(1938, 137) 1

I D 114-115 I D 128-129 V AB ii 20-21 II AB iii 13-16 II AB i 1-23 I* AB i 16-17 I D 114-115 I D 128-129 II AB vi 47-48

(KTU 1.19 iii 6-7) (KTU 1.19 iii 20-21) (KTU 1.3 ii 20-21) (KTU 1.4 iii 13-16) (KTU 1.4 i 1-23) (KTU 1.5 i 16-17) (KTU 1.19 iii 6-7) (KTU 1.19 iii 20-21) (KTU 1.4 vi 47-48)

Or as Watson (1984a, 70-71) put it, ‘spontaneity’ and ‘tradition’.

328

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Although KTU 1.19 iii 6-7 has indeed a y]qf{l and a q*f~l, the two verbs belong to two adjacent verses and are not part of the same couplet. The next sample is a couplet with three volitive verbs. Another debatable sample follows, because this verbal sequence occurs as part of a triplet (Smith 1997, 108; Wyatt 1998, 75). It is true, the last line is verbless, but such a triplet, as a later development from a more regular couplet, stays out of our limited interest. The passage in 1.4 i 1-23 is by definition too large to accommodate a couplet with QYYQ and, besides, it does not contain such a verbal sequence. Provided one defines the limits of the verse more clearly, KTU 1.4 iii 14-16 is indeed a genuine sample. The most problematic of all is the passage in KTU 1.4 vi 47-48. Cassuto reconstructs the y]qf{l of the secondary line from a ‘yn’ left at the end of the primary line after dropping the ‘n’. Similarly, one can apply the same change to the following two couplets. Smith (1987, 134) follows Cassuto in this matter. Wyatt (1998, 107) gives credit to KTU2 and reads a ‘y[n]’ or ‘[yn]’ at the end of each of rows 47-51. Another good illustration of this case could be KTU 1.5 i 16-18 (Wyatt 1988, 117-8), against Smith’s suggestion to integrate these two lines into a triplet with the following line (1987, 142). Gordon touched in passing the idea of QYYQ verbal sequence, but his samples do not reflect our narrow understanding of this phenomenon. Instead, Gordon speaks of sequence of verbs whereby one or more y]qf{l verbal forms can follow one or more q*f~l verbal forms or vice versa. Actually most of his samples offered in a new section ‘Sequence of Verbs’ in his Ugaritic Textbook (1965, 122-3) belong to this category.1 There are some, though, that draw our attention. Gordon

(1955, 54 n. 1)

(1955, 54 n. 2) (1965, 122)

51: VII: 21 51: V: 87-88 127: 20 51: IV: 8 51: III: 14-16 127: 20-21

(KTU 1.4 vii 21-22) (KTU 1.4 v 25-26) (KTU 1.16 vi 20-21) (KTU 1.4 iv 8-9) (KTU 1.4 iii 14-16) (KTU 1.16 vi 20-21)

In both 1.4 vii 21-22 and 1.4 v 25-26 couplets, the secondary line has two y]qf{l verbal forms and, therefore, do not fit our narrow description of 1 E.g., 51: IV: 9-15; 1 Aqht: 155-6; 1 Aqht: 162-3; Krt 80-83; 51: II: 8-11; 51: III: 23-26; 51: VII: 21-22; 52: 37; 127: 32-34; 1 Aqht: 75-76; 2 Aqht: II: 25; 1 Aqht: 170-2; Krt: 157-160; 67: V: 17-20; 121: II: 5-6; 1 Aqht: 114-5.

THE CASE OF UGARITIC POETRY

329

QYYQ verbal sequence.1 The sample in Keret’s legend has only (w~y)y]qf{l verbal forms. As for the sample in 1.4 iv 8-9, even if one would admit line 8 along with line 9, which is in itself less probable because verse-units in line 8 do not parallel any of those in line 9, one does not have a proper QYYQ sample because the secondary line would have two q*f~l verbal forms. Gordon recalls one of Cassuto’s samples, that is 1.4 iii 14-16, and this is actually the only genuine sample admitted in our database. Held took into consideration only those cases where verbs involved in QYYQ verbal sequence were of the same root and added to them samples where w~Y]qf{l replaced the y]qf{l. Held

(1962, 283) (1962, 284) (1962, 285) (1962, 285 n. 6) (1962, 288)

II AB vi 38-40 V AB, B 38-41 V AB, D 86-88 I* AB i 16-17 I D 114-115 I D 128-129 I D 142-143 V AB B 40-41 II AB iii 14-16

(KTU 1.4 vi 38-40) (KTU 1.3 ii 38-41) (KTU 1.3 iv 43-44) (KTU 1.5 i 16-17) (KTU 1.19 iii 6-7) (KTU 1.19 iii 20-21) (KTU 1.19 iii 34-35) (KTU 1.3 ii 40-41) (KTU 1.4 iii 14-16)

The sample that opens Held’s list can be accepted without further comment, as well as the following one, provided we allow a finer tuning of its reference. Indeed, lines 40-41 host the couplet with QYYQ verbal sequence (cf. Held 1962, 288). Those samples identified in Aqhat’s Legend can be accepted as valid as long as the material is carefully displayed in poetic lines and verses. Against Parker (1987, 72), we prefer to see lines 1.19 iii 8-11 (22-25) organized in two verses, a couplet with a QYYQ verbal sequence and a triplet (del Olmo Lete 1981, 394). As for KTU 1.19 iii 6-7 (repeated in 20-21 and 34-35), we believe that there is enough reason to separate the y]qf{l from its q*f~l counterpart as parts of two adjacent distinct couplets (del Olmo Lete 1981, 393). Actually, in all these cases, the y]qf{l stands as the verb of a couplet’s secondary line concluding Dan’el’s speech, and the q*f~l is the verb of another couplet’s primary line starting a narrative section. Held recalls both ‘traditional’ samples in 1.4 iii 14-16 and 1.5 i 16-18. In conclusion, our original number of QYYQ samples has been enlarged with five new ones.

1

Tropper (1992) takes ‘xjq’ as infinitive.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

330

Held (1962, 281 n. 2) noticed that the y]qf{l//q*f~l verbal sequence is more frequent than the q*f~l//y]qf{l variant with verbs of same root since y]qf{l is preferred in the primary line of a parallelistic couplet. Taking into consideration samples with verbal forms built on different verbal roots, this situation remains unaltered. Moreover, Held’s inference that the QYYQ verbal sequence ‘is primarily attested in the Baal epic, rarely in Aqht, and never in evidence in Keret’ (Held 1962, 286 n. 4) is not substantiated by our findings. Two more samples will be added to our database, another one (1.14 ii 43-45) to join Held’s sample from Keret’s legend, and a new sample from Aqhat’s legend (1.19 iv 58-59). The latter has indeed its finite verbal forms built on the same root. As for its antiquity and later adoption in HPy, we have to postpone our conclusions until a proper analysis of the database is completed. Marcus adds to our database three more samples, potentially even five, but two of them employ the QYYQ verbal sequence outside the limits of a couplet. KTU 1.3 ii 20-21 (Marcus 1969, 56) stand as a triplet, even though the ternary line is verbless. Similarly, KTU 1.23:33-35 is a four lines verse (Marcus 1969, 56), whose first pair of lines is entirely repeated by the second pair with the exception of the verbal form.1 Noticing that the last three samples are verbatim reiteration of the same formula, the following samples stand as valid cases of couplets with QYYQ verbal sequence: Marcus

(1969, 57)

Marcus

(1970-71, 107)

I AB vi 30-31 I* AB II 6-7 II AB ii 17-20 V AB D: 30-32 1 D: 94-96

(KTU 1.6 vi 30-31) (KTU 1.5 ii 6-7) (KTU 1.4 ii 17-20) (KTU 1.3 iii 34-35 ) (KTU 1.19 ii 45-49)

Watson promotes Niccacci’s distinction between ‘discourse’ and ‘narrative’ as the two main text types in use in Ugaritic and Hebrew literature, whereby ‘narrative’ stands for the narrative verse, not for narratives per se. Watson (1989, 441) extended the tendency noted by Held to include verbs of different root, but for that he took into account the verbal sequence of this sort that appears across verses larger than the couplet. Restricting the samples to couplets with QYYQ verbal sequence one has to consider the following samples, wherein one can find some new ones: 1

J.B. Gray (1915, 63) called this ‘alternate parallelism’.

THE CASE OF UGARITIC POETRY Watson (q*f~l//y]qf{l)

(1989, 440) narrative

discourse

(y]qf{l//q*f~l)

narrative discourse

331

KTU 1.4 v 25-26 KTU 1.4 vii 21-22 KTU 1.20 ii 3 KTU 1.22 ii 22 KTU 1.100: 67-68 KTU 1.2 iv 32-34 KTU 1.14 ii 43-45 KTU 1.114: 17-18 KTU 1.4 vi 38-40 KTU 1.24: 30-31 KTU 1.4 v 26-27 KTU 1.19 iii 8-9

Two of Held’s samples are valid and, therefore, accepted: 1.4 vi 38-40 and 1.19 iii 8-9. Two others do not qualify for our research because they have two y]qf{l verbal forms in the secondary line: 1.4 v 25-26 and 1.4 vii 21-22. The samples in 1.2 iv 32 and 1.22 ii 22 are too fragmentary to make a case out of them. Although 1.20 ii 3 is echoed in 1.22 ii 22, the former has a supplementary line that makes of it a triplet. Lines 30-32 in KTU 1.24 are difficult to split into their respective verse-lines, containing three finite verbs that function as Predicates: yUrr, wyU[n], and wnU[n]. To these, one has to add a clause with a non-finite verb. Wyatt (1998, 339) drops the second and leaves one finite verb on each line, but these lines are not part of the same couplet. Del Olmo Lete (1981, 460) places the first two finite verbs in one line and produces a couplet with parallel lines and, therefore, his solution comes out of our interest. Following Gibson (1977, 129), Watson prefers to see each finite verb on a different line, thus producing three, of which the last two are part of the same couplet. Since the whole poem seems to be very irregular in terms of building verse lines across written lines, we consider the latter suggestion more credible. Even so, we are left with a final line that contains both a finite verb and a projected non-finite verbal clause. Therefore, we have to drop this sample. Additional valid samples can include: 1.4 v 26-27; 1.14 ii 43-45; 1.100: 67-68; and 1.114: 17-18. Whilst investigating the usages of qatala in Ugaritic literature, Smith (1994, 49-51) touches in passing the matter of ‘the parallelistic use of the indicative prefix and suffix forms’. Among his suggested samples, one can trace several already extant in the ‘tradition’ of the topic such as: 1.4 iii 1416, 1.4 vi 47-54 and 1.5 i 17 from Cassuto, 1.3 ii 40-41, 1.4 vi 38-40, 1.19 iii 8-9 from Held, 1.5 ii 7, 1.3 ii 20-21, 1.6 vi 30-31 and 1.23:33-35 from

332

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Marcus. His original four samples do not qualify to enter in our database – with one welcome exception though, that is KTU 1.19 iv 58-59. Two of the samples are denied entrance for having the QYYQ verbal sequence part of a triplet (1.5 v 19-21 and 1.5 vi 12-13), and the third for having multiple finite verbs on both lines of the couplet. Our survey will conclude only after retrieving Tropper’s contribution. In his monumental Ugaritische Grammatik, as a peculiar phenomenon specific to the verbal system, the German Ugaritologist included a generous section on ‘suffix forms next to a preffix form in parallelismus membrorum’ (pp. 711712). Giving credit to the previous work of Held and Watson, Tropper divides his samples into two categories: prefix form//suffix form and suffix form//preffix form. Some of the samples quoted by Tropper have already been proposed by other scholars, namely 1.19 ii 8-9, 1.19 iii 22-23, 1.3 ii 40-41, 1.3 iv 43-44, 1.4 vi 38-40 (y]qf{l//q*f~l), 1.20 ii 2-3, 1.100: 67-68 and 1.114: 17 (q*f~l//y]qf{l). Couplet 1.10 iii 7-8 is too fragmentary to be accepted. Other couplets have similar verb forms in both lines, such as 1.14 iv 21-23 (unlike 1.4 ii 43-45), 1.23: 37 and 1.114: 14-15. Four QYYQ samples cited by Tropper are arguably integrated in tercets (1.4 v 15-17, 1.5 v 19-21, 1.14 ii 32-34, 1.114: 19-20), or in quatrains, respectively (1.114: 11-12). As for 1.19 iii 53-54 and 1.14 ii 39-41, the y]qf{l’s are admittedly volitives (Tropper 2000, 726). After all these samples were pruned, the only genuine samples to enter our database are the following two, both displaying the q*f~l//y]qf{l verbal sequence: Tropper

(2000, 692) (2000, 732)

KTU 1.3 ii 3-5 KTU 1.4 vii 23-25

So far, our database includes fifteen samples, without considering their verbatim reiteration elsewhere in the Ugaritic literature. These texts are translated and analysed from a SFG perspective in Appendix 5. They are arranged in two categories, the Ugaritic couplets with a q*f~l//y]qf{l verbal sequence preceding those with a y]qf{l//q*f~l verbal sequence. The numbering of these samples includes even the repetitive couplets, although we quote the passage only once. Below the reference we give the encoded clause complex analysis. At the opposite right side of the reference we include the verbal roots of the verbal forms in the couplet. Various translation options are given in the footnotes.

THE CASE OF UGARITIC POETRY

333

Distributed according to the verbal sequence, our database includes now the following samples: y]qf{l//q*f~l 1.3 ii 40-41 1.3 iii 34-35 1.4 v 26-27 1.4 vi 38-40 1.5 i 16-18 1.5 ii 6-7 1.6 vi 30-31 1.19 iii 8-9

First proponent Held Marcus Watson Held Cassuto Marcus Marcus Held

q*f~l//y]qf{l 1.3 ii 3-5 1.4 iii 15-16 1.4 vii 23-25 1.14 ii 43-45 1.19 iv 58-59 1.100: 67-68 1.114: 17-18

First proponent Tropper Cassuto Tropper Watson Smith Watson Watson

Our next section will collect and interpret observations from the four levels of SFG analysis we have conducted on these sample texts.

7.4. THE SFG ANALYSIS OF UPY COUPLETS WITH QYYQ VERBAL SEQUENCE

Our database includes nineteen samples, producing a total of thirty eight verse-lines coextensive with as many clauses. 7.4.1. Thematic f-structure of Ugaritic couplets with QYYQ verbal sequence After surveying the thematic structure of the couplets in our database, the following results emerge: MTheme//

Theme Theme//MTheme MTheme// MTheme Theme//Theme

q*f~l//y]qf{l (7) (2) (4) (1), (3), (6)

y]qf{l//q*f~l wanting (13) (14), (15), (18-19) (10-12), (16), (17)

Table 7.1: Distribution of Ugaritic couplets with QYYQ verbal sequence according to their thematic f-structure

The missing samples (5) and (8-9) do not have a thematic structure per se on one line or on both, respectively. SFG analysis of the thematic structure of the couplets in our database shows a preference for the y]qf{l//q*f~l verbal sequence against the other. Six couplets have a q*f~l//y]qf{l verbal sequence, and ten couplets have a y]qf{l//q*f~l verbal sequence. Of the thirty-two lines/clauses, nineteen have familiar

334

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

thematic structure (~60 per cent). The rest of them, thirteen lines/clauses representing ~40 per cent, display a defamiliarized thematic structure. Therefore, one cannot speak of a clear tendency towards markedness in couplets with QYYQ verbal sequence in Ugaritic literature. Numerically the statistics in Table 7.1 look as follows: MTheme//Theme

Theme//MTheme MTheme// MTheme Theme//Theme Total

q*f~l//y]qf{l 1 1 1 3 6

y]qf{l//q*f~l nil 1 4 5 10

Total 1 2 5 8 16

These thirteen lines with altered thematic structure are of particular interest to us. Subtracting the number of those cases where a Marked Theme appears in both lines (ten lines/clauses), one is left with three. In relation to the association of Marked Theme and the verbal form of its respective line/clause, one notices that there is only one line/clause with a Marked Theme in the y]qf{l clause (~33 per cent), whereas cases with a Marked Theme in the q*f~l clause are two (~66 per cent). Provided we had more samples, this might have suggested that in Ugaritic literature the q*f~l is less stable than the y]qf{l. It is to be noted that the cases of couplets with the thematic structure altered in both lines favour those with the y]qf{l//q*f~l verbal sequence (three versus one). 7.4.2. Mood f-structure of Ugaritic couplets with QYY Q verbal sequence The clauses in our database display a high degree of uniformity. All of them are declarative and affirmative clauses, without noting variations in terms of mood from one clause to the other. There is very notable difference between the two verbal sequences in terms of the way they employ the Subject (Table 7.2). With only one exception (sample (4)), the q*f~l//y]qf{l verbal sequence prefers to have a different Subject in the secondary line and an implicit one too. The y]qf{l//q*f~l verbal sequence displays more interest for having the same type of Subject in both lines and an explicit one too. The only exception is sample (13). When Subject is explicit, there is a strong tendency to mark it, especially in clauses with three different constituents. Whenever the clause has only two constituents, there is a clear tendency to leave the finite verb at

THE CASE OF UGARITIC POETRY

335

the head of the clause (samples (10-12), (16), (17)). Sample (6) could have been included too, provided its primary line had lost one of its three constituents, as its secondary line has had. In this case, the Subject being elided in the secondary line, had to be mentioned at least in the primary line. Since it stands as an exception, one should expect variations such as that in sample (7), where the explicit Subject is marked. It is to be noted that, in all the other cases, where the Subject is conflated with Marked Theme (samples (4)AB, (13)B, (14)AB, (15)AB, (18-19)AB), the clause has three components. explicit 3rd person 2nd person 1st person implicit 3rd person 2nd person 1st person

q*f~l//y]qf{l (4) AB; (5)A; (6)A; (7)A wanting wanting q*f~l//y]qf{l (1)B; (5)B; (6)B; (7)B (1)A; (2)A; (3)B (2)B; (3)A

y]qf{l//q*f~l (10-12)AB; (13)B; (14)AB; (15)AB; (16)AB; (17)AB; (18-19)AB wanting wanting y]qf{l//q*f~l wanting (13)A wanting

Table 7.2: Distribution of clauses in Ugaritic couplets with QYYQ verbal sequence according to the quality of their Subjects

Intriguingly, the conflation of Subject with Marked Theme is noted in seven cases with q*f~l (samples (4)A, (7)A, (13)B, (14)B, (15)B, (18-19)B) and in only two with y]qf{l (samples (4)B and (15)A). Considering that in most of them the markedness of the Subject comes along the q*f~l verbal forms, it seems natural to assume that it is bound to the form of the verb. Conflation of Complement with Marked Theme is noted in four cases and all in clauses with a y]qf{l finite verb (Cf. samples (2)B, (14)A, and (1819)A). In the case of the first sample, the poet preferred a casus pendens (Sivan 2001, 218), whilst the Subject is implicit in both lines. As for the last three samples, the Complement is conflated with Marked Theme although the clause has an explicit Subject. Nonetheless, in the secondary line the Complement goes back to its rightful place and the explicit Subject takes upon the Marked Theme function. Apparently, the poet sought this variation in order to refine the parallelistic structure of his couplet, resulting in a terrace structure (Watson 1984, 208-13), even when the repeated material is not reiterated verbatim but follows the same order of constituents.

336

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

7.4.3. Transitivity f-structure of Ugaritic couplets with QYYQ verbal sequence The SFG analysis of the transitivity structure reveals that of the thirty eight Processes used, the large majority are material (31/38), six are behavioural (samples (15)AB, (16)AB, (17)AB), and only one is verbal (sample (3)A). Similar Processes are used in both lines of the couplets in our database with only one exception (sample (3)). Again, parallelism proves to have had a strong grip on the grammatical structure of the couplets in Ugaritic literature. From an ergative perspective, one can notice that the preferences of the clauses in our database divide unevenly between effective: operative (transitive Process with Subject as Agent)1 and middle (intransitive Process with Subject as Medium)2. Again, the Process type proposed by the primary line is followed by the secondary line with two exceptions. In both cases, the primary line is middle. In sample (3), the secondary line is middletransitive, whereas in sample (13), the secondary line is receptive. Lines (5)A and (8-9)AB could not be included because they do not have an ergativity structure as clauses. 7.4.4. Logical structure of Ugaritic couplets with QYY Q verbal sequence There are nineteen clause nexuses in our database, the relationship between their clauses being described by parataxis. Two of them (samples (8-9)) are produced from embedded clauses paratactically joined together. Clause nexuses with the q*f~l//y]qf{l verbal sequence display a greater variety of logico-semantic relationships, and clause nexuses with the y]qf{l//q*f~l verbal sequence stay with expansion through elaboration (Table 7.3).

1 Cf. samples (1)AB, (2) AB, (4)AB, (5)B, (14)AB, (15)AB, (16)AB, (17)AB, (1819)AB. 2 Cf. samples (3)A, (6)AB, (7)AB, (10-12)AB, (13)A.

THE CASE OF UGARITIC POETRY

Taxis 1^2

Logico-semantic relationship elaboration (=)

q*f~l//y]qf{l

y]qf{l//q*f~l

(6), (7)

extending (+) enhancing (×) quotation (“) reporting (‘)

(1), (4) (2), (5) (3) wanting wanting wanting

(10-12), (13), (14), (15), (16), (17), (18-19) wanting wanting wanting wanting wanting (8-9)

a^b embedded clauses

337

Table 7.3: Distribution of clause nexuses in Ugaritic couplets with QYYQ verbal sequence according to their logico-semantic relationships

7.5. S UMMARY After surveying the samples in our Ugaritic database, it becomes clear that Ugaritic couplets with QYYQ verbal sequence display a strong tendency towards applying parallelism even at the inner level of the order of constituents, producing very regular expected structures. The SFG analysis reveals a secondary line that follows in most cases the directions proposed by the primary line in terms of the order of constituents, markedness, mood, transitivity, ergativity and even clause type. This uniformity betrays the fact that grammatical parallelism was indeed the driving force behind the composition of Ugaritic literature. In terms of QYYQ verbal sequence, we could notice that y]qf{l verbal forms promote stability and familiarity, whereas q*f~l verbal forms are closely related to marked Themes, Subject change, Process variation and irregular clause complex structure. The most familiar verbal content of an Ugaritic couplet is the one with a double-duty finite verb, mainly a y]qf{l, as the predicate of the primary line/clause, whose effect is felt over to the clause in the secondary line. Pivot structures, frequently employed both in UPy and HPy serve very well the purpose of a double-duty verb (Sivan and Yona 1998; Sivan 2001, 212-3).1 1 It is to be noted, though, that Sivan’s understanding of pivot is wider than Watson’s (1983, 214-21), the former including Watson’s terrace structure alongside the phenomenon of pivot.

338

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

This situation contradicts our findings from the Hebrew database in two directions. First, as opposed to Hebrew, Ugaritic does not display a clear preference for employing markedness in couplets with QYYQ verbal sequence. Second, unlike in HPy, in UPy q*f~l is the verbal form that hosts the defamiliarized structure and displays an increased interest towards markedness, explicit Subjects, middle realization and logico-semantic type. This evidence may be interpreted as a consequence of the fact that parallelism has a drive on the verb use stronger in Ugaritic than in Hebrew. Theoretically, this can be explained in two ways: first, as a departure of the Hebrew poetic tradition from its Ugaritic bed; second, as a departure from the proto-Semitic tradition of which both Ugaritic and Hebrew traditions are later derivations. At the same time, knowing that the samples in our Ugaritic database are rather epic than lyric as opposed to those derived from the Hebrew Psalms, one could also infer that the two literatures developing separately (not entirely without contact) produced different literary forms to fit each genre. A study of the Akkadian psalms discovered at Ugarit would have brought some more light on this issue.1 Unfortunately the study of the very few Ugaritic psalms (KTU 1.24, 1.101, and 1.108)2 did not reveal anything relevant to our topic. Avishur (1998, 253 ff) proposed three more lyric Ugaritic texts, KTU 1.119, KTU 101, KTU 108, KTU 1.65, but this did not change the situation at all. It is quite possible that HPy imitated literary norms of Akkadian tradition, but if that is true some evidence could be traced in the Akkadian poetry at Ugarit. This argument extends the long sensed connection between Hebrew Psalms and the Canaanite religious poetry,3 whether applied to particular Psalms or not.4

Mowinckel (1955, 14 n. 1) includes here RS 94 and RS 95 as published by R. de Langhe in Les Textes de Ras Shamra-Ugarit et leur Rapports avec le Milieu Biblique de l’Ancien Testament I-II (Paris 1945), vol. I, pp. 211 ff. 2 Mowinckel (1955, 15) refers to two fragments from De Langhe’s book, RS 13 and RS 43. 3 Cf. Dussaud 1941; de Vaux 1937; Mowinckel 1939; Baumgartner 1939; Albright 1942, 1945 inter alia for more general connections between the two literatures, as well as Patton 1944 and Coppens 1946 for more specific ones. 4 Cf. Psalm 82 (Morgenstern 1939; Mullen 1980), Psalm 48 (Morgenstern 1941), Psalm 110 (Widengren 1941), Psalm 88 (Widengren 1945), Psalm 64 (Albright 1950-51), Psalm 29 (Kloos 1986, 55-93; Avishur 1994, 39-110) inter alia. Coppens (1947, 173-7) discussed in passing Psalm 110.3, 6b and Psalm 82.7. 1

8 CONCLUSIONS

8.1. CONTRIBUTIONS OF THIS RESEARCH Given the standard use of verbal forms and verbal sequences in SH PROSE, the irregular and sudden shift from q*f~l to y]qf{l and vice versa as noted in numerous couplets of HPy cannot go unnoticed. This study argues on the evidence found in the Psalter that the QYYQ verbal sequence is a poetic device uniquely present in POETRY that allows the reader to distinguish POETRY from PROSE. Its presence is noted in 63 of the 150 Psalms of the Psalter, decreasing in later books. This result gives credit to the redactional theory that considers the Davidic Hymn Book is pre-eminent to other hymn collections, such as the Songs of Ascent or the Hallelu-Yah Psalms, which must have been added later. Psalms

Book I Book II Book III Book IV Book V TOTAL

with QYYQ verbal sequence

percentage

22 15 9 5 12 63

53% 48% 53% 30% 28% 42%

without QYYQ verbal sequence 19 16 8 12 32 87

percentage

TOTAL

47% 52% 49% 70% 72% 58%

41 31 17 17 44 150

Table 8.1: Distribution of couplets with QYYQ verbal sequence in the Hebrew Psalter

The number of couplets with the QYYQ verbal sequence is very low in comparison to the total number of couplets in the Psalter, only 102 out of 2126 to be precise (cf. Appendix 7). Consequently, it can be stated that such a poetic device is far from having the generalized status parallelismus 339

340

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

membrorum enjoys in HPy. Its role must have been more specialised than sponsoring similarity between adjacent lines. The QYYQ verbal sequence is unsatisfactorily linked with authorial style, being noted in use in poems of various poetic tradition, whether Davidic, Asaphic, Korahic or otherwise. It can be argued that its use belongs rather to diachronic instantiations than to a particular authorial style. With its eight samples (cf. Appendix 6), Ancient HPy suggests that this poetic device was long known before the Classical era when the arts flourished in Ancient Israel. Previously noted as occurring only with similar root verbal forms (Held, Watson), the study of the QYYQ verbal sequence was extended to include samples of different root verbal forms (Dahood, Buth). It was initially believed to be a rare poetic feature with the y]qf{l//q*f~l more frequent than the q*f~l//y]qf{l. More recently, the odds favour the latter rather than the former. Dahood identified sixty-two samples (42 q*f~l//y]qf{l, 20 y]qf{l//q*f~l), and Buth reported ninety-one samples (54 q*f~l//y]qf{l, 37 y]qf{l//q*f~l), without providing the evidence, though. Even after adding the samples with a w~Y]qf{l instead of q*f~l, the results still not favour the y]qf{l//q*f~l verbal sequence. Looking closely, only some of these samples qualified to enter in the database created. The one hundred and two samples accepted in the database are all couplets with one finite verb per line, of which one is a q*f~l and the other an y]qf{l. Consequently, many of the samples quoted by the aforementioned scholars did not fit this definition. Although variations are noted, they were not given access, so that a more focused study would be allowed. So far, the solutions offered discussed the phenomenon as an expression of grammatical parallelism or with reference to pragmatics, mostly information structure theory, as an expression of markedness. This study confirms Kugel’s theory, followed by Berlin, Buth and Watson, that parallelism creates a ‘connection’ between two verse-lines and ‘separates’ them at the same time (Kugel 1981, 54-55), a perception that applies to Psalm couplets with the QYYQ verbal sequence too. The QYYQ verbal sequence accomplishes similarity and dissimilarity at the same time. Despite the suggestions that, in some instances, information structure is not sufficient to explain the deviations from the typical word order (Rosenbaum, Gross, and Lunn), this study remains within the limits of the linguistic theory selected. Apart from the situations where Lunn interpreted

CONCLUSIONS

341

a Marked Theme as defamiliarisation (altered only for aesthetic reasons),1 there are only few that were interpreted differently. Some spring from a different interpretation of various constituents,2 others from a different verse-line delimitation,3 with some still remaining insufficiently qualified.4 Despite this agreement, Lunn’s theory on the prominence of line A over line B does not seem to hold in the case of couplets with a QYYQ verbal sequence. It appears that a q*f~l verse-line with a regular Theme is more likely to trigger a Marked Theme in the y]qf{l parallel verse-line than vice versa, irrespective of the order of the lines. Thus, a q*f~l verse-line is more stable than a y]qf{l one. With its assumption that texts are the instantiation of both cultural values and functions intrinsic to lexicogrammar, the Systemic Theory appeals to the modern exegete of the HB. Having to reconstruct meaning from a text written in a dead language in order to reach the ideology that created that text, the exegete can find useful the tools that SFG offers. This work holds the view that the application of the Systemic Theory to the text of the HB allows the reader to understand linguistic phenomena that are less understood in traditional grammars. Besides offering an alternative interpretation of word order to that of the information structure theory, SFG comprehensively integrates it into an analysis that includes the observation of the same constituents at other levels too. The Systemic Theory holds the view that in lexicogrammar meaning is produced at three different levels: thematic (including information structure), interpersonal, and ideational (experiential and logical). This research attempts to show that SFG principles are of great value to exegetes. Although not new in the world of linguistics, with the exception of one work on the clause complex only, the Systemic Theory has not been applied to the text of the HB yet. This study proposes a tentative Systemic Grammar of Hebrew. When SFG principles are applied to Hebrew lexicogrammar, several peculiarities can be observed. 1 Examples include: 2.1B; 6.10B; 7.14AB; 13.6aB; 18.5B; 26.4B; 26.5B; 37.23AB; 63.7B; 63.8B; 65.12B; 73.3B; 78.20B; 81.7B; 83.6B; 92.5B; 102:15b (q*f~l//y]qf{l); 44.11B; 44.12B; 73.18B; 73.27A (y]qf{l//q*f~l). 2 Such as an expletive (118.10), mood Adjunct (127.1bB; and 127.1cB), circumstantial Adjunct (132.17A), and embedded relative clause (47.5B; and 142.4bA). 3 Examples include: 3.7; 23.6; 101.3a; and 132.11a. 4 Examples include: 17.11; 33.21; 58.9; and 119.100.

342

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

In SH, it is argued that the unmarked choice of mood is more variable than in English, being negotiated by two sub-systems, mood and verb finiteness. It is also held that most frequently Finite conflates with Predicator and, occasionally, even with Complement. Arguably, the system of Agency also includes the middle-transitive choice, where the process has a reflexive form and Subject is conflated with Reflexor. The prepositive question particle is used for the construction of polar questions. Polarity is never ambiguous, being always accomplished by negation adverbs. Vocatives are strongly preferred towards the end of the clause. On analysing the Psalm couplets with QYYQ verbal sequence, it could be noticed variations in terms of Theme content, explicit Subject, Process type, transitivity/ergativity and clause type. The couplets with a QYYQ verbal sequence are very unstable in terms of thematic structure. In such a couplet, at least one of the verbs has to be accompanied by a Marked Theme. When explicit, Subject tends to conflate with the Marked Theme, preferably in a y]qf{l verse-line. When Subject is implicit, Adjunct is more likely to function as Marked Theme preferably after a y]qf{l. HPy makes special use of personal pronouns as Subject/Marked Theme. The couplets that display a swap in the voice of their respective verselines allow one to notice the preference for y]qf)l verbal forms for a middle realization. Most couplets display a preference for similar processes in both verse-lines, with a slightly higher average among the couplets with the q*f~l//y]qf{l verbal sequence. At the logical level, the most favourable solution is paratactic elaboration. Summed up, all the differences we have noted between the verse-lines of a couplet with a QYYQ verbal sequence are of thematic and mood nature and justify employing such a device for aesthetic reasons. There are others, though, such as Process type, transitivity/ergativity, as well as clause type that follow closely the synonymous pattern. The results of this study concur with the general consensus among scholars that the Ugaritic literary tradition, conclusively older than its Hebrew counterpart, must have epitomized a peak of Canaanite artistry. The list of similarities between the literary texts produced by these cultures, generally explained as a loan from Ugaritic to Hebrew, cannot include the QYYQ verbal sequence, though. This work argues that the QYYQ verbal sequence as found in UPy differs from that found in HPy, in that the y]qf{l promotes stability and

CONCLUSIONS

343

familiarity, whereas the q*f~l is closely related to marked Themes, Subject change, process variation and irregular clause complex structure. Theoretically, this can be explained as a departure of the Hebrew poetic tradition from its Ugaritic bed, as well as a departure from the proto-Semitic tradition of which both Ugaritic and Hebrew traditions are later derivations.

8.2. RESOURCES FOR FURTHER RESEARCH A lot of ink has been spilt over the delimitation of verse-lines in poems in general, and the Psalms in particular. Without proposing to offer a definitive solution to an old debatable matter, this study produces a database of the Psalms, in Masoretic reading, with the text displayed in poetic verse-lines. The original text, as conveyed by the Codex Leningradensis (Masoretic accents included), and imported from Bible-Works 5.0, was placed at the core of the table that is annotated with the variants noticed in the works consulted. Verse-lines are numbered according to the Masoretic tradition, supplemented by the technical numbering of verses and verse-lines. In the margin, the verb content of each verse-line is supplied. Although the matter of translation was not approached as much as originally intended, based on the more refined understanding of verbal parallelism, this study confirms Dahood’s intuition to translate similarly both verbal forms that are part of the QYYQ verbal sequence. Since they are interchangeable in couplets with a QYYQ verbal sequence, they have to be translated similarly. This rule can be followed with confidence as long as the two verse-lines closely reflect each other. The context makes a highly valuable contribution to the understanding of time/aspect frame and function of each verbal form. Hence, the logicosemantic type indexes best the relationship between adjacent clauses. Thus, it can be argued that logico-semantic types such as elaborating and expanding can allow adjacent verbal forms to be translated in a similar fashion. Conversely, logico-semantic types of enhancing, locution and idea reflect a disjunctive relationship between two adjacent clauses and should allow for diversity in time/aspect function of the verbal forms.

8.3. RECOMMENDATIONS FOR FURTHER RESEARCH This study could have benefited from a careful analysis of the verbal group from a SFG perspective, with application to time and aspect references, which are not completely transparent in Hebrew grammar. More insights

344

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

could have been gained from an in-depth prosodic approach to the information structure, too. Although this study is limited to couplets with a QYYQ verbal sequence, the Psalter is rich in various alternatives to this regular structure. Nothing compares, though, with the analysis of couplets with the QYYQ verbal sequence in the context of the original poems to which they belong. Such a practice could be extended to good effect for the understanding of the phenomenon in HPy. A more detailed study of the QYYQ verbal sequence in HPy could contribute to the understanding of the poetic register, which can become a very valuable tool for the understanding of the HB texts, provided the narrative register, the genealogies register, the predictive register, etc. are understood from a SFG perspective. For more of a diachronic approach to the topic, one should consider the QYYQ verbal sequence in Akkadian poetry, particularly the Akkadian from Ugarit, as well as its presence in other poetic books both from SH and LH, whether lyrical or prophetic. Among the most significant documents in Akkadian from Ugarit the following can be mentioned: the Atrahasis fragment (RS 22.421), ‘The Righteous Sufferer’ fragment (RS 25.460), various prayers and hymns (RS 1979.25; 1980.389; RS 1980.383), as well as some wisdom precepts (RS 15.10; 22.439; 25.130; 23.034; 25.424) (Huehnergard 1989, 337-8).

BIBLIOGRAPHY Periodicals, Reference Works, and Serials ABC AET AnOr ANSS ANET ANSS ASTHLS BASOR BDB BHS BJRL BJS Bib Bsac BVSAWL-PH BWATH BZAW CBOTS CBQ CTL CBL DJD DULAT

Anchor Bible Commentary Abhandlungen zur evangelischen Theologie Analecta Orientalia Abr-Nahraim Supplement Series Ancient Near Eastern Texts (ed. J.B. Pritchard) Abr-Nahraim Supplement Series Amsterdam Studies in the Theory and History of Linguistic Science Bulletin of the American Schools of Oriental Research Hebrew and English Lexicon of the Old Testament (ed by F. Brown, S.R. Driver and C.A. Briggs) Biblia Hebraica Stuttgartensia Bulletin of the John Rylands Library Biblical and Judaic Studies Biblica Bibliotheca Sacra Berichte über die Verhandlungen der Sächsischen Akademie der Wissenschaften zu Leipzig - Philologischhistorische Klasse Beiträge zur Wissenschaft vom Alten Testament Herausgegeben Beihefte zur Zeitschrift für die alttestamentliche Wissenschaft Coniectanea Biblica, Old Testament Series Catholic Biblical Quarterly Cambridge Textbooks in Linguistics Collectanea Biblica Latina Discoveries in the Judaean Desert (of Jordan) A Dictionary of the Ugaritic Language in the Alphabetic Tradition 345

346

EB EJ FCI FOTL HS HSM HSS HTIBS HTR HUCA IDB IBHS IFG ILBS IOS IOSOT IEJ ISBL JAAR JBL JCS JETS JJS JLSP JNES JNSL JQR JQRS JR JSOT JSOTS JSS JTS KTU LSS

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS Encyclopaedia Britannica Encyclopaedia Judaica Foundations of Contemporary Interpretation The Forms of Old Testament Literature Hebrew Studies Harvard Semitic Monographs Harvard Semitic Studies Historic Texts and Interpreters in Biblical Scholarship Harvard Theological Review Hebrew Union College Annual The Interpreter’s Dictionary of the Bible An Introduction to Biblical Hebrew Syntax (Waltke & O’Connor 1990) An Introduction to Functional Grammar (3rd ed, 2004) Indiana Literary Biblical Series Israel Oriental Studies International Organization for the Study of the Old Testament Israel Exploration Journal Indiana Studies in Biblical Literature Journal of the American Academy of Religion Journal of Biblical Literature Journal of Cuneiform Studies Journal of the Evangelical Theological Society Journal of Jewish Studies Janua Linguarum Series Practica Journal of Near Eastern Studies Journal of Northwest Semitic Languages Jewish Quarterly Review Jewish Quarterly Review Supplement Journal of Religion Journal for the Study of the Old Testament Journal for the Study of the Old Testament Supplement Series Journal of Semitic Studies Journal of Theological Studies Keilalphabetische Texte aus Ugarit (ed by M. Dietrich, O. Loretz and J. Sanmartin) Language and Style Series

BIBLIOGRAPHY NHLS NIDOTTE NT OBO OLA OTS OTG PIB PPFBRHUJ PSWRA RB RSR RSV SBL SBEC SBLDS SBLMS SBLRBS SBLSCSS SBLWAWS SEL Sem SIL SILPLRF SILUTAPL SLCS SPSM SSN START SVTS TLL TRu TynB

347

North-Holland Linguistic Series New International Dictionary of Old Testament and Exegesis (ed by Willem van Gemeren) Novum Testamentum Orbis Biblicus et Orientalis Orientalia Lovaniensia Analecta Oudtestamentische Studien Old Testament Guides Pontificium Institutum Biblicum Publications of the Perry Foundation for Biblical Research, The Hebrew University of Jerusalem Pericope: Scripture as Written and Read in Antiquity Revue Biblique Religious Studies Review English Bible: Revised Standard Version Society of Biblical Literature Studies in the Bible and Early Christianity Society of Biblical Literature, Dissertation Series The Society of Biblical Literature, Masoretic Studies Society of Biblical Literature, Resources for Biblical Study Society of Biblical Literature, Septuagint and Cognate Studies Series Society of Biblical Literature, Writings from the Ancient World Series Studi epigrafici e linguistici sul Vicino Oriente antico Semitics Summer Institute of Linguistics Summer Institute of Linguistics Publications in Linguistics and Related Fields Summer Institute of Linguistics and The University of Texas at Arlington Publications in Linguistics Studies in Language Companion Series Studia Pohl, Series Major Studia Semitica Neerlandica Selected Technical Articles Related to Translation Supplements to Vetus Testamentum Series Topics in Language and Linguistics Theologische Rundschau Tyndale Bulletin

348

UBSMS UF VT WBC WFG ZA ZAW

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS United Bible Societies Monograph Series Ugarit-Forschungen Vetus Testamentum Word Biblical Commentary Working with Functional Grammar (Martin, Matthiessen & Painter 1997) Zeitschrift für Althebraistik Zeitschrift für die alttestamentliche Wissenschaft

BIBLIOGRAPHY

349

Aistleitner, Josef. Wörterbuch der ugaritischen Sprache. BVSAWLPH, 106.3. Berlin: Akademie-Verlag, 1963. Aitken, Kenneth T. The Aqht Narrative: A Study in the Narrative Structure and Composition of an Ugaritic Tale. JSS Monograph, 13. Manchester: Manchester University Press, 1990. Albright, W. F. Archaeology and the Religion of Israel. The Ayer Lectures of the Colgate-Rochester Divinity School 1941. Baltimore, 1942. _____. “The Oracles of Balaam.” JBL 63 (1944): 207-233. _____. “The Old Testament and the Canaanite Language and Literature.” CBQ 7 (1945): 5-31. _____. “A Catalogue of Early Hebrew Lyric Poems.” HUCA 23 (1950-51): 1-39. _____. Yahweh and the Gods of Canaan: A Historical Analysis of Two Contrasting Faiths. London: Athlone Press, 1968. Alexander, Patrick H. et al. The SBL Handbook of Style For Ancient Near Eastern, Biblical, and Early Christian Studies. Peabody, Mass.: Hendrickson Publishers, 1999. Allen, Leslie. Psalms 101-150. WBC, 21. Waco, TX: Word Books, 1983. Allgeier, Arthur. Die altlateinischen Psalterien: Prolegomena zu einer Textgeschichte der Hieronymianischen Psalmenübersetzungen. Freiburg im Breisgau: Herder, 1928. Alonso-Schökel, Luis. “Nota estilística sobre la partícula hNEhi.” Biblica 37 (1956): 74-80. _____. A Manual of Hebrew Poetics. Subsidia Biblica, 11. Trans. by Adrian Graffy. Rome: PIB, 1988. Spanish original, 1963. Alter, Robert. The Art of Biblical Narrative. London: George Allen and Unwins. 1981. _____. The Art of Biblical Poetry. Edinburgh: T. & T. Clark, 1990. Originally published, 1985. _____. “The Poetic and Wisdom Books.” In The Cambridge Companion to Biblical Interpretation, edited by John Barton, 226-40. Cambridge: Cambridge University Press, 1998. Althann, Robert. Studies in Northwest Semitic. Biblica et Orientalia, 45. Rome: PIB, 1997. Andersen, Francis I. The Sentence in Biblical Hebrew. JLSP, 231. The Hague: Mouton, 1974. _____. “Salience, Implicature, Ambiguity, and Redundancy in ClauseClause Relationships in Biblical Hebrew.” In Biblical Hebrew and

350

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Discourse Linguistics, edited by Robert D. Bergen. 99-116. Dallas, Tex.: SIL, 1994. Andersen, Francis I. and A. Dean Forbes. A Key-Word in Context Concordance to Psalms, Job and Proverbs. The Computer Bible, 34. Wooster, Ohio: Biblical Research Associates, 1991. Arnold, Bill T. and John H. Choi. A Guide to Biblical Hebrew Syntax. Cambridge: Cambridge University Press, 2003. Arnold, C.T., F.E. Kitchener, Alex. Potts and J.S. Phillpotts. The Psalms Chronologically Arranged: An Amended Version with Historical Introductions and Explanatory Notes. 2nd ed. London: Macmillan, 1870. Aronoff, Mark. “Orthography and Linguistic Theory: The Syntactic Basis of Masoretic Hebrew Punctuation.” Language 61 (1985): 28-72. Aviram, Amittai F. Telling Rhythm: Body and Meaning in Poetry. Ann Arbor, Mich.: University of Michigan, 1994. Avishur, Yitzak. Studies in Hebrew and Ugaritic Psalms. PPFBRHUJ. Jerusalem: Magnes Press, 1994. Auffret, Pierre. Hymnes d’Édypte et d’Israël: Etudes de structures littéraires. OBO, 34. Fribourg: Éditions Universitaires, Göttingen: Vandenhoeck & Ruprecht, 1981. _____. La Sagesse a bati sa maison : Etudes de structures littéraire dans l’Ancien Testament et spécialement dans les Psaumes. OBO, 49. Fribourg: Éditions Universitaires, Götingen: Vandenhoek & Ruprecht, 1982. _____. Voyez de vos yeux: Étude structurelle de vingt psaumes dont le psaume 119. SVTS, 48. Leiden: Brill, 1993. _____. Merveilles à nos yeux: Étude structurelle de vingt psaumes dont celui de 1Ch 16,8-36. BZAW, 235. Berlin: Walter de Gruyter, 1995. _____. Là montent les tribus : Etude structurelle de la collection des Psaumes de Montées, d’Ex 15, 1-18 et des rapports entre eux. BZAW, 289. Berlin : Walter de Gruyter, 1999. _____. Que seulement de tes yeux tu regardes…: Etude structurelle de treize psaumes. BZAW, 330. Berlin: Walter de Gruyter, 2003. Baillet, M., J.T. Milik and Roland de Vaux. Les ‘Petit Grottes’ de Qumran : Exploration de la falaise, Les grottes 2Q, 3Q, 5Q, 6Q, 7Q à 10Q, Le rouleau de cuivre. DJD, 3. Oxford: Clarendon Press, 1962. Bandstra, Barry L. A Review of (Meier 1992). Biblica 76 (1995): 568-70. Barr, James. The Variable Spellings of the Hebrew Bible: The Schweich Lectures of the British Academy 1986. Oxford: Oxford University Press, 1989. Barthélemy, D., and J.T. Milik et.al. Qumran Cave 1. DJD, 1. Oxford: Clarendon Press, 1955.

BIBLIOGRAPHY

351

Barton, John. “Classifying Biblical Criticism.” JSOT 29 (1984): 19-35. Bateman, John et. all. “Resources for Multilingual Text Generation in Three Slavonic Languages.” http://www.cdi.uni-aarland.de/publikationen/ softcopies/Bateman:2000:RMT.pdf, 2000. Retrieved at 20/10/2005. Baumgartner, W. “Ras Šamra und das Alte Testament.” Theologische Rundschau 12 (1940): 163-88. _____. “Ras Šamra und das Alte Testament.” Theologische Rundschau 13 (1941): 1-20, 85-102, 157-83. Bauer, Hans. “Die Tempora im Semitischen.” Beiträge zur Assyriologie und semitischen Sprachwissenschaft 8/1 (1910): 1-53. Beit-Arié, Malachi. The Makings of the Medieval Hebrew Book: Studies in Paleography and Codicology. Jerusalem: Magnes Press, 1993. Benoit, Pierre. “The Jerusalem Bible.” Review and Expositor 76: 341-9. Trans. by John B. Polhill, 1979. Bergen, Robert D., ed. Biblical Hebrew and Discourse Linguistics. Dallas, Tex.: SIL, 1994. Bergsträsser, G. Hebräische Grammatik. Band II: Verbum. Leipzig, 1929. Berlin, Adele. The Dynamics of Biblical Parallelism. Bloomington, Ind.: Indiana University Press, 1985. _____. Biblical Poetry Through Medieval Jewish Eyes. ISBL. Bloomington and Indianopolis, Ind.: Indiana University Press, 1991. Bliese, Loren F. “Symmetry and Prominence in Hebrew Poetry: With Examples from Hosea.” Discourse Perspective on Hebrew Poetry in the Scriptures. Edited by Ernst R. Wendland. UBSMS, 7. Reading: UBS, 1994. Bobzin, Hartmut. Die ‘Tempora’ im Hiobdialog. Marburg: Görich & Weiershäuser, 1974. Boecker, H.J. Redeformen des Rechtslebens im Alten Testament. WMANT, 14. Neukirchen: Neukirchener Verlagsgesselschaft, 1964. Bordreuil, Pierre and Dennis Pardee. Manuel d’Ougaritique. 2 vols. Paris: Geuthner Manuels, 2004. Brenner, Martin. The Song of the Sea: Exod. 15:1-21. BZAW, 195. Berlin: Walter de Gruyter, 1991. Bright, John. Review of (Kraus 1960) in Interpretation 15 (1961): 193-5. Briggs, Charles A. and Emilie G. Briggs. A Critical and Exegetical Commentary on the Book of Psalms. The International Critical Commentary. 2 vols. Edinburgh: T & T Clark, 1906. Brockelmann, C. “Zur Syntax der Sprache von Ugarit.” Orientalia 10 (1941): 223-40.

352

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Brongers, H. A. “Die Partikel ![;m;l. in der biblisch-hebräischen Sprache.” In Syntax and Meaning: Studies in Hebrew Syntax and Biblical Exegesis, 84-96. OTS, 18. Leiden: Brill, 1973. _____. “Some Remarks on the Biblical Particle h&Oa.” In Remembering All the Way...: A Collection of Old Testament Studies Published on the Occasion of the Fortieth Anniversary of the Oudtestamentisch Werkengezelschap in Nederland, edited by B. Albrektson et al., 180-4. OTS, 21. Leiden: Brill, 1981. Brown, Dennis. “Jerome and the Vulgate.” In A History of Biblical Interpretation, vol. 1: The Ancient Period, edited by Alan J. Hauser and Duane F. Watson, 355-79. Grand Rapids, Mich.: Eerdmans, 2003. Buth, Randall. “The Taxonomy and Function of Hebrew Tense-Shifting in the Psalms (qatal-yiqtol-yiqtol-qatal, antithetical grammatical parallelism).” START 15 (1986): 26-32. _____. Word Order in Aramaic from the Perspectives of Functional Grammar and Discourse Analysis. Ph.D. diss., University of California, Los Angeles, 1987. _____. “Topic and Focus in Hebrew Poetry — Psalm 51.” In Language in Context: Essays for Robert E. Longacre, edited by Shin Ja J. Hwang, and William R. Merrifield, 83-96. SILUTAPL, 107. Dallas, Tex.: SIL, 1992. _____. “Methodological Collision Between Source Criticism and Discourse Analysis: The Problem of ‘Unmarked Temporal Overlay’ and the Pluperfect/Nonsequential Wayyiqtol.” In Biblical Hebrew and Discourse Linguistics, edited by Robert D. Bergen, 138-54. Dallas, Tex.: SIL, 1994. _____. “Word-Order in the Verbless Clause: A Generative-Functional Approach.” In The Verbless Clause in Biblical Hebrew: Linguistic Approaches, edited by Cynthia L. Miller, 79-108. LSAWS, 1. Winona Lake, Ind.: Eisenbrauns, 1999. Butler, Cristopher S. Systemic Linguistics: Theory and Application. London: Batsford Academic and Educational, 1985. _____. Structure and Function: A Guide to Three Major Structural-Functional Theories. 2 vols. SLCS, 63, 64. Amsterdam and Philadelphia: John Benjamins, 2003. Butt, D. et al. Living with English. Sydney: Macquarie University Press, 1990. Buttenwieser, Moses. The Psalms Chronologically Treated with a New Translation. Chicago: University Pres, 1938. Callow, John. “Units and Flow in the Song of Songs 1:2-2:6.” In Biblical Hebrew and Discourse Linguistics, edited by Robert D. Bergen, 462-88. Dallas, Tex.: SIL, 1994.

BIBLIOGRAPHY

353

Callow, Kathleen. Man and Message: A Guide to Meaning-Based Text Analysis. Lanham, MY: University Press of America, 1998. Callow, Kathleen and John Callow. “Text as Purposive Communication: A Meaning-based Analysis.” In Discourse Description: Diverse Linguistic Analyses of a Fund-Raising Text, edited by W.C. Mann and Sandra A. Thompson, 5-37. Pragmatics and Beyond New Series, 16. Amsterdam: John Benjamins, 1992. Caquot, André, Maurice Sznycer and Andrée Herdner. Textes Ougaritiques. Tome I: Mythes et Légendes. Literatures Anciennes du Proche-Orient. Paris: Cerf, 1974. Cassuto, Umberto (David Moshe). “The Death of Baal (Tablet IAB of Ras Shamra).” In Biblical and Oriental Studies II, 146-67, 1975. Italian original, 1937. _____. “The Palace of Baal in Tablet II AB of Ras Shamra.” In Biblical and Oriental Studies II, 113-45. Italian original, 1938. _____. “Baal and Moth in the Ugaritic Texts.” In Biblical and Oriental Studies II, 168-77. Hebrew original, 1942. _____. “Biblical and Canaanite Literature.” In Biblical and Oriental Studies II, 16-59. Hebrew original, 1942-43. _____. The Goddess Anath: Canaanite Epics of the Patriarchal Age: Texts, Hebrew Translation, Commentary and Introduction. Trans. by Israel Abrahams. Jerusalem: Magnes, 1971. Hebrew original. _____. Biblical and Oriental Studies. Vol I: Bible. Trans. by Israel Abrahams. PPFBRHUJ. Jerusalem: Magnes Press, 1973. _____. Biblical and Oriental Studies. Vol II: Bible and Ancient Oriental Texts. Trans. by Israel Abrahams. PPFBRHUJ. Jerusalem: Magnes Press, 1975. Ceresko, Anthony R. Review of (Loretz 1979) in CBQ 43 (1981): 279-80. _____. Review of (Auffret 1993) in JBL 114 (1995): 304-5. Christensen, Duane L. Prose and Poetry in the Bible: The Narrative Poetics of Deuteronomy 1, 9-18. ZAW 97 (1985a): 179-89. _____. Review of (Auffret 1982) in JBL 104 (1985b): 133-4. _____. “The Numeruswechsel in Deuteronomy 12.” In Proceedings of the Ninth World Congress of Jewish Studies, Jerusalem, August 4-12, 1985. Division A: The Period of the Bible, 61-8. Jerusalem: WUJS, 1986. _____. “Narrative Poetics and the Interpretation of the Book of Jonah.” In Directions in Biblical Hebrew Poetry, edited by Elaine R. Follis, 29-48. JSOTS, 40. Sheffield: Sheffield Academic Press, 1987.

354

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

_____. “The Masoretic Accentual System and Repeated Metrical Refrains in Nahum, Song of Songs, and Deuteronomy.” In The Eighth International Congress of the International Organization for Masoretic Studies, Chicago, 1988, edited by E.J. Revell, 31-6. SBL-MS, 6. Missoula, Mo: Scholars Press, 1988. Claasen, W. T. “Speaker-Oriented Functions of K' in Biblical Hebrew.” JNSL 11 (1983): 29-46. Clifford, Richard J. A review of (Margalit 1980). CBQ 44 (1982): 125-7. Cohen, Harold R. (Chaim). Biblical Hapax Legomena in the Light of Akkadian and Ugaritic. SBLDS, 37. Missoula, Mo: Scholars Press, 1978. Collins, Terence. Line-Forms in Hebrew Poetry: A Grammatical Approach to the Stylistic Study of the Hebrew Prophets. SPSM, 7. Rome: PIB, 1978. Comrie, Bernard. Language Universals and Lunguistic Typology: Syntax and Morphology. Oxford: Blackwell. Chicago, Ill.: University Press, 1981. Cook, John A. “The Hebrew Verb: A Grammaticalization Approach.” ZA 14 (2001): 117-43. _____. “The Semantics of Verbal Pragmatics: Clarifying the Roles of w~Y]qf)l and w=q*f~l in Biblical Hebrew Prose.” JSS 49 (2004): 247-73. Coppens, Joseph. “Les Parallèls du Psautier avec les Texted de Ras-ShamraOugarit.” Museón 59 (1946): 113-42. _____. “Miscellanées Bibliques xi-xviii.” Ephemerides theologicae Lovaniensis. 23 (1947): 173-90. Craigie, Peter C. “Ugarit and the Bible: Progress and Regress in 50 Years of Literary Study.” In Ugarit in Retrospect: Fifty Years of Ugarit and Ugaritic, edited by G.D. Young, 99-111. Winona Lake, Ind.: Eisenbrauns, 1981. _____. Psalms 1-50. WBC, 19. Waco, TX: Word Books, 1983. Creach, Jerome F.D. Review of (Tate 1990) in Interpretation 47 (1993): 71-2. Cross, F.M., Jr., and D. N. Freedman. “The Blessing of Moses.” JBL 67 (1948): 191-210. _____. Studies in Ancient Yahwistic Poetry. Grand Rapids, Mich.: Eerdmans, 1975. Crow, Loren D. The Songs of Ascents (Psalms 120-134): Their Place in Israelite History and Religion. SBL-DS, 148. Atlanta, Ga.: Scholars Press, 1996. Crystal, David. A Dictionary of Linguistics and Phonetics. 4th ed. Oxford: Blackwell, 1980. Culley, Robert C. Review of (Auffret 1995) in CBQ 59 (1997): 105-6.

BIBLIOGRAPHY

355

Dahood, Mitchell. “Ugaritic-Hebrew Philology: Marginal Notes on Recent Publications.” Biblica et Orientalia, 17. Rome: PIB, 1965a. _____. Psalms I : 1-50. ABC, 16. Garden City, N.Y.: Doubleday, 1965b. _____. Psalms II : 51-100. ABC, 17. Garden City, N.Y.: Doubleday, 1968. _____. Psalms III : 101-150. ABC, 17A. Garden City, N.Y.: Doubleday, 1970. _____. “Vocative lamedh in Psalm 74,14.” Biblica 59 (1978): 262-3. _____. “Interrogative kî in Psalm 90,11; Isaiah 36,19 and Hosea 13,9.” Biblica 60 (1979): 573-4. _____. “Vocative lamedh in Exodus 2,14 and Merismus in 34,21.” Biblica 62 (1981): 413-5. Dawson, David Allan. Text-Linguistics and Biblical Hebrew. JSOTS, 177. Sheffield: Academic Press, 1994. Delitzsch, F. Biblical Commentary on the Psalms I-III. Translation by Francis Bolton. 2nd ed. London: Hodder & Stoughton, 1888. German original, 1867. Di Lella, Alexander A. Review of NJB in CBQ 51 (1989): 358-9. Dik, Simon C. Functional Grammar. NHLS. Amsterdam: North-Holland, 1978. _____. Functional Grammar. 3rd rev. edition. New York, NY: Academic Press, 1981. _____. The Theory of Functional Grammar, Part 1: The Structure of the Clause. Dordrecht-Holland/Providence, R.I.: Foris, 1989. Dotan, Aron. “Masorah.” In Encyclopaedia Judaica, vol. 16, 1401-82. Jerusalem: Keter, 1971. ____ (ed.). rva !b hvm!b !rha lv hrwsmhw ~ym[jh dwqynh yp l[ bjyh ~yqywdm ~ybwtkw ~yaybn hrwt drgnynl dy btkb. Tel-Aviv: University Press, 1973. ____ (ed.). Biblia Hebraica Leningradensia: Prepared According to the Vocalisation, Accents, and Masora of Aaron ben Asher in the Leningrad Codex. Peabody, Mass.: Hendrickson, 2001. Dresher, Bezalel E. “The Prosodic Basis of the Tiberian Hebrew System of Accents.” Language 70 (1990): 1-52. Driver, G.R. Problems of the Hebrew Verbal System. OTS, 2. Edinburgh: T. & T. Clark, 1936. _____. Canaanite Myths and Legends. OTS, 3. Einburgh: T. & T. Clark, 1956. Driver, S.R. A Treatise on the Use of the Tenses in Hebrew: and some other Syntactical Questions. 3rd ed. Oxford: Clarendon Press, 1892. Drysdale, A.H. Early Bible Songs: With Introduction on the Nature and Spirit of Hebrew Song. London: The Religious Tract Society, 1890.

356

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Ducrot, Oswald and Tzvetan Todorov. Encylopedic Dictionary of the Sciences of Language. Baltimore, Md.: Johns Hopkins University, 1979. Dussaud, R. Les découverts de Ras Shamra (Ugarit) et l’Ancien Testament. 2nd ed. Paris: Paul Geuthner, 1937. Eaton, J. H. “Music’s Place in Worship: A Contribution From the Psalms.” In Prophets, worship and Theodicy: Studies in Prophetism, Biblical Theology and Structural and Rhetorical Analysis and on the Place of Music in Worship. Papers Read at the Joint British-Dutch Old Testament Conference Held at Woudschoten, 1982, 85-107. OTS, 23. Leiden: Brill, 1984. Emerton, J. A. Review of (Craigie 1983) in VT 34 (1984): 248-9. Eisenberg, Azriel. 1976. The Book of Books: The Story of the Bible Text. London: Soncino. Eissfeldt, Otto. The Old Testament: An Introduction including Apocrypha and Pseudepigrapha, and Also the Works of Similar Type from Qumran. The History of the Formation of the Old Testament. Translated by Peter R. Ackroyd. 3rd ed. Oxford: Basil Blackwell, 1965. German original, 1934. Elliger, K., and W. Rudolph, eds. Biblia Hebraica Stuttgartensia. Stuttgart: Württembergische Bibelanstalt, 1969. Esh, Saul. tyrb[b rvy rwbyd ynpl hxyt-twlm l[. Leshonenu 22 (1958): 48-53. Eskhult, Mats. Studies in Verbal Aspect and Narrative Technique in Biblical Hebrew Prose. Studia Semitica Upsaliensia. Stockholm: Uppsala, Almqvist & Wiksell International Acta Universitatis Upsaliensis, 1990. Even-Shoshan, Abraham. A New Concordance of the Bible. 3 vols. Jerusalem: Kiryat Sepher, 1982. Ewald, G. Heinrich. Commentary on the Psalms. 3rd ed. 2 vols. Trans. into English by E. Johnson. London: Williams and Norgate, 1880. German original, 1866. Fariss, Sherry L. Word Order in Biblical Hebrew Poetry. Ph.D. diss. The University of Texas at Arlington, 2003. Fassberg, Steven E. “Lengthened Imperative hl'j.q' in Biblical Hebrew.” HS 40 (1999): 7-14. Fensham, F.C. “The Use of the Suffix Conjugation and the Prefix Conjugation in a Few Old Hebrew Poems.” JNSL 6 (1978): 9-18. Fenton, Terry L. The Ugaritic Verbal System. Ph.D. diss. Oxford: Oxford University Press, 1963. _____. Passages in Ugaritic Discourse – Restoration and Observations. UF 1 (1969a): 199-200. _____. Command and Fulfillment in Ugaritic - ‘TQTL : YQTL’ and ‘QTL : QTL’. JSS 14 (1969b): 34-38.

BIBLIOGRAPHY

357

_____. “The Hebrew ‘Tenses’ in the Light of Ugaritic.” In Proceedings of the Fifth World Congress of Jewish Studies, vol. 4, 31-9. Jerusalem: WUJS, 1973. Fishoff, E. “Piyyut.” In Encyclopaedia Judaica, vol. 13, 574-602. Jerusalem: Keter, 1971. Floor, Sebastiaan Jonathan. From Information Structure, Topic, Focus, to Theme in Biblical Hebrew Narrative. D.Lit. dissert. Stellenbosch: University, 2004. Retrieved from http://academic.sun.ac.za/as/cbta/Floor.pdf., at 24/05/2006. Follis, Elaine R., ed. Directions in Biblical Hebrew Poetry. JSOTS, 40. Sheffield: Academic Press, 1987. Fokkelman, J. P. Narrative Art and Poetry in the Books of Samuel: A Full Interpretation Based on Stylistic and Structural Analyses. Vol. 1: King David (2 Sam. 9-20 & 1 Kings 1-2). SSN, 20. Assen: Van Gorcum, 1981. _____. Narrative Art and Poetry in the Books of Samuel: A Full Interpretation Based on Stylistic and Structural Analyses. Vol. 2: The Crossing Fates (1 Sam. 13-31 & 2 Sam. 1). SSN, 23. Assen: Van Gorcum, 1986. _____. The Structure of Psalm LXVIII. Pp. 72-83 in In Quest of the Past: Studies on Israelite Religion, Literature, and Prophetism. Papers Read at the Joint British-Dutch Old Testament Conference, Held at Elspeet, 1988. Ed. by A.S. Van der Woude. OTS, 26. Leiden: Brill, 1990. _____. Narrative Art and Poetry in the Books of Samuel: A Full Interpretation Based on Stylistic and Structural Analysis. Vol. 4: Vow and Desire (1 Sam. 112). SSN, 31. Assen: Van Gorcum, 1993. _____. “The Song of Deborah and Barak: Its Prosodic Levels and Structure.” In Pomegranates and Golden Bells: Studies in Biblical, Jewish, and Near Eastern Ritual, Law, and Literature in Honor of Jacob Milgrom, edited by David P. Wright, David Noel Freedman, and Avi Huritz, 595-628. Winona Lake, Ind.: Eisenbrauns, 1995. _____. Major Poems of the Hebrew Bible at the Interface of Hermeneutics and Structural Analysis. Volume I: Ex. 15, Deut. 32, and Job 3. SSN, 37. Assen: Van Gorcum, 1998. _____. Major Poems of the Hebrew Bible at the Interface of Prosody and Structural Analysis. Vol. II: 85 Psalms and Job 4-14. SSN, 41. Assen: Van Gorcum, 2000. _____. Reading Biblical Poetry: An Introductory Guide. Trans. by Ineke Smith. Louisville, Ky: Westminster John Knox, 2001. _____. The Psalms in Form: The Hebrew Psalter in its Poetic Shape. Tools for Biblical Study Series, 4. Leiden: Deo, 2002.

358

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

_____. Major Poems of the Hebrew Bible at the Interface of Prosody and Structural Analysis. Vol. III: The Remaining 65 Psalms. SSN, 43. Assen: Van Gorcum, 2003. Freeborn, Dennis. Style: Text Analysis and Linguistic Criticism. Houndmills: Macmillan, 1996. Freedman, David Noel. “Acrostics and Metre in Hebrew Poetry.” In Pottery, Poetry and Prophecy: Studies in Early Hebrew Poetry, 51-76. Winona Lake, Ind.: Eisenbrauns, 1980. Originally published in 1972. _____. “Another Look at Biblical Hebrew Poetry.” In Directions in Biblical Hebrew Poetry, edited by Elaine R. Follis, 11-28. JSOTS, 40. Sheffield: Academic Press, 1987. Freedman, David Noel (ed.). The Leningrad Codex: A Facsmile Edition. Grand Rapids, Mich.: Eerdmans. Leiden: Brill, 1998. Freedman, D.N. and D.M. Cross. Early Hebrew Orthography: A Study of the Epigraphic Evidence. American Oriental Series 36. New Haven, Conn.: American Oriental Society, 1952. Freedman, D. N. and Jeffrey C. Geoghegan. “Quantitative Measurement in Biblical Hebrew Poetry.” In Ki Baruch Hu: Ancient Near Eastern, Biblical, and Judaic Studies in Honor of Baruch A. Levine, edited by Robert Chazan, W.W. Wallo, and L.H. Shiffman, 229-49. Winona Lake, Ind.: Eisenbrauns, 1999. Friedman, R. E. “The Prophet and the Historian: The Acquisition of Historical Information from Literary Sources.” In The Poet and the Historian: Essays in Literary and Historical Biblical Criticism, edited by R.E. Friedman, 1-12. HSS, 26. Chico, CA.: Scholars Press, 1983. Garr, W. Randall. “The Linguistic Study of Morphology.” In Linguistics and Biblical Hebrew, edited by Walter R. Bodine, 49-64. Winona Lake, Ind.: Eisenbrauns, 1992. _____. !he Revue Biblique 111 (2004): 321-44. Gaster, T. H. Myth, Legend and Cult in the Old Testament. New York, N.Y.: Harper & Row, 1969. Geller, Stephen A. Parallelism in Early Biblical Poetry. HSM, 20. Missoula, Mo.: Scholars Press, 1979. _____. “Theory and Method in the Study of Biblical Poetry.” JQR 73 (1982): 65-77. _____. “Cleft Sentences with Pleonastic Pronoun: A Syntactic Construction of Biblical Hebrew and Some of Its Literary Uses.” JANES 20 (1999): 16-33.

BIBLIOGRAPHY

359

Gerstenberger, Erhard S. Psalms (part 1) with an Introduction to Cultic Poetry. FOTL, 14. Grand Rapids, Mich.: Eerdmans, 1988. Gevaryahu, H.M.I. “Biblical Colophons: A Source for the ‘Bibliography’ of Authors, Texts, and Books.” In Congress volume: Edinburgh, 1974, edited by IOSOT, 42-59. SVTS, 28. Leiden: Brill, 1975. Gevirtz, Stanley. “Evidence of Conjugational Variation in the Parallelization of Selfsame Verbs in the Amarna Letters.” JNES 32 (1973a): 99-104. _____. “On Canaanite Rhetoric: The Evidence of the Amarna Letters from Tyre.” Orientalia n.s. 42 (1973b): 162-77. Gianto, Agustinus. “Mood and Modality in Classical Hebrew.” In Past Links: Studies in the Languages and Cultures of the Ancient Near East, edited by Shlomo Isre’el, et. all., 183-98. IOS, 18. Winona Lake, Ind.: Eisenbrauns, 1998. Gibson, J.C.L. Canaanite Myths and Legends. 2nd ed. Edinburgh: T. & T. Clark, 1977. Gillingham, S. E. One Bible, Many Voices: Different Approaches to Biblical Studies. London: SPCK, 1998. Ginsberg, H.L. “Ugaritic Myths, Epics, and Legends.” In Ancient Near Eastern Texts Relating to the Old Testament, edited by James B. Pritchard, 129-58. 3rd ed. Princeton, NJ: University Press, 1969. Ginsburg, David (ed.). Twrh nByaym wKTwbym Ucrm w’rBUh sPry hqDv mDwyqym hyfB Ul Py hmswrh wUl Py DPwsym ravwnym Um jlwPym whGhwT mn KTBy yD UTyqym wTrGwmym yvnym. London: Trinitarian Bible Society, 1894. Ginsburg, Christian D. Introduction to the Massoretico-Critical Edition of the Hebrew Bible. New York, N.Y.: KTAV, 1966. Givón, Talmy. “Verb Complements and Relative Clauses: A Diachronic Case Study in Biblical Hebrew.” Afroasiatic Linguistics. 1/4 (1974): 1-22. _____. “The Drift from VSO to SVO in Biblical Hebrew: The Pragmatics of Tense-Aspect.” In Mechanisms of Syntactic Change, edited by Charles N. Li, 181-254. Austin, Tx.: University of Texas, 1977. _____. Syntax: A Functional-Typological Introduction. Vol. 1. Amsterdam: John Benjamins, 1984. _____. Syntax: A Functional-Typological Introduction. Vol. 2. Amsterdam: John Benjamins, 1990. _____. Functionalism and Grammar. Amsterdam: John Benjamins, 1995. Goetze, Albrecht. The So-Called Intensive of the Semitic Languages. JAOS 62 (1942): 1-8. Goldenberg, Gideon. On Direct Speech and the Hebrew Bible. In Studies in Hebrew and Aramaic Syntax Presented to Professor J. Hoftijzer on the Occasion

360

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

of his Sixty-Fifth Birthday, edited by K. Jongeling, H.L. Murre-van den Berg and L. van Rompay, 79-96. Brill: Leiden, 1991. Goldfajn, Tal. Word Order and Time in Biblical Hebrew Narrative. Oxford: Clarendon, 1998. Goulder, Michael D. The Psalms of the Sons of Korah. JSOTSpl., 20. Sheffield: JSOT Press, 1982 _____. The Prayers of David (Psalms 51-72): Studies in the Psalter, II. JSOTS., 102. Sheffield: Academic Press, 1990. _____. The Psalms of Asaph and the Pentateuch: Studies in the Psalter III. JSOTS, 233. Sheffield: Academic Press, 1996. _____. The Psalms of the Return (Book V, Psalms 107-150): Studies in the Psalter, IV. JSOTS, 258. Sheffield: Academic Press, 1998. Gordon, Cyrus Hertzl. Ugaritic Literature: A Comprehensive Translation of the Poetic and Prose Texts. Scripta Pontificii Instituti Biblici, 98. Rome: PIB, 1949. _____. Ugaritic Manual. Newly Revised Grammar. Texts in Transliteratin. Cuneiform Studies. Paradigms. Glossary. Indices. AnOr, 35. Roma: PIB, 1955. _____. Ugaritic Textbook: Grammar, Texts in Transliteration, Cuneiform Selections, Glossary, Indices. AnOr, 38. Rome: PIB, 1965. Gosling, F.A. The Syntax of Hebrew Poetry: An Examination of the Use of Tense in Poetry with Particular Reference to the Book of Job 3/1-42/6. Doctoral Dissertation, University of St. Andrews, 1992. Goshen-Gottstein, Moshe H. The Authenticity of the Aleppo Codex. Textus 1 (1960): 17-58. Goshen-Gottstein, Moshe H. (ed.). The Aleppo Codex. Jerusalem: Magnes Press, 1976. Gray, George Buchanan. The Forms of Hebrew Poetry. London: Hodder and Stoughton, 1915. Gray, John. The KRT Text in the Literature of Ras Shamra: A Social Myth of Ancient Canaan. 2nd ed. Documenta et Monumenta Orientis Antiqui, 15. Leiden: Brill, 1964. _____. The Legacy of Canaan: The Ras Shamra Texts and Their Relevance to the Old Testament. SVTS, 5. Leiden: Brill, 1957. Greenberg, Joseph H., et al. Universals of Human Language. 4 vols. Stanford, Ca.: University Press, 1978. Greenfield, Jonas C. “The Root šql in Akkadian, Ugaritic, and Aramaic.” UF 11 (1979): 325-7.

BIBLIOGRAPHY

361

Greenstein, Edward L. “Two Variations of Grammatical Parallelism in Canaanite Poetry and Their Psycholinguistic Background.” JANES 6 (1974): 87-105. _____. “How Does Parallelism Mean?” In A Sense of Text: The Art of Language in the Study of Biblical Literature (Papers from a Symposium at the Dropsie College for Hebrew and Cognate Learning, May 11, 1982), 41-70. JQRS. Winona Lake, Ind: Eisenbrauns, 1983. _____. “An Introduction to a Generative Phonology of Biblical Hebrew.” In Linguistics and Biblical Hebrew, edited by Walter R. Bodine, 29-40. Winona Lake, Ind.: Eisenbrauns, 1992. Gross, Walter. Verbform und Funktion wayyiqtol für die Gegenwart? Ein Betrag zur Syntax poetischer althebräischer Texte. Arbetiten zu Text und Sprache im alten Testament 1. St Ottilien: EOS, 1976.. _____. “Zur syntaktischen Struktur des Vorfelds im hebräischen Verbalsatz.” ZA 7.2 (1994): 203-214. _____. Die Satzteilfolge im Verbalsatz alttestamentlicher Prosa: Untersucht an den Büchern Dtn, Ri und 2 Kön. Forschungen zum Alten Testament, 17. Tübingen: J.C.B. Mohr, 1996. _____. Doppelt besetztes Vorfeld: Syntaktische, pragmatische und übersetzungstechnische Studien zum althebräischen Verbalsatz. BZAW, 305. Berlin/New York: de Gruyter, 2001. Gunkel, Hermann. The Psalms: A Form Critical Introduction. Translated by T. Horner. Philadelphia, Pa.: Fortress Press, 1967. German original, 1927. Habermann, Abraham Meir. “Poetry.” In Encyclopaedia Judaica, vol. 13, 670-93. Jerusalem: Keter, 1971. Hadas-Lebel, Mireille. Histoire de la langue hébraïque des origines à l’époque de la Mishna. Collection de la Revue des Etudes juives, 21. Paris-Louvain: E. Peeters, 1995. Haglund, Erik. Historical motifs in the Psalms. CBOTS, 23. Malmö: CWK Gleerup, 1984. Halliday, M. A. K. Intonation and Grammar in British English. JLSP, 48. The Hague: Mouton, 1967. _____. “Language Structure and Language Function.” In New Horizons in Linguistics, edited by John Lyons, 140-65. Harmondsworth: Penguin, 1970. _____. “Modes of Meaning and Modes of Expression: Types of Grammatical Structure and Their Determination by Different Semantic Functions.” In Function and Context in Linguistic Analysis: A

362

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Festschrift for William Haas, edited by D.J. Allerton, E. Carney and D. Holdcroft, 57-79. Cambridge: Cambridge University Press, 1979. _____. “Learning How to Mean.” In Foundations of Language Development: A Multidisciplinary Perspective, edited by Eric Lennenberg and Elizabeth Lennenberg, 239-65. London: Academic Press, 1975. _____. “Meaning and the Construction of Reality in Early Childhood.” In Models of Perceiving and Processing of Information, edited by H.L Pick, Jr., and Elliot Saltzman, 67-96. Hillsdale, NJ: Lawrence Erlbaum Associates, 1978. _____. “The Ontogenesis of Dialogue.” In Proceedings of the Twelfth International Congress of Linguistics, edited by W.U. Dressler, 539-44. Innsbruck: Innsbrucker Beiträge zur Sprackwissenschaft, 1979. _____. Systemic Background. In Systemic Perspectives on Discourse. vol 1: Selected Theoretical Papers from the Ninth International Systemic Workshop, edited by J.D. Benson and W.S. Greaves, 1-15. Advances in Discourse Processes, 15. Ablex Publishing, 1985a. _____. An Introduction to Functional Grammar. [2nd ed. 1994 (Halliday 1994b]. London: Edward Arnold, 1985b. _____. Systemic Theory. In The Encyclopedia of Language and Linguistics, edited by R.E. Asher & J.M.Y. Simpson, vol. 8, 4505-8. Oxford: Pergamon, 1994. _____. On Grammar and Grammatics. In Functional Descriptions: Theory in Practice, edited by Ruqaiya Hasan, Carmel Cloran, and David Butt, 1-38. ASTHLS, 121. Amsterdam: John Benjamins, 1996. Halliday, M. A. K. and Christian M. I. M. Matthiessen. An Introduction to Functional Grammar. 3rd ed. London: Arnold, 2004. Hatav, Galia. The Semantics of Aspect and Modality: Evidence from English and Biblical Hebrew. SLCS, 34. Amsterdam: John Benjamins, 1997. Heimerdinger, Jean-Marc. Topic, Focus and Foreground in Ancient Hebrew Narratives. JSOTS, 225. Sheffield: Academic Press, 1999. Held, Moshe. “The YQTL-QTL (QTL-YQTL) Sequence of Identical Verbs in Biblical Hebrew and in Ugaritic.” In Studies and Essays in Honor of Abraham A. Neuman, President, Dropsie College for Hebrew and Cognate Learning, Philadelphia, edited by Ben-Horin, Meir, Bernard D. Weinryb, and Solomon Zeitlin, 281-90 . Leiden: Brill, 1962. _____. “The Action-Result (Factitive-Passive) Sequence of Identical Verbs in Biblical Hebrew and Ugaritic.” JBL 84 (1965): 272-82. Herner, Sven. Syntax der Zahlwörter im Alten Testament. Berlin/Lund, 1893.

BIBLIOGRAPHY

363

Herzog, Av. “Masoretic Accents (Musical Rendition).” In Encyclopedia Judaica, vol. 11, 1098-111. Jerusalem: Keter, 1972. Holman, J. “A Semiotic Analysis of Psalm CXXXVIII (LXX).” In In Quest of the Past: Studies on Israelite Religion, Literature, and Prophetism. Papers Read at the Joint British-Dutch Old Testament Conference, Held at Elspeet, 1988, edited by A.S. Van der Woude, 84-100. OTS, 26. Leiden: Brill, 1990. Hoop, Raymond de. Genesis 49 in Its Literary and Historical Context. OTS, 29. Leiden: Brill, 1999. _____. “The Colometry of Hebrew Verse and the Masoretic Accents: Evaluation of a Recent Approach (Part I).” JNSL 26/1 (2000a): 47-73. _____. “The Colometry of Hebrew Verse and the Masoretic Accents: Evaluation of a Recent Approach (Part II).” JNSL 26/2 (2000b): 65-100. _____. “Lamentations: The Qinah-Metre Questioned.” In Delimitation Criticism: A New Tool in Biblical Scholarship, edited by M.C.A. Korpel and J.M. Oesch, 80-104. PSWRA, 1. Assen: Van Gorcum, 2000c. Horwitz, W.J. “A Study of Scribal Practices and Prosody in CTA 2:4.” UF 5 (1973): 165-73. _____. “Our Ugaritic Mythological Texts: Copied or Dictated?” UF 9 (1977): 122-9. _____. “The Ugaritic Scribe.” UF 11 (1979): 389-94. Howard, David M., Jr. The Structure of Psalms 93-100. BJS, 5. Winona Lake, Ind.: Eisenbrauns, 1997. Hrushovski, Benjamin. “Prosody, Hebrew.” In Encyclopaedia Judaica., vol. 13, 1195-240. Jerusalem: Keter, 1971. Huehnergard, John. The Akkadian of Ugarit. HSM, 34. Atlanta, Ga: Scholars Press, 1989. Hwang, Shin Ja J., and William R. Merrifield (eds.). Language in Context: Essays for Robert E. Longacre. SILUTAPL, 107. Dallas, Tex.: SIL, 1992. Isaksson, Bo. Studies in the Language of Qoheleth with Special Emphasis on the Verbal System. Acta Universitatis Upsaliensis: Studia Semitica Upsaliensia, 10. Uppsala: Almqvist & Wiksell, 1987. Jakobson, Roman. Language in Literature, edited by Krystyna Pomorska and Stephen Rudy. Cambridge, Mass. and London: Belknap, 1987. Janis, Norman. A Grammar of the Biblical Accents. Ph.D. diss. Cambridge, MA: Harvard University Press, 1987. Jenni, Ernst. Das hebräische Pi’el: Syntaktisch-semasiologische Untersuchung einer Verbalform im Alten Testament. Zürich: EVZ, 1968.

364

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

_____. “Zur Funktion der Reflexiv-passiven Stamformen in BiblischeHebräischen.” In Proceedings of the 5th World Congress of Jewish Studies, vol 4 (1973), 61-70. Jerusalem, 1969. _____. Die hebräische Präpositionen, Band 1: Die Präposition Beth. Stuttgart: W. Kohlhammer, 1992. _____. Die hebräische Präpositionen, Band 2: Die Präposition Kaph. Stuttgart: W. Kohlhammer, 1994. _____. Die hebräische Präpositionen, Band 3: Die Präposition Lamed. Stuttgart: W. Kohlhammer, 2000. Jongeling, B. La particule qr;. In Syntax and Meaning: Studies in Hebrew Syntax and Biblical Exegesis, edited by C.J. Labuschagne et al., 97-107. OTS, 18. Leiden: Brill, 1973. _____. l*k}n dans L’Ancien Testament. In Remembering All the Way...: A Collection of Old Testament Studies Published on the Occasion of the Fortieth Anniversary of the Oudtestamentisch Werkengezelschap in Nederland, edited by B. Albrektson et al., 190-200. OTS, 21. Leiden: Brill, 1981. Joosten, Jan. “The Predicative Participle in Biblical Hebrew.” ZA 2 (1989): 128-159. _____. “The Indicative System of the Biblical Hebrew Verb and its Literary Exploitation.” In Narrative Syntax and the Hebrew Bible, ed. by E. van Wolde, 51-71. Biblical Interpretation Series, 29. Leiden: Brill, 1997. _____. “The Long Form of the Prefix Conjugation Referring to the Past in Biblical Hebrew Prose.” Hebrew Studies 40 (1999): 15-26. _____. “Do the Finite Verbal Forms in Biblical Hebrew Express Aspect?” JANES 29 (2002): 49-70. _____. “The Distinction between Classical and Late Biblical Hebrew as Reflected in Syntax.” Hebrew Studies 46 (2005): 327-39. Joüon, P. A Grammar of Biblical Hebrew. 2 vols. 2nd ed. Subsidia biblica 14/I, II. Translated and revised by T. Muraoka. Rome: PIB, 1996. French original, 1947. Kaddari, M. Z. “A Semantic Approach to Biblical Parallelism.” JJS 24 (1973): 167-75. Kahle, Paul E. The Cairo Geniza. 2nd ed. Oxford: Basil Blackwell, 1959. Kautzsch, E. Gesenius’ Hebrew Grammar. Trans. by A.E. Cowley. Oxford: Clarendon, 1910. German original, 1909. Keel, Othmar. The Symbolism of the Biblical World. Ancient Near Eastern Iconography and the Book of Psalms. Translated by Timothy J. Hallett. New York, N.Y.: Crossroad, 1985. German original, 1972. Kennedy, David B. Review on (Allen 1983) in JETS 27 (1984): 346-7.

BIBLIOGRAPHY

365

Kilpatrick, G. D. Review of NJB in NT 28 (1986): 381-2. Kiparski, P. “The Role of Linguistics in a Theory of Poetry.” In Language as a Human Problem, edited by M. Bloomfield, and E. Haugen, 233-46. New York, N.Y.: Norton, 1974. Kloos, Carola. Yhwh’s Combat with the Sea: A Canaanite Tradition in the Religion of Ancient Israel. Leiden: Brill/Amsterdam: G.A. van Oorschot, 1986. Kogut, Simcha. “On the Meaning and Syntactical Status of hNEhi in Biblical Hebrew.” In Studies in Bible 1986, edited by Sara Japhet, 133-54. Scripta Hierosolymitana, 31. Jerusalem: Magnes Press, 1986. Koopmans, William T. Joshua 24 as Poetic Narrative. JSOTS, 93. Sheffield: Academic Press, 1990. Korpel, Marjo C. A. “Introduction to the Series Pericope.” In Delimitation Criticism: A New Tool in Biblical Scholarship, edited by Marjo C.A. Korpel and Josef M. Oesch, 1-50. PSWRA, 1. Aasen: Van Gorcum, 2000. _____. “Unit Delimitation in Ugaritic Cultic Texts and Some Babylonian and Hebrew Parallels.” In Layout Markers in Biblical Manuscripts and Ugaritic Tablets, edited by M.C.A. Korpel and J.M. Oesch, 141-60. PSWRA, 5. Aasen: Van Gorcum, 2005. Korpel, Marjo C. A. and Johannes C. de Moor. “Fundamentals of Ugaritic and Hebrew Poetry.” In The Structural Analysis of Biblical and Canaanite Poetry, edited by Willem van der Meer and Johannes C. de Moor, 1-61. JSOTS, 74. Sheffield: JSOT Press, 1988. _____. The Structure of Classical Hebrew Poetry: Isaiah 40-55. OTS, 41. Leiden: Brill, 1998. Kosmala, Hans. “Form and Structure in Ancient Hebrew Poetry (A New Approach).” VT 14 (1964): 423-45. _____. “Form and Structure in Ancient Hebrew Poetry (continued).” VT 16 (1966): 152-80. Knudtzon, J.A. Om det saakaldte PERFECTUM OG IMPERFECTUM iHebraisk. Denmark: Kristiania, 1889. Kraus, Hans-Joachim. Psalms 1-59: A Commentary. Trans. from German by Hilton C. Oswald. Augsburg, Minn.: Augsburg, 1988. German original, 1978. _____. Psalms 60-150: A Commentary. Trans. from German by Hilton C. Oswald. Augsburg, Minn.: Augsburg, 1989. German original, 1978. Krašovec, Joze. Antithetic Structure in Biblical Hebrew Poetry. SVTS, 35. Brill: Leiden, 1984. Kroeger, Paul R. Analyzing Grammar: An Introduction. Cambridge: University Press, 2005.

366

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Kugel, James L. The Idea of Biblical Poetry: Parallelism and Its History. New Haven, Conn.: Yale University Press, 1981. _____. “Some Thoughts on Future Research into Biblical Style: Addenda to ‘The Idea of Biblical Poetry.’ ” JSOT 28 (1984): 107-17. _____. Review of (Krašovec 1984) in JBL 105 (1986): 704-5. Kutscher, E.Y. A History of the Hebrew Language, edited by R. Kutscher. Jerusalem: Magnes/Leiden: Brill, 1982. Labuschagne, C. J. “The Particles !he and hNEhi.” In Syntax and Meaning: Studies in Hebrew Syntax and Biblical Exegesis, edited by C.J. Labuschagne, 1-14. OTS, 18. Leiden: Brill, 1973. Lake, Helen and Kirsopp Lake. Codex Sinaiticus Petropolitanus et FridericoAugustanus Lipsiensis. Oxford: Clarendon Press, 1922. Landy, Francis. “Poetics and Parallelism: Some Comments on James Kugel’s ‘The Idea of Biblical Poetry.’ ” JSOT 28 (1984): 61-87. Langa, Tertulian. Psaltirea: text versificat conform originalului ebraic (TOB). Cluj-Napoca: ALC Media, and Paris: CERF, 2000. Launderville, Dale. Review of (Craigie 1983) in CBQ 46 (1984): 745-6. _____. Review of (Allen 1983) in CBQ 47 (1985): 691-2. van Leeuwen, C. “Die Partikel ~ai.” In Syntax and Meaning: Studies in Hebrew Syntax and Biblical Exegesis, edited by C.J. Labuschagne, 15-48. OTS, 18. Leiden: Brill, 1973. Levinson, Stephen C. Pragmatics. Cambridge: Cambridge University Press, 1983. Lode, Lars. “A Discourse Perspective on the Significance of the Massoretic Accents.” In Biblical Hebrew and Discourse Linguistics, edited by Robert D. Bergen, 155-74. Dallas, Tex.: SIL, 1994. Loewe, Raphael. “The Medieval History of the Latin Vulgate.” In The Cambridge History of the Bible, vol. 2: The West from the Fathers to the Reformation, edited by G.W.H. Lampe, 102-54. Cambridge: Cambridge University Press, 1969. Loewenstamm, Samuel E. “The Expanded Colon in Ugaritic and Biblical Verse.” JSS 14 (1969): 176-196. Hebrew original, 1964. Loewinger, D.S. “The Aleppo Codex and the Ben Asher Tradition.” Textus 1 (1960): 59-111. Longacre, Robert E. Joseph: A Story of Divine Providence. A Text Theoretical and Textlinguistic Analysis of Genesis 37 and 39-48, Winona Lake, Ind.: Eisenbrauns, 1989. _____. The Grammar of Discourse. 2nd ed. TLL. New York, N.Y.: Plenum, 1996.

BIBLIOGRAPHY

367

Longman, Tremper III Jr. Literary Approaches to Biblical Interpretation. FCI, 3. Grand Rapids, Mich.: Zondervan/Leicester: Apollos, 1987. Loretz, Oswald. Die Psalmen II: Beitrag der Ugarit-Texte zum Vertändnis von Kolometrie und Textologie der Psalmen – Psalm 90-150. AOAT, 207/2. Kevalaer: Butzon&Bercker, Neukirchen-Vluyn: Neukirchener Verlagsgesselschaft, 1979. _____. Psalmstudien: Kolometrie, Strophik und Theologie ausgewählter Psalmen. BZAW, 309. Berlin: Walter de Gruyter, 2002. Lowth, Robert. De Sacra Poesi Hebraeorum: praelectiones academicae Oxonii habitae. Oxford: Clarendon Press, 1753 Lundbom, Jack R. Review of (Auffret 1982) in CBQ 48 (1986): 99-101. Lugt, Pieter van der. Cantos and Strophes in Biblical Hebrew Poetry with Special Reference to the First Book of Psalter. OTS, 53. Brill: Leiden, 2006. Lunn, Nicholas P. Word Order Variation in Biblical Hebrew Poetry: The Role of Pragmatics and Poetics in the Verbal Clause. Ph.D. diss. London: Brunel University, 2004. Lyons, John. Linguistic Semantics: An Introduction. 2 vols. Cambridge: Cambridge University Press, 1995. Mabie, Frederick J. “The Syntactical and Structural Function of Horizontal Dividing Lines in the Literary and Religious Texts of the Ugaritic Corpus (KTU 1).” UF 36 (2004): 291-311. Mallinson, Graham and Barry J. Blake. Language Typology: Cross-Linguistic Studies in Syntax. NHLS, 46. Amsterdam: North-Holland, 1981. Maori, Y. “The Tradition of Pisqa’ot in Ancient Hebrew mss: The Isaiah Texts and Commentaries from Qum’ran.” Textus. 10 (1982): a-! (Hebrew), 1982. March, Eugene W. Review of (Craigie 1983) in JBL 104 (1985a): 313-5. _____ Review of (Allen 1983) in JBL 105 (1985b): 703-4. Marcotte de Sainte-Marie, Henri. Sancti Hieronymi Psalterium iuxta Hebraeos. CBL, 11. Rome: Abbaye Saint-Jerôme, 1954. Marcus, David. “Studies in Ugaritic Grammar I.” JANES 1/2 (1969): 55-61. _____. “The qal Passive in Ugaritic.” JANES 3 (1970-71): 102-111. Margalit, Baruch. “Studia Ugaritica I: Introduction to Ugaritic Prosody.” UF 7 (1975): 289-313. _____. A Matter of Life and Death: A Study of the Baal-Moth Epic (CTA 4-5-6). AOAT, 206. Kevelaer: Butzon and Bercker, Neukirchener-Vluyn: Neukirchener Verlagsgesselschaft, 1980.

368

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

_____. The Ugaritic Poem of Aqht: Text. Translation. Commentary. BZAW, 182. Berlin: Walter de Gruyter, 1989. Margoliouth, G. Catalogue of the Hebrew and Samaritan MSS in the British Museum. part I. London: British Museum, 1899. Martin, J. R. “Systemic Grammar.” In International Encyclopedia of Linguistics, edited by William Bright, vol. 4, 120-1. New York and Oxford: Oxford University Press, 1992. Martin, J. R., Christian M. I. M. Matthiessen, and Clare Painter. Working with Functional Grammar. London: Arnold, 1997. Martin, J. R. and R. McCormack. Mapping Meaning: Profiling with Integrity in a Post-Modern World, 2001. http://www.solki.jyu.fi/apples/ 010101/Mapping.htm, retrieved at 20/05/2005. Mays, James Luther. Review of (Kraus 1989) in Interpretation 45 (1991): 195. McCann, Clinton J. Review of (Tate 1990) in CBQ 55 (1993): 782-3. McFall, Leslie. The Enigma of the Hebrew Verbal System: Solutions from Ewald to the Present Day. HTIBS. Sheffield: Almond Press, 1982. Meer, Willem van der, and Johannes C. de Moor (eds.). The Structural Analysis of Biblical and Canaanite Poetry. JSOTS, 74. Sheffield: Sheffield Academic Press, 1988. Meier, Samuel A. Speaking of Speaking: Marking Direct Discourse in the Hebrew Bible. SVTS, 46. Leiden: Brill, 1992. _____. Bookreview on Rosenbaum (1997) in RBL 02/15/1999. Meschonnic, Henri. Les États de la poétique. Paris: Presses Universitaires de France, 1985. _____. Critique du rythme: anthropologie historique du langage. Lagrasse: Verdier, 1982. Meyer, Rudolf. Hebräische Grammatik. Band 3. Berlin, 1972. Meyer, Rudolf and Georg Beer. Hebräische Grammatik. Sammlung Göschen Bd. 769/769a. Berlin: Walter de Gruyter, 1960. Michel, Diethelm. Tempora und Satzstellung in den Psalmen. AET, 1. Bonn: H. Bouvier, 1960. Miller, Cynthia L. The Representation of Speech in Biblical Hebrew Narrative: A Linguistic Analysis. HSM, 55. Atlanta, Ga.: Scholars Press, 1996. Miller, Patrick D. Vocative Lamed in the Psalter: A Reconsideration. UF 11 (1979): 617-37. _____. Review of (O’Connor 1980a). JBL 102 (1983): 628-9. _____. Review of (Craigie 1983) and (Allen 1983) in Interpretation 39 (1985): 83-4.

BIBLIOGRAPHY

369

Moor, Johannes C. de. “The Art of Versification in Ugarit and Israel. I: The Rhythmical Structure.” In Studies in the Bible and the Ancient Near East presented to Samuel E. Loewenstamm on His Sevetienth Birthday, edited by Yitschak Avishur and Yoshua Blau, 119-39. Jerusalem: E. Rubinstein, 1978. _____. An Anthology of Religious Texts from Ugarit. Religious Texts Translation Series NISABA, 16. Leiden: Brill, 1987. Moor, Johannes C. de and Wilfred G.E. Watson. “General Introduction.” In Verse in Ancient Near Eastern Prose, edited by De Moor, Johannes C. & Wilfred G.E. Watson, ix-xviii. AOAT, 42. Neukirchener-Vluyn: Neukirchener Verlagsgesselschaft, Butzon & Bercker Kevalaer, 1993. Mulder, M. J. Die Partikel ![;y:. In Syntax and Meaning: Studies in Hebrew Syntax and Biblical Exegesis, edited by C.J. Labuschagne, 49-84. OTS, 18. Leiden: Brill, 1973. Mullen, Theodore E. Jr. The Assembly of the Gods: The Divine Council in Canaanite and Early Hebrew Literature. HSM, 24. Chico, CA: Scholars Press, 1980. Muraoka, T. Emphatic Words and Structures in Biblical Hebrew. Jerusalem: Magnus Press, 1985. _____. Semantics of Ancient Hebrew. ANSS, 6. Louvain: Peeters, 1998. Murphy, Roland E. Review of (Kraus 1960) in CBQ 23 (1961): 76-7. _____. Review of (Kraus 1988) and (Kraus 1989) in CBQ 53 (1991): 474. Narkiss, Bezalel, and Aliza Cohen-Mushlin. The Kennicott Bible. London: Facsimile editions, 1985. Nasuti, Harry P. Tradition History and the Psalms of Asaph. SBLDS, 88. Atlanta, Ga.: Scholars Press, 1988. _____. Defining the Sacred Songs: Genre, Tradition and the Post-Critical Interpretation of the Psalms. JSOTS, 218. Sheffield: Sheffield Academic Press, 1999. Neubaeuer, Adolf, and Arthur Ernest Cowley. Catalogue of the Hebrew Manuscripts in the Bodleian Library. vol 2. Oxford: Clarendon Press,1906. Niccacci, Alviero. Review of Terence Collins ‘Line-Forms in Hebrew Poetry: A Grammatical Approach to the Stylistic Study of the Hebrew Prophets’. Liber Anuus 30 (1980): 450-3. _____. “A Neglected Point of Hebrew Syntax: Yiqtol and Position in the Sentence.” Liber Anuus 37 (1987): 7-19. _____. “Basic Principles of the Biblical Hebrew Verbal System in Prose.” Liber Anuus 38 (1988): 7-16.

370

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

_____. “An Outline of the Biblical Hebrew Verbal System in Prose.” Liber Anuus 39 (1989): 7-26. _____. The Syntax of the Verb in Classical Hebrew Prose. Trans. into English by W.G.E. Watson. JSOTS, 86. Sheffield: Academic Press, 1990. Italian original, 1986. _____. “On the Hebrew Verbal System.” In Biblical Hebrew and Discourse Linguistics, edited by Robert D. Bergen, 117-37. Dallas, TX: SIL, 1994a. _____. “Analysis of Biblical Narrative.” In Biblical Hebrew and Discourse Linguistics, edited by Robert D. Bergen, 175-98. Dallas, TX: SIL, 1994b. _____. “Analysing Biblical Hebrew Poetry.” JSOT 74 (1997): 77-93. _____. “Types and Functions of the Nominal Sentence.” In The Verbless Clause in Biblical Hebrew: Linguistic Approaches, edited by Cynthia L. Miller, 215-48. LSAWS, 1. Winona Lake, Ind.: Eisenbrauns, 1999. Norman, W. M. “Grammatical Parallelism in Quiché Ritual Language.” BLS 6 (1980): 378-99. O’Connor, Michael Patrick. Hebrew Verse Structure. Winona Lake, Ind.: Eisenbrauns, 1980a. _____. A Review of (Collins 1978). CBQ 42 (1980b): 91-2. _____. “‘I only am Escaped to Tell thee’: Native American and Biblical Hebrew Verse.” RIL 3 (1986): 121-32. Oesch, Josef M. Petucha und Setuma: Untersuchungen zu einer überlieferten Gliederung im hebräischen Text des Alten Testaments. OBO, 27. Freiburg: Universitätsverlag, Göttingen: Vandenhoeck & Ruprecht, 1979. Olley, John W. ‘Righteousness’ in the Septuagint of Isaiah: A Contextual Study. SBLSCSS. Missoula, Mo.: Scholars Press, 1979. _____. “‘Hear the Word of YHWH’: The Structure of the Book of Isaiah in 1QISAa.” VT 43 (1983): 19-49. del Olmo Lete, Gregorio. Mitos y Leyendas de Canaan segun la tradicion de Ugarit. Fuentes de la Ciencia Biblica, 1. Madrid: Ediciones Cristiandad, 1981. del Olmo Lete, Gregorio and Joaquín Sanmartín. A Dictionary of the Ugaritic Language in the Alphabetic Tradition. 2 vols. Handbook of Oriental Studies, Section One: The Near and Middle East, 67. Trans. by W.G.E. Watson. Leiden: Brill, 2003. Pardee, Dennis. “The Preposition in Ugaritic.” UF 7 (1975): 329-78. _____. “The Preposition in Ugaritic.” UF 8 (1976): 215-322. _____. “More on the Preposition in Ugaritic.” UF 11 (1979): 685-92.

BIBLIOGRAPHY

371

_____. “Ugaritic and Hebrew Metrics.” In Ugarit in Retrospect: Fifty Years of Ugarit and Ugaritic, edited by G.D. Young, 113-30. Winona Lake, In.: Eisenbrauns, 1980. _____. “Ugaritic.” AfO 28 (1982): 259-272. _____. Ugaritic and Hebrew Poetic Parallelism: A Trial Cut (‘nt I and Proverbs 2). SVTS, 39. Leiden: Brill, 1988. _____. “Structure and Meaning in Hebrew Poetry: The Example of Psalm 23.” Maarav 5-6 (1990): 239-80. _____. “Ugaritic Myths.” In The Context of Scripture, vol 1: Canonical Compositions from the Biblical World, edited by W.W. Hallo and K.L. Younger, 241-83. Leiden: Brill, 1997. _____. “Ugaritic Studies at the End of the 20th Century.” BASOR 320 (2000): 49-86. Parker, Simon B. “Parallelism and Prosody in Ugaritic Narrative Verse.” UF 6 (1974): 283-94. _____. “Aqhat.” In Ugaritic Narrative Poetry, edited by Simon B. Parker, 49-80. SBLWAWS, 9. Missoula, Mo: Scholars Press, 1987. _____. The Pre-Biblical Narrative Tradition: Essays on the Ugaritic Poems Keret and Aqhat. SBLRBS, 24. Atlanta, Ga.: Scholars Press, 1989. _____. “The Beginning of the Reign of God – Psalm 82 as Myth and Liturgy.” RB 102 (1995): 532-59. Patton, J.H. Canaanite Parallels in the Book of Psalms. Baltimore: John Hopkins University Press, 1944. Perdue, Leo G. Review of (Craigie 1983) in JAAR 52 (1984): 378-9. Polzin, Robert. Late Biblical Hebrew: Toward an Historical Typology of Biblical Hebrew Prose. Missoula: Scholars Press, 1976. _____. A review of (Fokkelman 1981 & 1986). RSR 15 (1989): 297-306. Pope, Marvin H. “Ugaritic Enclitic -m.” JCS 5 (1951): 123-8. _____. “Vestiges of Vocative Lamedh in the Bible.” UF 20 (1988): 201-207 Price, James D. The Syntax of Masoretic Accents in the Hebrew Bible. SBEC, 27. Lewinston, Canada: Edwin Mellen, 1990. _____. Concordance of the Hebrew Accents in the Hebrew Bible. 5 volumes. SBEC, 34E. Lewinston: Edwin Mellen, 1996. Qimron, Elisha. The Hebrew of the Dead Sea Scrolls. Atlanta, Ga.: Scholars Press, 1986. Rata, Christian G. The Finite Verb in the Book of Job. Doctoral Dissertation, University of Toronto, 2004. de Regt, Lénart J. “Macrosyntactic Functions of Nominal Clauses Referring to Participants.” In The Verbless Clause in Biblical Hebrew: Linguistic

372

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Approaches, edited by Cynthia L. Miller, 273-96. LSAWS, 1. Winona Lake, Ind.: Eisenbrauns, 1999. Rendsburg, Gary A. Linguistic Evidence for the Northern Origin of Selected Psalms. SBL-MS, 43. Atlanta, Ga: Scholars Press, 1990. Revell, E. J. “The Oldest Evidence for the Hebrew Accent System.” BJRL 54 (1970-71): 214-22. _____. “Pausal Forms in Biblical Hebrew: Their Function, Origin and Significance.” JSS 25.2 (1980): 165-79. _____. “Pausal Forms and the Structure of Biblical Poetry.” VT 31 (1981): 186-99. _____. “Masoretic Accents.” In The Anchor Bible Dictionary, edited by D. N. Freedman, vol. 4, 594-6. New York: Doubleday, 1992. _____. “Thematic Continuity and the Conditioning of Word Order in Verbless Clauses.” The Verbless Clause in Biblical Hebrew: Linguistic Approaches, edited by Cynthia L. Miller, 297-319. LSAWS, 1. Winona Lake, Ind.: Eisenbrauns, 1999. Riffaterre, Michael. “Criteria for Style Analysis.” Word 15 (1959):154-74. _____. Semiotics of Poetry. London: Methuen, 1978. Robinson, Theodore H. The Poetry of the Old Testament. Studies in Theology. London: Duckworth, 1947. Rosenbaum, Michael. Word-Order Variation in Isaiah 40-55: A Functional Perspective. SSN, 36. Van Gorcum, 1997. Rubinstein, A. “A Finite Verb Continued by an Infinitive Absolute in Hebrew.” VT 2 (1952): 362-7. Sáenz-Badillos, Angel. A History of the Hebrew Language. Trans. by John Elwolde. Cambridge: Cambridge University Press, 1993. Spanish original, 1988. Sanders, J. A. The Psalms Scroll of Qumrân Cave 11 (11QPsa). DJD, 4. Oxford: Clarendon, 1965. _____. The Dead Sea Psalms Scroll. Ithaca, NY: Cornell, 1967. Sasson, Jack M. “Literary Criticism, Folklore Scholarship, and Ugaritic Literature.” In Ugarit in Retrospect: Fifty Years of Ugarit and Ugaritic, edited by G.D. Young, 81-98. Winona Lake, Ind.: Eisenbrauns, 1981. Schirmann, J. H. “Yannai.” Encyclopaedia Judaica, vol. 16, 712-4. Jerusalem: Keter, 1971. Schoors, A. “The Particle yKi.” In Remembering All the Way…: A Collection of Old Testament Studies Published on the Occasion of the Fortieth Anniversary of the Oudtestamentisch Werkengezelschap in Nederland, edited by B. Albrektson et al., 240-76. OTS, 21. Leiden: Brill, 1981.

BIBLIOGRAPHY

373

Schökel, Alonso. A Manual of Hebrew Poetics. SubBib, 11. Roma: PIB, 1988. Schneider, Wolfgang. Grammatik des Biblischen Hebräisch. Ein Lehrbuch. München: Claudius, 1994. Segert, Stanislav. “Ugaritic Poetry and Poetics: Some Preliminary Observations.” UF 11 (1979): 729-38. _____. A Basic Grammar of the Ugaritic Language: With Selected Texts and Glossary. Berkeley, CA: University Press, 1985. _____. “Words Spread over Two Lines.” UF 19 (1987): 283-8. Seybold, Klaus. Die Psalmen. Handbuch zum Alten Testament I/15. Tübingen: J.C.B. Mohr (Paul Siebeck), 1996. Sheppard, Gerald T. Review of (Allen 1983) in JAAR 53 (1985): 290. Shimasaki, K. Focus Structure in Biblical Hebrew. Bethesda, MD: CDL (2002). Shulman, Ahouva. “The Particle aN" in Biblical Hebrew Prose.” HS 40 (1999): 57-82. Siebesma, P.A. The Function of the niph’al in Biblical Hebrew in Relationship to Other Passive-Reflexive Verbal Stems and to the pu’al and hoph’al in Particular. SSN. Assen: Van Gorcum, 1991. Siewerska, Anna. Functional Grammar. London and New York: Routledge, 1991. Sinclair, Cameron. “The Valence of the Hebrew Verb.” JANES 20 (1991): 63-81. Sirat, Collette. Hebrew Manuscripts of the Middle Ages. Trans. by Nicholas de Lange. Cambridge: Cambridge University Press, 2002. French original, 1994. Sivan, Daniel. “The Use of QTL and YQTL Forms in the Ugaritic Verbal System.” In Israel Oriental Studies vol. XVIII: Past Links – Studies in the Languages and Cultures on the Ancient Near East, edited by Shlomo Izre’el, Itamar Singer, and Ran Zadok, 89-103. Winona Lake, Ind.: Eisenbrauns, 1998. _____. A Grammar of the Ugaritic Language. Handbuch der Orientalistik. Leiden: Brill, 2001. Hebrew original, 1997. Sivan, Daniel and Avi Yonah. Pivot Words of Expressions in Biblical Hebrew and in Ugaritic Poetry. VT 48 (1998): 399-407. Slager, Donald. “The Use of ‘Behold’ in the Old Testament.” Optat 3.1 (1989): 50-79. Smith, Mark S. “Poetic Structure in KTU 1.3 I 4-21.” UF 22 (1990): 317-9. _____. The Ugaritic Baal Cycle. Vol. I: Introduction with text, translation and commentary of KTU 1.1-1.2. SVTS, 40. Leiden: Brill, 1994.

374

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

_____. “The Baal Cycle.” In Ugaritic Narrative Poetry, edited by Simon B. Parker, 81-180. SBLWAWS, 9. Missoula, Mo.: Scholars Press, 1997. von Soden, Wolfram. Ergänzungsheft zum Grundriss der akkadischen Grammatik. AO, 47. Roma: PIB, 1969. Sperber, Alexander. A Historical Grammar of Biblical Hebrew: A Presentation of Problems with Suggestions to Their Solution. Leiden: Brill, 1966. Sperber, Dan, and Deirdre Wilson. Relevance: Communication and Cognition. 2nd ed. Cambridge, Ms.: Harvard University Press, 1995. Sternberg, Meir. The Poetics of Biblical Narrative: Ideological Literature and the Drama of Reading. ILBS. Bloomington, Ind.: Indiana University Press, 1985. Talstra, Eep. “A Review of T. Collins ‘Line-Forms in Hebrew Poetry: A Grammatical Approach to the Stylistic Study of the Hebrew Prophets.’” Bibliotheca Orientalis 41 (1984): 453-7. _____. “Hebrew Syntax: Clause Types and Clause Hierarchy.” In Studies in Hebrew and Aramaic Syntax Presented to Professor J. Hoftijzer, edited by K. Jongeling, et al., 180-93. Leiden: Brill, 1991. _____. Narrative and Comment. Amsterdam/Kampen: Kok, 1995. _____. “Reading Biblical Hebrew Poetry: Linguistic Structure of Rhetorical Device?” JNSL 25.2 (1999): 101-26. Tate, Marvin E. Psalms 51-100. WBC, 20. Waco, TX: Word Books, 1990. Tatu, Silviu. “The Study of Classical Hebrew Poetry: Terminological issues.” Studia Theologica III.2 (2005): 70-9, http://www.studiatheologica.cnet.ro/numar200502.asp?id=1, retrieved at 10/01/2008. _____. “Graphic Devices Employed by the Editors of Ancient Manuscripts and Mediaeval Incunabula to Mark Verse-lines in Classical Hebrew Poetry.” In Method in Unit Delimitation, edited by M.C.A. Korpel, J.M. Oesch, and S.E. Porter, 92-140. PSWRA, 6. Leiden: Brill, 2007a. _____. “Systemic Functional Grammar and Its Value for the Analysis of Plot in Biblical Hebrew Narratives: A Thematic Investigation of Isaac Narrative Cycle (Gen 25.19-28.9).” In Literary-Linguistic Approaches to Narrative: The Ancient Near East (including Egypt), and Neighbouring Regions, edited by Fredrik Hagen et al. OLA; Leuven: Peeters, forthcoming 2007b. Thacker, T. W. The Relationship of the Semitic and Egyptian Verbal System. Oxford: Clarendon Press, 1954.

BIBLIOGRAPHY

375

Thibault, P.J. Formalism and Functionalism in Linguistic Criticism. Pp. 1284a-1287a in vol. 3 of The Encyclopedia of Language and Linguistics. Ed. by R.E. Asher. Oxford: Pergamon, 1994. Thompson, Sandra A. “Functional Grammar.” In International Encyclopedia of Linguistics, vol. 2, edited by William Bright, 37-39. New York and Oxford: Oxford University Press, 1992. Tov, Emmanuel. Textual Criticism of the Hebrew Bible. 2nd ed. Minneapolis, Minn.: Fortress Press, 2001. _____. “The Background of the Sense Divisions in the Biblical Texts.” In Delimitation Criticism: A New Tool in Biblical Scholarship, edited by Marjo C.A. Korpel and Josef M. Oesch, 312-50. Pericope, 1: Scripture as Written and Read in Antiquity. Aasen: Van Gorcum, 2000. _____. “Scribal Features of Early Witnesses of Greek Scriptures.” In The Old Greek Psalter: Studies in Honour of Albert Pietersma, edited by Robert J.V. Hiebert, Claude E. Cox, and Peter J. Gentry, 125-48. JSOTS, 332. Sheffield: Academic Press, 2001. Tropper, Josef. “Das ugaritische Verbalsystem: Bestandsaufnahme der Formen und statistische Auswertung.” UF 24 (1992): 313-37. _____. Ugaritische Grammatik. AOAT, 273. Münster: Ugarit-Verlag, 2000. Tsevat, Matitiahu. A Study of the Language of the Biblical Psalms. JBL-MS, 9. Philadelphia, Penn.: SBL, 1955. Turner, V. “A Review of ‘Ethnopoetics.’” Boundary 2/6 (1978): 583-90. Ullmann, Stephan. Language and Style. LSS, 1. Oxford: Basil Blackwell, 1964. van der Merwe, Christo. “Explaining fronting in Biblical Hebrew.” JNSL 25.2 (1999a): 173-86. _____. “The Elusive Biblical Hebrew Term : A Perspective in Terms of Syntax, Semantics, and Pragmatics in 1 Samuel.” HS 40 (1999b): 83114. Van Valin, Robert D., Jr., and Randy J. La Polla. Syntax: Structure, Meaning and Function. CTL. Cambridge: Cambridge University Press, 1997. Vantoura, Suzanne Haïk. La Musique de la Bible Révélée (with cassette). 2nd ed. Paris: Dessain et Tolra, 1978. Vaux, Roland de. “Les textes de Ras Shamra et l’Ancien Testament.” Revue Biblique 46 (1937): 526-55. Vogels, Walter A. A review of (Fokkelman 1998). CBQ 61 (1999): 332-3. _____. A review of (Fokkelman 2000). CBQ 63 (2001): 516-7. Waltke, Bruce K. “The New International Version and Its Textual Principals in the Book of Psalms.” JETS 32/1 (1989): 17-26. _____. Superscripts, postcripts, or both. JBL 110 (1991): 583-96.

376

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Waltke, Bruce K., and M. O’Connor. An Introduction to Biblical Hebrew Syntax. Winona Lake, Ind.: Eisenbrauns, 1990. Warnsbrough, Henry. The New Jerusalem Bible. London: Darton, Longman & Todd, and Garden City, NY: Doubleday, 1985. Watson, Wilfred G. E. “Lineation (Stichometry) in Ugaritic Verse.” UF 14 (1982): 311-2. _____. “Review of M. P. O’Connor ‘Hebrew Verse Structure’.” Biblica 64 (1983): 131-4. _____. Classical Hebrew Poetry: A Guide to Its Techniques. JSOTS, 26. Sheffield: JSOT Press, 1984a. _____. A Review of (Kugel 1981) in JSOT 28 (1984b): 89-94. _____. “Internal Parallelism in Ugaritic Verse.” SEL 1 (1984c): 53-67. _____. “Internal Parallelism in Classical Hebrew Verse.” Biblica 66 (1985a): 365-84. _____. “Internal Parallelism in Ugaritic Verse: Further Examples.” UF 17 (1985b): 345-56. _____. “Antithesis in Ugaritic Verse.” UF 18 (1986): 413-9. _____. A review of (Krašovec 1984) in CQR 50 (1988a): 503-5. _____. “Internal (Half-line) Parallelism in Ugaritic Once More.” UF 20 (1988b): 365-74. _____. “Parallelism with Qtl in Ugaritic.” UF 21 (1989a): 435-42. _____. “Internal or Half-line Parallelism Once More.” LA 39 (1989b): 27-36. _____. Internal or Half-Line Parallelism in Classical Hebrew. VT 29 (1989c): 44-66. _____. Number Parallelism in Mesopotamian Texts. Maarav 7 (1991): 241-52. _____. “Half-Line Parallelism as Indicative of Verse in Hebrew Prose.” In Verse in Ancient Near Eastern Prose, edited by Johannes C. de Moor and Wilfred G.E. Watson, 331-44. AOAT, 42. Neukirchener-Vluyn: Neukirchener Verlagsgesselschaft, Butzon & Bercker Kevalaer, 1993. _____. Traditional Techniques in Classical Hebrew Verse. JSOTS, 170. Sheffield: Sheffield Academic Press, 1994. _____. “The ‘Split Couplet’ in Ugaritic Verse.” SEL 14 (1997): 29-42. _____. “Ugaritic Poetry.” In Handbook of Ugaritic Studies, edited by W.G.E. Watson and N. Wyatt, 165-92. Handbook of Oriental Studies. Leiden: Brill, 1999. Watts, James W. Review of (Auffret 1995) in JBL 116 (1997): 541-3.

BIBLIOGRAPHY

377

Waxman, M. A History of Jewish Literature, vol. 1: From the Close of the Canon to the End of the Twelfth Century, New York, N.Y.: Thomas Yoseloff, 1960. Weinberg, Werner. Transliteration and Transcription of Hebrew. HUCA 40-1 (1969-70): 1-32. _____. Toward a World Standard in the Transliteration of Hebrew. Pp. 137-51 in Proceedings of the Fifth World Congress of Jewish Studies, Jerusalem, 1969. Vol. 4. 1973. World Union of Jewish Studies, 1969. _____. On Hebrew Transliteration. Biblica 56 (1975): 150-2. Weitzman, Steven. “The Shifting Syntax of Numerals in Biblical Hebrew: A Reassessment.” JNES 55 (1996): 177-85. _____. Song and Story in Biblical Narrative: The History of a Literary Convention in Ancient Israel. Bloomington and Indianopolis, Ind.: Indiana University Press, 1997. Wellhausen, J. The Book of Psalms: Critical Edition of the Hebrew Text. Trans. by J.D. Prince. Sacred books of the Old Testament. Leipzig: J.C. Hinrichs’sche Buchhandlung/London: David Nutt, 1895. Wendland, Ernst R. Comparative Discourse Analysis and the Translation of Psalm 22 in Chichewa, a Bantu Language of South-Central Africa. SBEC, 32. Lewinston, Canada: Edwin Mellen, 1990. _____. “The Discourse Analysis of Hebrew Poetry: A Procedural Outline.” Discourse Perspective on Hebrew Poetry in the Scriptures, edited by Ernst R. Wendland, 1-27. UBSMS, 7. Reading: UBS, 1994a. _____. “Continuity and Discontinuity in Hebrew Poetic Design: Patterns and Points of Significance in the Structure and Setting of Psalm 30.” In Discourse Perspective on Hebrew Poetry in the Scriptures, edited by Ernst R. Wendland, 28-66. UBSMS, 7. Reading: UBS, 1994b. _____. The Discourse Analysis of Hebrew Prophetic Literature: Determining the Larger Textual Units of Hosea and Joel. Mellen Biblical Press Series, 40. Lampeter: Edwin Mellen, 1995. _____. Analyzing the Psalms. 2nd ed. Dallas, TX: SIL/Winona Lake, Ind.: Eisenbrauns, 2002. Westermann, Claus. The Praise of God in the Psalms. Trans. by Keith R. Crim. London: Epworth, 1965. German original, 1961. Wickes, William. A Treatise on the Accentuation of the Three So-Called Poetical Books of the Old Testament, Psalms, Proverbs, and Job. Oxford: Clarendon Press, 1881. _____. A Treatise on the Accentuation of the Twenty-One So-Called Prose Books of the Old Testament. Oxford: Clarendon Press, 1887.

378

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Widengren, Geo. “Psalm 110 och det sakrala kungadömet i Israel.” Uppsala Universitets Aorskrift. 7.1, 1941. _____. “Konungens vistelse i dödsriket. En studio till Psalm 88.” Svensk Exegetisk Aorsbok. 10 (1945): 66-81. Wilson, Gerald H. “Ugaritic Word Order and Sentence Structure in KRT.” JSS 27 (1982): 17-32. Wittstruck, Thorne. The Book of Psalms: An Annotated Bibliography. Books of the Bible. New York, N.Y.: Garland, 1994. Wordsworth, William. Lyrical Ballads 1800. 2nd ed. London, 1800. Reprinted, 1997. Wright, W. A. The Hexaplar Psalter being the Book of Psalms in Six English Versions. Cambridge: Cambridge University Press, 1911. Würthwein, Ernst. The Text of the Old Testament: An Introduction to the Biblia Hebraica. 2nd rev. ed. Trans. by Erroll R. Rhodes. Grand Rapids, Mich.: Eerdmans, 1985. German original, 1972. Wutz, Franz. Die Psalmen: Textkritisch untersucht. München: Kösel & Pustet, 1925. Wyatt, Nicholas. Religious Texts from Ugarit. 2nd edition. The Biblical Seminar, 53. Sheffield: Academic Press, 2002. Yeivin, Israel 1968. The Aleppo Codex of the Bible: A Study of Its Vocalisation and Accentuation. Jerusalem: Magnes [Hebrew]. 1980. Introduction to the Tiberian Masorah. Trans. and ed. by E.J. Revell. SBLMS, 5. Missoula, MO.: Scholars Press. Young, G.D. Ugaritic Prosody. JNES 9 (1950): 124-33. Zevit, Ziony. Nondistinctive Stress, Syllabic Constraints and Wortmetrik in Ugaritic Poetry. UF 15 (1983): 291-8. Zim, Rivkah. English Metrical Psalms: Poetry as Praise and Prayer 1535-1601. Cambridge: Cambridge University Press, 1987.

APPENDIX 1: FUNCTIONAL STRUCTURE OF HEBREW PSALMS COUPLETS WITH q*f~l//y]qf{l VERBAL SEQUENCE ~yI+Ag Wvåg>r' hM'l'â 02.01A `qyrI)-WGh.y< ~yMiªaul.W÷ 02.01B1 ind: interr: fin: WH-

l*M> Why

goy]m ¨ the nations,

r*g+vW √conspire

CCause:

1

rGv // hGh

Process: sensing

reason

Senser Medium

Wh-/Adj Res-

Fin/ Mo-

Pred

Subject

-idue

-od

Topical Rheme

Theme ind: interr: fin: WH- [ellipsis WH- element] =2

(exposition)

Wl=a%M'm and the peoples Senser

y#hGW √ plot Process: sensing

Medium Adj

Phenomenon Range

Subj

Fin/

Mood Text Top MTheme

1

r'q ¨| vanity?

Pred Residue

Rheme

a merges line B with line A. 379

Complement

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

380

yti_N"xiT. hw"hy>â [m;äv' `xQ")yI ytiîL'piT.( hw"©hy>÷ ind: decl: fin: aff 1

=2

(exposition)

1

vmu // lGj

yhwh T=j]N*t' ¨ Lord my supplication, Senser Phenomenon Medium Range Pred Subj Complement Res-od -idue

v*m~u √hear Process: sensing Fin MoTop Theme

ind: decl: fin: aff

06.10A1 06.10B

yhwh Lord Senser Medium Subject MoTop MTheme

Missing line in a.

Rheme T=p]L*t' my prayer Phenom Range Compl ResRheme

y]Q*j ¨| √accept. Process: sensing Fin/ -od

Pred -idue

APPENDIX 1

1

twli Wtyviyª "÷ ~h,îynEy[e ind: decl: fin: aff 1

ind: decl: fin: aff =2

(clarification)

17.11A 17.11B2

383

sBB // vyT

¨ a~V%r?nW u~TT> s=b*bWnW our steps now [they] √surround us Goal Process: Goal' doing Medium Medium' Compl Adj Fin/ Pred Compl Res- Mood -idue Topical MTheme Rheme

u?n?h#m their eyes Goal Medium Compl ResTopical MTheme

y*v'tW [they] √put Process: doing Fin/ Mood

Pred -idue

l]nfot B*a*r#x ¨| to stretch to the land CCause: CLocation: purpose place Adj

Adjunct

Rheme

Following some Hebrew Mss (ynwrva), Ö prefers a 1st person direct object. ã follows Symmachus: ‘those who consider me fortunate’. Many Hebrew Mss, Targums, ä, K, Ö, ã, and Hieronymus prefer a 1st person direct object attached to the verb too. 2 Dahood (1965b, 92) splits this line into two, producing a tercet instead of the couplet (3+2+2). 1

384

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

twaow>) 18.41A `2~te(ymic.a; ya;ªn>f;m.W÷ 18.41B

ind: decl: fin: aff 1

ind: decl: fin: aff +2

(additive: positive)

385

nTn // xmT

¨ w=a)y=b~y n*t~T> L' u)r#p and my enemies √give me the back of the neck Actor Process: Client Scope doing Medium Beneficiary Range Subject Fin/ Pred Compl Complement Mood Residue Topical MTheme Rheme

Wm=c~n]a~y but my enemies Goal Medium Compl ResTopical MTheme

a~x]m't}m ¨| [I] √exterminate. Process: doing Fin/ Mood

Pred -idue

Rheme

ã follows the å prefering ‘breaks in front of me’. Few Mss prefer to move the conjunction in front of the finite verb from its place in front of the participle. Ö has a 2nd sg pronoun as direct object instead of the currently 3rd masc pl. 1 2

386

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

aw>v"+-ytem.-~[i yTib.v;y"â-al{ 26.04A1 `aAb)a' al{å ~ymiªl'[]n:÷ ~[iîw> 26.04B ind: decl: fin: neg 1

yvB // Bwa

Oa-y*v~bT' u]m-m=t}-v*w=a ¨ neither [I] √seat with deceitful man Process: behaving CAccompaniment: comitative Fin/ Mood Top Theme

Pred Residue

Adjunct

Rheme

ind: decl: fin: neg +2

(additive: negative)

w=u]m n~u&l*m'm Oa a*boa ¨| and with hypocrites nor [I] √go. CAccompaniment: Process: behaving comitative Adj

Adjunct ResText Topical Marked Theme

1

Fin/ Mood Rheme

A merges line 26.04A with line 26.03B, forming a ‘log’.

Pred -idue

APPENDIX 1

~y[i_rem. lh;äq. ytia nEf'â 26.05A `bve(ae al{å ~y[iªv'r>÷-~[iw> 26.05B ind: decl: fin: aff 1

cna // yvB

c*n}at' [I] √hate Process: sensing Fin/ Mood Top Theme

387

Pred Residue

q=h~l m=r}u'm ¨ the assembly of evildoers, Phenomenon Range Complement

Rheme

ind: decl: fin: neg +2

(addition: positive)

w=u]m - r=v*u'm and with the wicked CAccompaniment: comitative Adj Text MTheme

Adj ResTop

Oa a}v}b ¨| not [I] √seat Process: sensing Fin/ Mood Rheme

Pred -idue

388

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

rAv=ymib. hd'äm.['( ylig>r;â 26.12A `hw")hy> 1%rEïb'a] ~yliªheq.m;B.÷ 26.12B2 ind: decl: fin: aff 1

r~g=l' my foot Bahaver Medium Subject Mood Top MTheme

umD // BrK b=m'vor ¨ on level ground CLocation: place

u*m=d> √stand Process: behaving Fin/

Pred Adjunct Residue

Rheme

ind: decl: fin: aff +2

(additive: positive)

B=m~qh}l'm a&b*r}k in congregations [I] √bless CLocation: Process: place behaving Adjunct ResTopical MTheme

1 2

Ö adds a 2nd sg pronominal suffix. A renders this verse as a ‘log’.

Fin/ Mood Rheme

Pred -idue

yhwh ¨| the Lord Phenomenon Range Complement

APPENDIX 1

éhw"hy>-taeme( yTil.a;äv' Ÿtx;Ûa; vQEïb;ña] Ht'çAa ind: decl: fin: aff 1

ind: decl: fin: aff =2

(exposition)

27.04aA 27.04aB1

a~j~t One thing Phenom Range Compl ResTop MTheme aot*h that Phenom Range Compl ResTop MTheme

389

val // Bqv

v*a~lT' [I] √ask Process: sensing

m}a}t -yhwh ¨ from the Lord, CLocation: place

Fin/ Mood

Adjunct

Pred -idue

Rheme a&b~Q}v ¨| [I] √ seek Process: sensing Fin/ Mood

Pred -idue

Rheme

a, ç, and GHB merge line A with line B into one, which associated with the following two, results in a tercet. 1

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

390

1

ynIWb+z"[] yMiäaiw> ybiäa'-yKi 27.10A `ynIpE)s.a;y:) hw"åhyw:) 27.10B

ind: decl: fin: aff 1

ind: decl: fin: aff +2

(addition: adversative)

1

uzB // asP

K' - a*b' w=a]M' my father and my mother Behaver Medium Adj Subject Mood Text Topical Marked Theme wyhwh but the Lor Behaver Medium Adj Subj Mood Text Top MTheme

å prefers here an y]qf{l.

u&z*bWn' ¨ √forsake me Process: Behaviour behaving Range Fin/ Pred/ Compl Residue Rheme

y~a~sp}n' ¨| √receive me Process: Behaviour behaving Range Fin/ Pred Compl Residue Rheme

APPENDIX 1

ywI+l.v;b. yTir>m:åa' ynIa]w:â 30.07A `~l'(A[l. jAMïa,-lB; 30.07B1 ind: decl: fin: aff 1

ind: decl: fin: neg ′

2 (quoted idea)

1

amr // amf

w~a&n' a*m~rT' b=v~lw' ¨ but I √think in my security CLocation: Senser Process: sensing time Medium Adj Subj Fin/ Pred Adjunct Mood Residue Text Top MTheme Rheme B~l-a#Mof ‘not [I] √be shaken Process: sensing

l=uol*m ¨| forever’ CExtent: duration

Fin/ Mood Top Theme

Adjunct

Pred Residue Rheme

a scans this verset as a monoline.

391

392

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Wnn"©AK rb,g ^ål.-yKi( 38.16A `yh'(l{a/ 1yn"ïdoa] 2hn you Behaver Medium Subject Mood Top MTheme

yjl // unh

Voc

Fin/ Mood

Pred -idue

Rheme

t~u&n#h √answer, Process: behaving Fin/

hoj*l=T' ¨ [I] √await Process: behaving

yhwh Lord,

Pred Res

a&d)n*y a$Oh*y ¨| Lord God.

Vocative

Rheme

Many Heb Mss prefer hwhy instead. Some Heb Mss prefer a conjunction in front of the pronoun. ã prefers the verb ‘to hear’. 3 a scans this verset as a monoline. 1 2

APPENDIX 1

éWTv; lAaåv.li( Ÿ!aCoÜK; 49.15aA1 ~[eîr>yIò tw tod'u}n' ¨| wisdom [you] √make know me Phenomen Process: Recipient sensing Range Beneficiary Subject Fin/ Pred Compl Mood -idue Rheme

Ö conjugates the verb as the following one. Missing in ã. a transposes twxjb to the next line. A displays RFSO (1:2). 3 Ö and ã prefer to add a 2nd sg pron suffix and conjugate it as ~tsbw. 4 Dahood (1968, 1) scans this verse as a couplet and a monoline (2+2, 2). 1 2

398

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

WnT'_x.n:z> ~yhiäl{a/ 1hT'äa;-al{)h] 60.12A2 `Wnyte(Aab.ciB. 3~yhiªl{a/÷ aceîte-al{w>) 60.12B4 ind: interr: fin: polar: neg 1

ind: decl: fin: neg =2

h&Oa - a~TT> ¿not you, ProBehaver Medium Adj Subject MoText Top Marked Theme

znj // yxa

a$Oh'm God, Voc

z=n~jT*nW ¨ √reject us -cess: behaving Behaviour Range Fin/ Pred/ Compl -od Residue

Rheme

w+Oa-t}x}a and not [you] √go out, Process: behaving

a$Oh'm B=x]baot?nW ¨| God, in our warfare? CLocation: time

(clarification) Adj Text Theme

Fin/ Mood Top

Pred Res-

Voc

Rheme

Absent in two Heb Mss. A transposes hta-alh to the previous ‘log’. 3 Absent in two Heb Mss, in Symmachus and ã. 4 Kraus (1989, 2, note m) deletes ~yhla metri causa. 1 2

Adjunct -idue

APPENDIX 1

éx;yDIh;l. Wcå[]y" 1ŸAt’aeF.mi %a:Ü 62.05aA2 bz"ïk'ñ 3Wcár>yI 62.05aB ind: decl: fin: aff 1

ind: decl: fin: aff =2

(exposition)

399

yux // rxh

a~k m]C=a}to Certainly, from his exaltation CManner: CLocation: degree place Adjunct Adjunct MoResInterp Topical MTheme

y*u&cW l+h~DD'~j ¨ [they] √devise to impel CCause: Process: sensing purpose Fin/ Pred Adjunct -od -idue Rheme

y]rxW k*z*b ¨| [they] √enjoy lies Process: sensing Phenom Range Fin/ Pred Compl Mood Residue Top Rheme Theme

Ö follows ã reading ‘my honour’, but Hieronymus prefers ‘my share’. A transposes xydhl to the next ‘brick’. Dahood (1968, 89) prefers it too in the next line. Langa (2000, 162) splits this line into two (2+2). 3 Ö follows ã prefering a different verb ‘to run’. They also continue by adding the preposition B in front of the following noun, as Aquilla, Origen, and Symmachus do. 1 2

400

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

y['_Wcy>-l[; ^yTiîr.k;z>-~ai 63.07A `%B"-) hG[, t'yyIåh'-yKi( 63.08A `!NE)r;a] ^yp,än"K. lceÞb.W 63.08B1 ind: decl: fin: aff

K' -

Text Theme ind: decl: fin: aff ×2

(cause: reason^ effect)

hyh // rnn u#zr*t> L' ¨ help to me CCause: Value behalf Range Pred Compl Adjunct Residue

h*y't* [you] √be Process: being

1 Adj

401

Fin/ Mood Top

Rheme

Wb=x}l K+n*p#yk* in the shadow of your wings CLocation: place

a&r~N}n ¨| [I] √sing Process: behaving

Adjunct ResTopical Marked Theme

Fin/ Mood

Pred -idue

Rheme

Dahood (1968, 95) extends the current couplet into a tercet, by splitting this line into two (3+2+1). 1

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

402

1

yNImmi ynIT:ïd>M;li ~yhiªl{a/ 71.17A1 `^yt,(Aal.p.nI dyGIïa; hN"he©÷-d[;w> 71.17B2 ind: decl: fin: aff

a$Oh'm God,

1 Vocative Interp Theme ind: decl: fin: aff 2 (cause: reason^ effect) ×

1 2

lmD // nGD

l]M~dT~n' [you] √teach me Process: doing Goal Medium Fin/ Pred/ Compl Mood Residue Top Rheme

w=u~d-h}N> a~GG'd and till hither [I] √declare CLocation: time Process: saying Adj

407

Adj ResText Top Marked Theme

Fin/ Mood Rheme

A displays RFSO (1:2). a renders this verset as a monoline.

Pred -idue

m]N=uWr*y ¨ from my youth CLocation: time Adjunct

n]pl=aot#yk* ¨| your miracles Verbiage Range Complement

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

408

~yli_l.AhB;( ytiaNEqiâ-yKi( 73.03A `ha,(r>a, ~y[iäv'r> ~Alßv. 73.03B1 ind: decl: fin: aff 1

ind: decl: fin: aff ×2

(effect^ cause: reason)

1

qna // rah

q]N}at' B~jol=l'm ¨ [I] √envy the arrogants Process: sensing Phenomenon Range Adj Fin/ Pred Complement Mood Residue Text Top Rheme Theme K' -

v=lom r=v*u'm a#ra#h¨| the welfare of the evil ones [I] √see Phenomenon Process: sensing Range Complement Fin/ Pred ResMood -idue Topical Rheme Marked Theme

a and ç read this verset as a monoline. A allows hara to join the following

‘brick’.

APPENDIX 1

hw"+a]g: Amt.q:ån"[] !kel'â `Aml'( 1sm'îx' tyvi-©÷ @j'[]y: ind: decl: fin: aff

l*k}n Therefore,

1 Adjunct Text Theme ind: decl: fin: aff +2

(additive: positive)

1 2

73.06A 73.06B2

409

unq // ufP

u&n*q~tmo √serve as necklace them Process: doing Goal Medium Fin/ Pred Compl MoResidue Top Rheme

g~a&w> ¨ pride Actor Agent Subject -od

y~u&f*p - v't j*m*s l*mo ¨| √cover [as] garment violence to them CRole: Process: doing Actor Goal guise Agent Medium Fin/ Pred Compl Subj Adj MoRes-od -idue Top Rheme Theme

The cvasi-totality of versions render it differently. A renders this verse into three ‘bricks’ (2+2+2).

410

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

~h,_yPi ~yIm:åV'b; WTåv; 73.09A `#r,a'(B' %l:ïh]Ti( ~n"©Avl.W÷ 73.09B ind: decl: fin: aff

v~TTW [they] √appoint Process: doing

1 Fin/ Mood Top Theme ind: decl: fin: aff +2

(additive: positive)

Pred Residue

vTh // hlK b~v*m~y]m P'h#m ¨ to heavens their mouth Scope Actor Range Medium Adjunct Complement

Rheme

Wl=von*m and their tongue Actor Medium Adj Subj Mood Text Top MTheme

T]h&l~k B*a*r#x ¨| √travel the earth Process: Scope doing Range Fin/ Pred Adj Residue Rheme

APPENDIX 1

xc;nbo*[]T; 4dWDïmi 5wyP'ªK;÷ 81.07B

swr // uBr

3

ind: decl: fin: aff 1

ind: decl: fin: aff =2

(exemplifica tion)

415

h&s'rot' m]S}b#l v]kmo ¨ [I] √remove from the load his shoulder CLocation: Process: Goal doing place Medium Fin/ Pred Adjunct Compl Mood Residue Top Theme Rheme K~PP*yw m]DDWd his hands from the basket CLocation: Goal Medium place Subject Adjunct MoResTop MTheme Rheme

T~u&b)r=n> ¨| √set free Process: doing Fin/ -od

Pred -idue

Ö prefers a 3 sg verbal form. A allows the previous line to halt ytwrysh for the formation of the ‘log’. 3 Ö reads √db[ instead. 4 Some Heb Mss read rwdm. Ö translates as if the preposition is b. 5 Some Heb Mss and ã have a conjunction in front of the noun. 1 2

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

416

wD'_x.y: bleä Wcå[]An yKiÛ `Wtro)k.yI tyrIåB. 1^yl,ª['÷ ind: decl: fin: aff 1

ind: decl: fin: aff =2

(clarification)

1

83.06A 83.06B

yux // KrT

nou&xW l}b y~jD*w ¨ [they] √plot a heart of unity Process: Behaviour behaving Medium Conj Fin/ Pred Complement Mood Residue Text Top Rheme Theme K'

u*l#yk* B=r't y]kr)tW ¨| against you a covenant [they] √cut down CCause: Behaviour Process: behaving behalf Medium Adjunct Compl Fin/ Pred ResMood -idue Top MTheme Rheme

ã and å have a conjunction in front of the preposition.

APPENDIX 1

yTi[.W:+vi hw"åhy> ^yl,äae ŸynIÜa]w: 88.14A1 `&'mq;t. ytiîL'piT.( rq,Boªb;W÷ 88.14B2 ind: decl: fin: aff 1

ind: decl: fin: aff +2

(addition: positive)

1 2

w~a&n' But I Behaver Medium Conj Subj MoText Top MTheme

417

vwu // qDm

a}l#yk* to you, CLocation: place Adj Res

yhwh v]ww~u=T' ¨ Lord, √cry out Process: behaving Voc Fin/ Pred -od -idue

Rheme

Wb~BB)q#r T=p]L*t' t=q~DD=m#K* ¨| and by morning my prayer √reach you CLocation: time Actor Process: Goal doing Agent Medium Conj Adj Subject Fin/ Pred Compl ResMood -idue Text Top MTheme Rheme

A displays RFSO (2:2). ç scans this verse as a monoline.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

418

~t'_ar'b. hT'äa; 1!ymiy"w>â !Apåc' 89.13A `WnNE)r;y> 2^ïm.viB. !Amªr>x,w>÷ rAbðT' 89.13B3 ind: decl: fin: aff 1

ind: decl: fin: aff =2

(clarification)

Bra // rnn

¨ x*pon w=y*m'n a~TT> b=r*at*m north and south you √create them Goal Actor Process: Goal' Medium Agent doing Medium' Complement Subj Fin/ Pred/ Compl ResMood -idue Topical MTheme Rheme

T*bor w=j#rmon B=v]mk* y=r~N}nW ¨| Tabor and Hermon at your name √rejoice C Behaver Cause: Process: behaving reason Medium Subject Adjunct Fin/ Pred MoRes-od -idue Topical MTheme Rheme

Some Greek Mss read ‘sea’ instead, with (the original Ö) or without article (Ö áG). 2 Two Heb Mss and ã replaced the preposition b with l. 3 A splits this line into two ‘bricks’ (2+2). 1

A L and

APPENDIX 1

yvi_d>q'b. yTi[.B;äv.nI tx;a;â 89.36A `bZE)k;a] dwIïd'l.-~ai( 89.36B1 ind: decl: fin: aff 1

ind: decl: fin: neg =2

(exposition)

1

419

vBu // KzB

a~j~t once CLocation: time

n]vB~uT' [I] √swear Process: behaving

Adjunct ResTop MTheme

Fin/ Mood

a]m surely not CManner: degree Adjunct MoMod MTheme

l=d*w]d to David Behaviour Range Compl ResTop

b+q"dv' ¨ on my holiness CManner: quality Pred -idue

Adjunct

Rheme

a and ç display this verset as a monoline.

a&k~ZZ}b ¨| [I] √lie Process: behaving Fin/ Pred -od -idue Rheme

420

1

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

^l÷-~ai 95.11B2 ind: decl: fin: aff

vBu // Bwa

¨ a&v#r - n]vB~uT' b=a~P' which [I] √swear in my wrath CManner: quality Process: saying

1 Adj Text Theme ind: decl: fin: neg ˝2

(quoted locution)

a]m “surely not CManner: degree Adj Mood Text Theme

1 2

Fin/ Mood Top

Pred Adjunct Residue Rheme

y=b)aWn a#l-m=nWj*t' ¨| [they] √enter in my rest” CLocation: place Process: happening Fin/

Pred Adj Residue

Top Rheme

Ö follows ã in placing a k in front of the relative pronoun. a, A, and ç render this verset as a monoline.

421

422

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

ytianE+f' 1~yjiîse-2hfo)[] 101.03bA3 `yBi( qB;äd>yI al{ß 101.03bB4 ind: decl: fin: aff 1

ind: decl: fin: aff +2

(additive: positive)

u&c#h-s}f'm the deed of the revolter Phenomenon Range Complement ResTopical MTheme

cna // DBq

c*n}at' ¨ [I] √hate Process: sensing Fin/ Pred Mood -idue Rheme

Oa y]dB~q B' ¨| not [he] √cling to me Process: sensing Phenom Range Fin/ Pred Adjunct Mood Residue Top Theme Rheme

Many Heb Mss renders this as ~yjf, as Ö does. Ö prefers a participle plural. 3 After transposing the last verse-unit to the next line, Dahood (1970, 1) admits the line along the previous two, thus expanding the current couplet into a tercet. 4 ç renders this verse as a couplet, merging the current lines by twos. A admits it too, but only after transposing hf[ to the previous ‘brick’. 1 2

APPENDIX 1

h'yn WWæqi yliÛ `1!n")ABt.a, ^yt,dª o[e÷ ind: decl: fin: aff 1

ind: decl: fin: aff +2

(adversative)

1 2

119.95A 119.95B2

433

qwh // Byn

l' q]wwW l=a~BB=d}n' ¨ for me √wait Phenom Process: sensing Range Adjunct Fin/ Pred MoResidue Top MThe Rheme me

r=v*u'm wicked men to destroy me CCause: Senser purpose Medium Subject Adjunct -od

u}d)t#yk* a#tBon*n ¨| your testimonies [I] √consider carefully Phenome Process: sensing non Range Compl Fin/ Pred ResMood -idue Top MTheme Rheme

ã and Hieronymus have a singular noun instead. ç displays this verse as a monoline.

434

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

yl'_g>r; ytialiäK' [r'â xr;aoå-lK'mi 119.101A `1^r q*n't* k]ly)t*y ¨ you √acquire my kidneys Actor Process: Goal doing Agent Medium Adj Subj Fin/ Pred Compl Mood Residue Text Top MTheme Rheme T=s%K}n' [you] √knit me Process: doing Goal Medium Fin/ Pred Compl Mood Residue Top Rheme Theme

Ö and ã have the preposition !m instead of b.

B=b#f#n a]M' ¨| in the womb of my mother CLocation: place Adjunct

438

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

ble_B. tA[år' Wbåv.x' rv-lK' ynIWpD>r>yIâ ds,x,äw" 1bAjÜ Ÿ%a:Ü 23.06A `~ymi(y" %r,aoål. hw"©hy>÷-tybeB. 2yTiîb.v;w> 23.06B3 ind: decl: fin: aff 1

ind: decl: fin: aff

a~k fob w*j#s#d Surely goodness and mercy CManner: Actor degree Agent Adj Subject ResMood Interp Topical MTheme

=2

w=v~bT' and [I] √return Process: happening

(exemplification)

Conj Text Theme

Fin/ Mood Top

rDP // yvB

y]rD=pWn' √follow me Process: Goal doing Medium Fin/ Pred Compl -idue

... ¨ ...

Rheme

B=b?t-yhwh l=a)r#k y*m'm ¨| in Lord’s house for the length of days CLocation: CExtant: duration place

Pred Adjunct Residue

Adjunct

Rheme

1 Ö does not have the first two words of this verse-line. As for the third one, Ö and ã add a 2nd sg pron suffix. 2 Ö renders this verbal form as a noun (‘my dwelling’), and ã reconstructs √bvy. 3 Dahood (1965b, 145), Craigie (1983, 204), NJB, Kraus (1988, 304), Langa (2000, 69), and Fokkelman (2003, 333) scan this verset as a double couplet (3+2, 2+2). On the unorthodox measures the lines of this Psalm display Briggs and Briggs agree too (1906, xliii). They see Psalm 23 as having three strophes of three, four and five tones, increasing with each strophe.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

444

1

!Ws+reh'yE) tAtV'h;â yKiä 11.03A `l['(P'-hm; qyDIªc;÷ 11.03B2

ind: decl: fin: aff β (temporal) ×

ind: interr: fin: WHα

K'

h~V*tot foundations Actor Medium Adj Subject Mood Text Top MTheme x~DD'q the righteous Actor Agent Subj MoTop MTheme

1 2

hrs // Pul y}h*r}sWn ¨ √tear down Process: happening Fin/

Pred Residue

Rheme

m~h what Goal Medium Wh/Compl Res-

P*u*l ¨| √do? Process: doing Fin/ -od

Pred -idue

Rheme

Ö and ã read here two verbs: ‘they have pulled down what you framed’. NJB scans this verset as one line attached to the previous verse.

APPENDIX 2

1

yT'_x.t; 2ydIä[]c; byxiär>T; 18.37A `yL'(sur>q; Wdª[]m'÷ al{ïw> 18.37B

ind: decl: fin: aff

rjB // muD

T~rj'b [you] √enlarge Process: doing

1 Fin/ Mood Top Theme ind: decl: fin: neg =2

(exposition)

445

Pred Residue

x~u&d' my step Actor Medium Compl

t~jT*y ¨ under me CLocation: place Adjunct

Rheme q~rs%L*y ¨| my anckles Actor Medium Pred Subject Res -od

w=Oa m*u&dW and not √slip Process: happening Adj Text Theme

Fin/ MoTop

Rheme

Some Heb Mss correct this to look as in 2 Sam 22.37 yntxt. Ö and ã follow the suggestion of the Origenic Hebrew text rendering this noun as plural. 1 2

446

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

yY"+x; ymeäy>-lK' ynIWpD>r>yIâ ds,x,äw" 1bAjÜ Ÿ%a:Ü 23.06A `~ymi(y" %r,aoål. hw"©hy>÷-tybeB. 2yTiîb.v;w> 23.06B3 ind: decl: fin: aff 1

ind: decl: fin: aff

a~k fob w*j#s#d Surely goodness and mercy CManner: Actor degree Agent Adj Subject ResMood Interp Topical MTheme

=2

w=v~bT' and [I] √return Process: happening

(exemplification)

Conj Text Theme

Fin/ Mood Top

rDP // yvB

y]rD=pWn' √follow me Process: Goal doing Medium Fin/ Pred Compl -idue

... ¨ ...

Rheme

B=b?t-yhwh l=a)r#k y*m'm ¨| in Lord’s house for the length of days CLocation: CExtant: duration place

Pred Adjunct Residue

Adjunct

Rheme

1 Ö does not have the first two words of this verse-line. As for the third one, Ö and ã add a 2nd sg pron suffix. 2 Ö renders this verbal form as a noun (‘my dwelling’), and ã reconstructs √bvy. 3 Dahood (1965b, 145), Craigie (1983, 204), NJB, Kraus (1988, 304), Langa (2000, 69), and Fokkelman (2003, 333) scan this verset as a double couplet (3+2, 2+2). On the unorthodox measures the lines of this Psalm display Briggs and Briggs agree too (1906, xliii). They see Psalm 23 as having three strophes of three, four and five tones, increasing with each strophe.

APPENDIX 2

1

yNIMAyð-~[i yTil.v;ªm.nIw>÷ 28.01cB2 ind: decl: fin: aff (condition: positive)

T#j$v#h m]M#N' ¨ [you] √be silent unto me Process: sensing CCause: behalf

Adj

Fin/ Mood Top

Text Theme ind: decl: fin: aff α

jvh // mvl

P#n if

×β

Pred Adjunct Residue Rheme u]m - yor=d? bor ¨| with the descendents into pit CAccompaniment: comitative

w=n]mv~lT' and [I] √compare Process: sensing Adj Text Theme

447

Fin/ Mood Top

Pred Adjunct Residue Rheme

Absent in few Heb Mss and in á. ç scans this verse as a couplet, merging the current lines by twos. GHB reads this verset as a tercet with lines 28.01cA and B merged as one. 1 2

448

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

yxiîWrñ dyqIáp.a; é^d>y"B. 31.06A `tm,(a/ laeä hw"©hy> ytiîAa ht'y dIÞP' 31.06B1 ind: decl: fin: aff 1

ind: decl: fin: aff +2

(additive: positive)

B=y*d=k* in your hand CLocation: place Adjunct ResTop MTheme

1

a~pq'd [I] √entrust Process: doing Fin/ Mood

Pred -idue

rWj' ¨ my spirit Goal Medium Compl

Rheme

P*d't> [you] √redeem Process: doing Fin/ Mood Top Theme

PqD // PDh

Pred Residue

aot' yhwh a}l a$m#t ¨| me, Lord, God of truth Goal Medium Compl Vocative

Rheme

A prefers to scans this verse as a 5+3 verse.

APPENDIX 2

WnBe_li xm;äf.yI Abâ-yKi 33.21A `Wnx.j'(b' Avåd>q' 1~veÞb. yKiÛ 33.21B ind: decl: fin: aff

K' - bo in him CMatter

1 Adj Interp MTheme ind: decl: fin: aff

K'

=2

Adj Res Top

449

cmj // Bfj y]cm~j l]BB}nW ¨ √rejoice our heart Process: Senser sensing Medium Fin/ Pred Subject Mo-idue -od Rheme

b=v}m q"dvo b*f*jnW ¨| in his holy name [we] √trust CMatter Process: sensing

(exposition) Adj Interp MTheme

1

Adjunct ResTop

Fin/ Mood Rheme

Ö reads this noun with a conjunction in front.

Pred -idue

450

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

#r,a'_-Wvr>yyI) ~ywIïn"[]w: 37.11A `~Al)v' broï-l[; Wg©N>[;t.hiw>÷ 37.11B1 ind: decl: fin: aff 1

ind: decl: fin: aff =2

yrv // unG

w~u&n*w'm y'r=vW - a*r#x ¨ and the poors √inherit land Actor Process: Goal doing Agent Medium Conj Subj Fin/ Pred Compl Mood Residue Text Top MTheme Rheme w=h]tu~N=gW and [they]√take delight Process: sensing

u~l-r)b v*lom ¨| in an abundance of peace CMatter

(exposition) Conj Text Theme

1

Fin/ Mood Top

Pred Adjunct Residue Rheme

A renders this verse as two ‘bricks’ (3:2).

APPENDIX 2

1Wdmo+[]y:

y[iäg>nI dg) 41.04A `Ay*l.x'b. 1T'k.p;îh' AbªK'v.mi÷-lK' 41.04B2 ind: decl: fin: aff 1

ind: decl: fin: aff =2

(exposition)

yhwh ¨

the Lord Actor Agent Subject Mood Topical MTheme

suD // hPK

y]su*d#NW √support him Process: Goal doing Medium Fin/ Pred Compl Residue

u~l -U#r#c D=w*y on his bed of sickness time

CLocation:

Adjunct

Rheme

K"l-m]vK*bo h*p~kT* b=j*l=yo ¨| all his laying around [you] √overturn in his sickness CLocation: Goal Process: doing time Medium Complement Fin/ Pred Adjunct ResMood -idue Topical MTheme Rheme

A minuscule Ms of Ö and ã read here an y]qf{l. A renders this verse as a ‘log’ and a ‘brick’ (5 // 2). Dahood (1965b, 248) scans this verse as a double couplet (2+2, 1+2). 1 2

APPENDIX 2

rAxa'â Wnbeäy viT. 44.11A `2Aml'( Wsv'ä Wnyaeªn>f;m.W÷ 44.11B

1rc"+-yNImi

ind: decl: fin: aff

T=v'b}nW [you] √make turn Process: doing

1 Fin/ Mood Top Theme ind: decl: fin: aff +2

(additive: positive)

1 2

Pred Residue

453

vwB // vsh

us Goal Medium Compl

a*jor m]N'-x*r ¨ back from the enemy CLocation: Scope place Range Compl Adjunct

Rheme

Wm=c~na?nW and our enemies Actor Medium Adj Adjunct Mood Text Top MTheme

v*sW l*mo ¨| √plunder at his will CManner: Process: happening quality Fin/ Pred Adjunct Residue Rheme

Absent in ã. Ö reads it ‘before our foes’. Few Heb Mss, ã and å prefer a 1st pl pron suffix instead.

454

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

lk'_a]m; !acoåK. WnnET.Tiâ 44.12A `Wnt'(yrIzE ~yI©AGb;W÷ 44.12B ind: decl: fin: aff

T]T=n}nW [you] √set Process: doing

1 Fin/ Pred Mood Residue Top Theme Rheme ind: decl: fin: aff =2

Wb~Ggoy]m and among nations CLocation: place

(clarification) Adj Text MTheme

Adjunct ResTop

nTn // zrh

us Goal Medium Compl

K=x)an m~a&k*l ¨ like sheep for eating CManner: comparison Adjunct

z}r?t*nW ¨| [you] √scatter us Process: doing Goal Medium Fin/ Pred Compl Mood -idue Rheme

APPENDIX 2

!Ah+-al{b. ^ïM.[;-rKo*m.Ti 44.13A `1~h,(yreyxim.Bi t'yBiªrI÷-al{ïw> 44.13B ind: decl: fin: aff 1

ind: decl: fin: neg =2

(exposition)

T]mK)r - u~M=k* b=Oa-hon ¨ [you]√sell your people for no gain CManner: Process: doing Goal quality Medium Fin/ Pred Compl Adjunct Mood Residue Top Rheme Theme w=Oa-r]BB't* B]mj'r?h#m ¨| and not [you] √gain from their price CLocation: Process: doing place Adj Text Theme

1

fPl // nxr

Fin/ Mood Top

Pred Adjunct Residue Rheme

Hieronymus prefers here a sg pron suffix.

455

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

456

1Wnte_l' x]n:-ta,

`2hl's,(

Wnl'î-rx;b.yI bheäa'-rv,a] bqOß[]y: !Aa’G> ta,î

ind: decl: fin: aff 1

47.05A 47.05B3

Bjr // ahB

y]bj~r - l*nW a#t-n~j&l*t}nW ¨ [he]√chose for our inheritance CCause: Process: Phenomenon sensing behalf Range Fin/ Pred Adj Complement Mood Residue Top Theme Rheme

ind: decl: fin: aff =2

(exemplification)

a#t G=aon y~u&q)b the glory of Jakob Phenomenon Range Compl ResTop Marked Theme

Ö and ã prefer a 3rd sg pron suffix. Absent in ã. 3 A displays RFSO (3:1). 1 2

[a&v#r - a*h}b] s#l> ¨| [× whom [he] √love] Process: sensing Fin/ Pred Mood -idue Rheme

APPENDIX 2

!Ayðd>Pi 2rq;yEw>â 49.09A3 `~l'(A[l. 4ld;îx'w> 49.09B5

1~v'ªp. n:

ind: decl: fin: aff 1

457

yqr // jDl P]dyon n~pv*m ¨ the redemption of their souls Carrier Medium Pred Subject Residue

w=y}q~r and √be precious Process: being Adj Text Theme

Fin/ Mood Top

Rheme

ind: decl: fin: aff =2

(exposition)

l=uol*m ¨| forever CExtant: duration

w=j*d~l and [he] √cease Process: existing Adj Text Theme

Fin/ Mood Top

Pred Adjunct Residue Rheme

Ö prefers a 3rd sg pron suffix. Following the Heb text of Origen, Ö renders this verb as a noun (‘the value’). ã rather reads a Qal into this. 3 A merges line 49.09A with line 48.08B in order to form a ‘log’. 4 The Origenic Heb text has the y]qf{l. 5 a transposes the content of this line to the next line. Wellhausen (1895) scans this verse as a monoline. Kraus (1988, 479, note h) discards the whole line as a parenthesis, obstructing the fluency of verse 7 to verse 9. 1 2

458

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

lke_aTo wyn"ïp'l.-vae 50.03bA `dao)m. 1hr'î[]f.nI wyb'ªybis.W÷ 50.03bB ind: decl: fin: aff 1

ind: decl: fin: aff +2

(additive: positive)

a}v a fire Actor Medium Subject MoTop MTheme

l=p*n*yw in front of him CLocation: place Adjunct Res-

T)ak}l ¨ √consume Process: happening Fin/ Pred -od -idue

Rheme

Ws=b'b*yw and around him CLocation: place

n]cu&r> m=a)d ¨| [he]√rage [a tempest] very CManner: Process: happening degree

Adj

Fin/ Mood

Text MTheme

1

aKl // vur

Compl ResTop

Many Heb Mss have √r[s.

Rheme

Pred -idue

Adjunct

APPENDIX 2

yBi_r>qiB. lyxiäy" yBiliâ 55.05A tw 89.44B2 ind: decl: fin: aff 1

ind: decl: fin: neg =2

(exposition)

1 2

467

vwB // qwm

a~p T*v'b xWr j~rBo ¨ even [ you] √turn back the edge of his sword Process: doing Goal Medium Adj Fin/ Pred Complement Mood Residue Interp Top Rheme Theme w=Oa h&q?m)to and not [you] √support Process: doing

him Goal Medium Adj Fin/ Pred Compl Mood Residue Text Top Rheme Theme

å reads lajwra instead. ç and A scan this verse as a monoline.

B~M]lj*m> ¨| in battle CLocation: time Adjunct

468

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

1hw"+ hy>

yceä[] W[B.f.yIâ `[j'(n" rv yleäWaG> Wrm.ayOâ 107.02A `rc")-dY:mi ~l'ªa'G>÷ rvmi 119.100A `yTir>c")n" ^yd,äWQpi yKiÞ 119.100B1 ind: decl: fin: aff α

ind: decl: fin: aff β (effect^ cause: reason) ×

1

Bnn // nxr

m]ZZ=q}n'm a#tBon*n ¨ than the elders [I] √understand CManner: Process: sensing degree Adjunct Fin/ Pred ResMood -idue Topical MTheme Rheme K' p]QWd#yk* because your precepts Phenomenon Range Adj Complement ResText Topical MTheme

n*x*r=T' ¨| [I] √obey Process: sensing Fin/ Pred Mood -idue Rheme

a and ç display this verse as a monoline.

473

474

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

!n"+ABt.a, 1^yd,îWQPimi 119.104A `rq,v'( xr;aoì-lK' Ÿytia nEÜf' !Ke-©÷ l[; 119.104B2 ind: decl: fin: aff

Bnn // cna

m]PP]QWd#yK* a#tBon*n ¨ from your precepts [I] √gain understanding CLocation: place Process: sensing

α Adjunct ResTopical MTheme ind: decl: fin: aff β (cause: reason^ effect)

u~l-K}n therefore

×

Adj Text Theme

Fin/ Mood

Pred -idue

Rheme c*n}at' K"l-a)r~j v*q#r¨| [I] √ hate every way of deception Process: sensing Phenomenon Range Fin/ Pred Complement Mood Residue Top Rheme

ã has the preposition b instead. ç and A render this verse as a monoline. a adds a new line: o[ti su. evnomoqe,thsa,j moi. Dahood (1970, 167) produces a new line from !k-l[, interpreted as ‘Most High Honest One’, allowing the rest of the line to produce a distinct monoline. Allen (1983, 131) and Kraus (1989, 406) split this line into two, developing the verse into a tercet (2+2+2). 1 2

APPENDIX 2

~yI+AGb; Wråm.ayO za'â 126.02bA1 `hL,a-e( ~[i tAfï[]l; hw"©hy>÷ lyDIîg>hi 126.02bB2 ind: decl: fin: aff

y)am=rW [they] √say Process: saying

Adj

Fin/ Mood Top

Text Theme ind: decl: fin: aff "2

(quoted locution)

amr // GDl

b~Ggoy]m¨ among nations CLocation: place Pred Adjunct Residue

a*z then

1

475

Rheme

yhwh l~u&cot u]m-a}L#h ¨| Lord to do with these” CCause: behalf Actor Process: happening Fin/ Pred Subj Adj Adj MoRes-od -idue Top Rheme Theme h]gD'l “ √make great

A transposes za to the previous ‘brick’. ç scans this verse as a couplet, merging the current lines by twos. A, Dahood (1970, 217), Allen (1983, 170) and Kraus (1989, 447) split this line into two, developing the verse into a tercet (3+2+2). NJB transposes lydgh to the previous line. 1 2

476

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

tyIb;ª hn ¨ with wickedness CManner: quality Adjunct

Rheme

n*c%a l~V*w=a √take in falsehood CManner: Process: behaving quality Fin/ Pred Adjunct MoResidue Top Rheme Theme

u*r#yk* ¨| your adversaries Behaver Medium Subject -od

Origen suggested ‘rebelled against you’ instead. Alternatively, following many versions, this verb has a conjunction and a plural ending. 3 A transposes afn to the following ‘brick’. 1 2

480

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

%LE+h;a] Wzð-xr;ao)B. 142.04bA `yli( xp;ä Wnàm.j' 142.04bB1 ind: decl: fin: aff ×β

(spatial)

ind: decl: fin: aff α

hlK // fmn

B=a)r~j - ÆzW a&h~L}kØ ¨ on the path Æ× that [I] √walkØ CLocation: Process: place happening Adjunct ResTop MTheme

Subj Fin/ Pred Mood -idue [Predicated Theme] Rheme

f*m=nW [they] √hide Process: doing Fin/ Mood Top Theme

Pred Residue

p~j a trap Goal Medium Compl

l' ¨| for me CCause: behalf Adj

Rheme

a scans this verset as a couplet, merging the current lines by twos. This couplet is inexplicably missing at Allen (1983, 275). ç merges line 142.04bB with line 142.04bA. 1

APPENDIX 3: AUTHORITIES CONSULTED FOR VERSE-LINES DELIMITATION IN P SALMS

Authorities\ Codex Sinaiticus (a) Vulgate ( ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1965b Auffret 1982 Craigie 1983 NJB 1985 Kraus 1988 Fokkelman 2000 Langa 2000 Loretz 2002

1             

2              

3       

4       

5       

     

    

    

PSALMS 6 7                          

8       

9       

10       

11       

12       

    

    

    

     

    





Fokkelman 2003 Auffret 2003

481

482

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Authorities\ Codex Sinaiticus (a) Vulgate ( ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1965b Auffret 1982 Craigie 1983 NJB 1985 Kraus 1988 Fokkelman 2000 Langa 2000 Loretz 2002 Fokkelman 2003 Auffret 2003

13              

14       

15  

16  

17  

PSALMS 18 19 20      

21  

22  

23  

24  

   

   

   

   

   

   

   

   

   

   

    

  

  

  

    

    

    

  

 

    

  



    

 

 







  

APPENDIX 3

Authorities\ Codex Sinaiticus (a) Vulgate ( ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1965b Auffret 1982 Craigie 1983 NJB 1985 Kraus 1988 Fokkelman 2000 Langa 2000 Fokkelman 2003

25             

26       

27       

28       

29       

    

    

    

  

37       

38       

39       

40       

41       

    

    

  

    

    

 

483

PSALMS 30 31                         

32            

33             

34           

44       

45       

46       

 47       

48       

  

  

  

  

  

 

 



 

Auffret 2003

Authorities\ Codex Sinaiticus (a) Vulgate ( ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1965b Auffret 1981 Craigie 1983 NJB 1985 Kraus 1988 Fokkelman 2000 Langa 2000 Fokkelman 2003 Auffret 2003



 

42             

43             

35       

36       

    

    



484

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Authorities\ Codex Sinaiticus (a) Vulgate ( ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1965b Dahood 1968 Auffret 1982 Craigie 1983 NJB 1985 Kraus 1988 Kraus 1989 Tate 1990 Auffret 1993 Fokkelman 2000 Langa 2000 Fokkelman 2003 Auffret 2003

49          

 

50          

 

PSALMS 54 55            

51      

52      

53      

 







 

 

 



   



 

1  

56      

57      

58      

59      

60      













 

 

 

 

 

 



 

   

   

 

 

 

 

   

 

    

1 Although this is the only psalm Fokkelman has treated in his two structural analysis volumes, published three years apart, the latest volume is considered. There are two main differences noted both occurring at the beginning of the poem. Previously Fokkelman scanned versets 2 and 3 as double distichs. Lately he reduced them into tristichs, by merging the first two lines into one.

APPENDIX 3

Authorities\ Codex Sinaiticus (a) Vulgate ( ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1968 Auffret 1982 NJB 1985 Kraus 1989 Tate 1990 Auffret 1993 Auffret 1995 Fokkelman 2000 Langa 2000 Loretz 2002 Fokkelman 2003 Auffret 2003

61       

62       

  

   

63           

485

PSALMS 66 67              

64       

65       

   

   

   



 

  

68       

69       

70       

71       

72       

  

  

  

  

  

 

 

  

  

 



















 



486

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Authorities\ Codex Sinaiticus (a) Vulgate ( ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1968 NJB 1985 Kraus 1989 Tate 1990 Auffret 1993 Auffret 1995 Fokkelman 2000 Langa 2000 Loretz 2002 Fokkelman 2003

Authorities\ Codex Sinaiticus (a) Vulgate ( ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1968 Loretz 1979 Auffret 1982 NJB 1985 Kraus 1989 Tate 1990 Auffret 1993 Auffret 1995 Howard 1997 Fokkelman 2000 Langa 2000 Loretz 2002 Fokkelman 2003

73          

74          

75          

76          

77          

PSALMS 78 79                      

80           

81          

82          

 

 

 

 

 

 

 



85       

86       

87       

88       

89       

90        

91        

 92        

   93        

   94        

    

  

  

  

  

  

   

   

   

83          

84          

 

 

95        

96        

  

  

  

  













  

 

 

  

 

 













APPENDIX 3

Authorities\ Codex Sinaiticus (a) Vulgate ( ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1968 Dahood 1970 Loretz 1979 Auffret 1981 Auffret 1982 Allen 1983 NJB 1985 Kraus 1989 Tate 1990 Auffret 1993 Auffret 1995 Howard 1997 Fokkelman 2000 Langa 2000 Fokkelman 2003 Auffret 2003

97       

98       

99       

100       









  

  

  

  

   

 

   

  

 

101      

487

PSALMS 102 103            

104      

105      

106      

107      

 

 

 

  

 

 

 

   

  

  

  

  

  

  

  

 

  

 

  

  

 

488

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Authorities\ Codex Sinaiticus (a) Vulgate (ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1970 Loretz 1979 Allen 1983 NJB 1985 Kraus 1988 Kraus 1989 Auffret 1993 Auffret 1995 Auffret 1999 Fokkelman 2000 Langa 2000 Loretz 2002 Fokkelman 2003 Auffret 2003

108          

109          

110          

111          

PSALMS 112 113                    

114          

115          

116          

117          

118          

 





















  

 

 































 

APPENDIX 3

Authorities\ Codex Sinaiticus (a) Vulgate (ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1970 Loretz 1979 Allen 1983 NJB 1985 Kraus 1989 Auffret 1993 Auffret 1995 Auffret 1999 Fokkelman 2000 Langa 2000 Loretz 2002 Fokkelman 2003

Authorities\ Codex Sinaiticus (a) Vulgate ( ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1970 Loretz 1979 Auffret 1982 Allen 1983 NJB 1985 Kraus 1989 Auffret 1995 Auffret 1999 Fokkelman 2000 Langa 2000 Loretz 2002 Fokkelman 2003

119            

120           

121              

122           

123           



  

  



  

 130        

131        

 132        

133        

134        

  

  

  

  

  

  

























489

PSALMS 124 125                      

126           

127           

128           

129           























 135        

 136        

 137        

   138        

 140        

   

  

  

  

 139            



  



 

 





     

490

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Authorities\ Codex Sinaiticus (a) Vulgate ( ç) Codex Aleppo (A) GHB 1894 Wellhausen 1895 Wutz 1925 Dahood 1970 Loretz 1979 Auffret 1981 Auffret 1982 Allen 1983 NJB 1985 Kraus 1988 Kraus 1989 Auffret 1995 Fokkelman 2000 Langa 2000 Loretz 2002 Fokkelman 2003 Auffret 2003

PSALMS 145 146                

141        

142        

143        

144        

 

 

 

 

 

 

   

   

 

   

  

  

147         

148        

149        

150        

 

  

 

 







 

 

 

 

   

  

APPENDIX 4: RELATIONS AMONG VERB FORMS, SUBJECT AND MARKED THEME

Explicit Subject

Subject

Compl.

Adjunct

Vocative

2.1A 2.1B 6.10A 6.10B 7.14A 7.14B 13.6aA 13.6aB 17.11A 17.11B 18.5A 18.5B 18.41A 18.41B 26.4A 26.4B 26.5A 26.5B

YIQTOL

Reference

Marked Theme

        

         -

     -

    -

  

-

491

492

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Explicit Subject

Subject

Compl.

Adjunct

Vocative

26.12A 26.12B 27.4aA 27.4aB 27.10A 27.10B 30.7A 30.7B 37.23A 37.23B 38.5A 38.5B 38.16A 38.16B 49.15aA 49.15aB 50.19A 50.19B 51.8A 51.8B 60.12A 60.12B 62.5aA 62.5aB 63.7A 63.7B 63.8A 63.8B

YIQTOL

Reference

Marked Theme

             

          -

         -

 -

        

-

APPENDIX 4

493

65.4A 65.4B 65.12A 65.12B 66.18A 66.18B 67.7A 67.7B 68.11A 68.11B 71.17A 71.17B 73.3A 73.3B 73.6A 73.6B 73.9A 73.9B 74.1bA 74.1bB 74.14A 74.14B 78.20bA 78.20bB 78.64A 78.64B 81.7A 81.7B

             

              

        

   -

 -

Vocative

Adjunct

Compl.

Subject

Explicit Subject

YIQTOL

Reference

Marked Theme

 -

494

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

83.6A 83.6B 88.14A 88.14B 89.13A 89.13B 89.36A 89.36B 92.5A 92.5B 95.11A 95.11B 101.3bA 101.3bB 102.15A 102.15B 104.5A 104.5B 104.6A 104.6B 108.12A 108.12B 111.5A 111.5B 116.2A 116.2B 118.10A 118.10B

             

       -

   -

     -

       -

Vocative

Adjunct

Complement

Subject

Explicit Subject

YIQTOL

Reference

Marked Theme

 -

APPENDIX 4

495

119.11A 119.11B 119.69A 119.69B 119.82A 119.82B 119.95A 119.95B 119.101A 119.101B 119.121A 119.121B 132.11aA 132.11bB 139.13A 139.13B 140.3A 140.3B

        

      -

  -

 -

   

Vocative

Adjunct

Compl.

Subject

Explicit Subject

YIQTOL

Reference

Marked Theme

-

496

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

3.7A 3.7B 5.6A 5.6B 8.7A 8.7B 9.8A 9.8B 10.14bA 10.14bB 11.3A 11.3B 18.37A 18.37B 23.6A 23.6B 28.1cA 28.1cB 31.6A 31.6B 33.21A 33.21B 37.11A 37.11B 38.12A 38.12B 41.4A 41.4B

              -

              -

        -

  

    -

Vocative

Adjunct

Compl.

Subject

Explicit Subject

YIQTOL

Reference

Marked Theme

-

APPENDIX 4

497

44.11A 44.11B 44.12A 44.12B 44.13A 44.13B 47.5A 47.5B 49.9A 49.9B 50.3bA 50.3bB 55.5A 55.5B 58.9A 58.9B 60.11A 60.11B 73.18A 73.18B 73.27A 73.27B 83.3A 83.3B 85.12A 85.12B 87.6A 87.6B

              -

             

          

  -

    -

Vocative

Adjunct

Compl.

Subject

Explicit Subject

YIQTOL

Reference

Marked Theme

-

498

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

89.44A 89.44B 104.16A 104.16B 104.26A 104.26B 107.2A 107.2B 108.11A 108.11B 119.93A 119.93B 119.100A 119.100B 119.104A 119.104B 126.2bA 126.2bB 127.1bA 127.1bB 127.1cA 127.1cB 132.17A 132.17B 139.20A 139.20B 142.4bA 142.4bB

              -

                -

   -

 -

       -

Vocative

Adjunct

Compl.

Subject

Explicit Subject

YIQTOL

Reference

Marked Theme

-

APPENDIX 5: FUNCTIONAL STRUCTURE OF UGARITIC COUPLETS WITH QYYQ VERBAL SEQUENCE

(a) q *f~l//y] qf{l (1) klat
.









DGrt





/
 bht
.
'nt 1.3 ii [she] √close the gates of the house of Anath1 3-5 Process: doing Goal Medium Fin/ Pred Complement Vocative Mood Residue Topical Rheme Theme 1 ^ +2 
 wtqry
.







Glmm


/
 bCt
.






Gr
. and [she] √meet the pages at the foot of the mountain CLocation: place Process: Scope happening Range Fin/ Pred Compl Adjunct Mood Residue Topical Rheme Theme

kla//qry

1 Pardee (1987, 250a) prefers a passive rendering of the verb and a singular noun: ‘the gate of Anatu’s house is closed’. Gibson (1977, 47), de Moor (1987, 5) and Wyatt (1998, 72) concur with our translation. Tropper (2000, 692) prefers a second person: “Sie verschloß die Tore des Palastes der Anatu, um die Burschen am Fuß des Berges zu treffen”. 499

500

(2) 1.4 iii 14-16

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Ctt










/
p[gl]t

.

bDlhny

 [you]√set [disgrace] upon my table1 CLocation: place Process: doing Goal Medium Fin/ Pred Compl Adjunct Mood Residue Topical Rheme Theme

Ct//Cty

1 ^ ×2 qlt

. / bks

. dishonour from [my] cup CLocation: Goal Medium place Compl Adjunct ResTopical MTheme Rheme

iCtynh
. [I] √drink2 it Process: doing Fin/ Mood

Pred -idue

Goal Medium Compl

1 Wyatt (2002, 96) takes this verb as an imperative: ‘take the [spindle] from my table’. All the other authorities concur with our translation, except the identity of the subject (Ginsberg 1969, 132; Caquot, Sznycer and Herdner 1974, 200; Gibson 1977, 58; de Moor 1987, 50; Pardee 1987, 258a). 2 Ginsberg (1969, 132), Gibson (1977, 58), Pardee (1987, 258a), Wyatt (2002, 96) and Tropper (2000, 884) see this verb as describing ‘the cup’ as a relative clause. Caquot, Sznycer and Herdner (1974, 201) and de Moor (1987, 50) concur with our rendering.

APPENDIX 5

(3) 1.4 vii 23-25








lrgmt
.









lk
.







lali/yn1 .
b'l
. certainly [I] √tell to you, O Mighty Ba‘al, CManner: Process: Receiver degree saying Beneficiary Adj Fin/ Pred Compl Vocative ResMood -idue Text Top Rheme Theme

501

rgm//Db

1 ^ "2 tDbn
.









b’l [you]√return, Ba‘al, Process: doing Fin/ Mood Topical Theme

Pred Res-

/ lhwty [.] to my word2 CLocation: place

Vocative

Adjunct -idue

Rheme

1 This is one of those instances where the word divisor split the word over two adjacent lines, just for practical reasons (Segert 1987, 288). 2 Gibson (1977, 65), de Moor (1987, 62), Pardee (1987, 262b), Wyatt (2002, 109) and Tropper (2000, 732) take this couplet as a long rhetorical question. Caquot, Sznycer and Herdner (1974, 216) prefer an affirmative rendering, though.

502

(4) 1.14 ii 43-45

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

yHd
.




bth
.





sgr / single man his house √shut Actor Goal Process: doing Agent Medium Subject Compl Fin/ Pred MoRes-od -idue Topical MTheme Rheme

sgr//Ckr

1 ^ +2 almnt
.



Ckr / tCkr
. the widow hireling √hire1 Actor Goal Process: doing Agent Medium Subject Compl Fin/ Pred MoRes-od -idue Topical MTheme Rheme

1 On account of parallelism, Ginsberg (1969, 143b and n.10) proposed ‘the widow locks herself in’ but he admitted that this meaning is ‘hard to confirm etymologically.’ Commentators do not agree whether the hiring person is the same with the hired one or not. Gray (1965, 13) understands that ‘the widow’ hires ‘a substitute,’ Gibson (1977, 84 n. 12) admits the alternative version and de Moor (1987, 195) interprets it as the widow’s giving of ‘a generous contribution.’ Wyatt (2002, 191) and Tropper (2000, 726) add to the dramatism of mobilisation by allowing the widow to enlist herself. Wyatt also reconstructs a N verbal form in the place of G form the text has. We retain the difficult reading. This is a verbal construction with an infinitive preceding the finite verbal form.

APPENDIX 5

(5) 1.19 iv 57-58

yd
.









mjxt
.
aq[h]t
G/zr1 . the hand [that] √smite Aqhat the hero Actor Medium Subject [[ embedded clause ]] Mood

503

mjx//mjx

Rheme 1 ^ ×2

tmjx
.








alpm


.
ib

 [she] √smite2 thousands of foe Process: doing Goal Medium Fin/ Pred Complement Mood Residue Topical Rheme Theme

1 This is one of those instances where the word divisor split the word over two adjacent lines, just for practical reasons (Segert 1987, 287). 2 Ginsberg (1969, 155b) translated the verb modally ‘can slay thousands of foes.’

504

(6) 1.100: 67-68

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

mGy
.







Hrn
.




lbth
.

 √come Horon to his house CLocation: Process: Actor happening Medium place Fin/ Pred Subject Adjunct MoRes- -od -idue Topical Rheme Theme

mGy//Cql

1 ^ =2 w/yCtql1

.




lHZrh
.
 [he] √proceed2 to his courtyard CLocation: place Process: happening Fin/ Pred Adjunct Mood Residue Topical Rheme Theme

Here Segert does not acknowledge the presence of a word divisor which split the word over two adjacent lines, although he noted several other examples (1.100: 33-34, 36-37, 39 cf. Segert 1987, 286). 2 Based on evidence from Akkadian (šaqālu), Ugaritic and Aramaic, Greenfield (1979) proposed that the lexical verb that produced this verbal form should be tstem √Cql instead, a verb of motion ‘to take, carry’ frequently used in pair with mGy, hlk or ba. Not to be confused with the Ug. √Dql ‘to weigh’. 1

APPENDIX 5

(7) 1.114: 16-18

i/l1
.






hlk

.







lbth
.
 El √walk to his house CLocation: Actor Process: happening place Medium Subject Fin/ Pred Adjunct Mood Residue Topical MTheme Rheme

505

hlk//Cql

1 ^ =2 yCtql


 / lHZrh
.
 [he]√proceed2 to his courtyard CLocation: place Process: happening Fin/ Pred Adjunct Mood Residue Topical Rheme Theme

1 Another word divisor splitting a word over adjacent lines, situation not recorded by Segert (1987). 2 Wyatt (1998, 410 n. 33) translates ‘stumble off’ as the verb is a Št form of √ql, invoking a falling motion, but Bordreuil and Pardee (2004, II: 45) does not see this necessary.

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

506

(b) y] qf{l //q *f~l (8-9) Tl
.





Cmm
.






tskh / 1.3 ii Dew [that] the heaven √pour on her 2 1 40-41 Goal Medium Compl [[ embedded clause]] Residue

nsk//nsk

Rheme 1^

=2

[r]bb

.



nskh
.





kbkbm drizzle [that]√pour on her the stars Goal Medium Compl [[ embedded clause]] Residue Rheme

Repeated in 1.3 iv 43-44. Ginsberg (1969, 136b) avoided to mention the Receiver of dew/drizzle. De Moor (1987, 7) reads ‘for her’ instead. Gibson (1977, 48) and Wyatt (2002, 76) concur with our translation. Tropper (2000, 711) agrees too, but he draws near the verb from a prior line. 1 2

APPENDIX 5

(10-12) 1.3 iii 34-351

34tGx
.









pnt





35kslh
.

507

nGx//anC

√be contracted2 the knuckels of her back Process: doing Actor Medium Fin/ Pred Subject MoResidue -od Topical Rheme Theme

1 ^ =2 anC
.










dt
.


Zrh
.
 √go weak3 the one of her back4 Process: doing Actor Medium Fin/ Pred Subject MoResidue -od Top Rheme Theme

It also appears fragmentarily in 1.4 ii 19-20 and again in 1.19 ii 45-49. Ginsberg (1969, 132a) read it as passive ‘are bent’. Marcus (1970-71) sees here a qal passive from √nGx corresponding to the N conjugation. 3 Gibson (1977, 50) and Wyatt (1998, 79) read here ‘muscles’ for what de Moor (1987, 10) has as ‘the small’. Taken as a verb, the parallelism becomes complete (Ginsberg 1969, 132a; Pardee 1987, 252a). Cf. DULAT. 4 Most probably ‘the spine’ (cf Pardee 1987, 252a). Ginsberg (1969, 132a n. 18) reads into this description ‘the standard reaction of a female character to an unexpected visit’. 1 2

508

(13) 1.4 v 26-27

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

26tbCr











b'l


bCr//ybl

√receive good tidings1 Ba‘al Process: happening Fin/ Pred Mood Residue Topical Theme

Vocative Rheme

1 ^ =2 27bCrtk
.



yblt

good tidings √to be given Goal Process: Medium happening Subject Fin/ Pred Mood Topical MTheme Rheme

to you2 Recipient Beneficiary Compl

1 Wyatt (2002, 102) translates ‘rejoice’ to what Gibson (1977, 61) has a passive form ‘be gladdened’. De Moor (1987, 55) and Pardee (1987, 260a) concur with our translation. Tropper (2000, 572) prefers to see the yqtl as an imperative. 2 Ginsberg (1969, 133b) reads the couplet as one line: ‘Receive, Baal, the glad tidings I bring thee.’

APPENDIX 5

(14) 1.4 vi 38-40

'dbt
.










bht[h
.


b']l
 the arrangements1 of his house Ba‘al Goal Actor Medium Agent Complement Subject ResMood Topical Rheme Marked Theme

509

/ y'db
. √arrange Process: doing Fin/ Pred -idue

'db//'db

1 ^ =2 hd
.





'db [.






'd]bt



 / hklh Hadu √arrange the arrangements of his palace Actor Process: Goal doing Agent Medium Subject Fin/ Pred Complement Mood Residue Topical MTheme Rheme

1 Gibson (1977, 63) translates neutrally ‘put his palace in order,’ but de Moor (1987, 60) and Pardee (1987, 261b) prefers ‘furnishings,’ whereas Wyatt (2002, 106 n. 148) prefers a cultic term ‘offerings.’ One thing is certain, though, that the verb and the noun of each clause are built on the same root. Tropper (2000, 711) concurs with our translation.

510

(15) 1.5 i 16-17

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

hm
.



brky
.





tkCd / rumm
. perhaps the pool √search wild bulls1 Behaver Process: Behaviour behavioural Medium Range Adjunct Subject Fin/ Pred Compl ResMood -idue Text Top MTheme Rheme

kCd//kfd

1 ^ =2 'n
.






kfd
.






aylt

 a spring √yearn2 a hind Behaver Process: Behaviour Medium behavioural Range Subject Fin/ Pred Complement Mood Residue Topical MTheme Rheme

1 Translators are divided on the matter of Subject-Object of this clause and the following one. Whereas, Gibson (1977, 68) and de Moor (1987, 70) see the pool as the entity that attracts wild-oxen, Pardee (1987, 264b cf. n. 205) and Wyatt (2002, 117) consider that the ‘wild bulls are yearning for pools’. The meaning of the verb (cf. DULAT) restricts the possibilities to the first one. 2 This verse-unit constitutes the most problematic issue of this line. Wyatt (2002, 118) follows Caquot, Sznycer and Herdner (1974, 242 n. r) by reading this verb as kCd, thus a qtl form of the verb root in the previous line.

APPENDIX 5

(16) 1.5 ii 6-7





yraun

.













aliyn
.
b'l / √fear him Mighty Ba‘al Process: Behaviour Behaver behavioural Range Medium Fin/ Pred/ Compl Subject MoResidue -od Topical Rheme Theme

1 ^ =2 Dt'.
nn

.
















rkb
.
'rpt
 √scare of him the Cloud-Rider Process: Behaviour Behaver behavioural Range Medium Fin/ Pred Compl Subject MoResidue -od Topical Rheme Theme

511

yra//Dt'

512

(17) 1.6 vi 30-31

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS






yru
.



bn



ilm t
. √afraid1 son of gods Mot Process: Behaver behaviour Medium Fin/ Pred Subject MoRes -od Topical Rheme Theme

yra//Dt'

1 ^ =2 



Dt'
.





y/dd2 .

il
.
Gzr
[.] √scare the beloved of El the hero Process: Behaver behaviour Medium Fin/ Pred Subject MoRes -od Topical Rheme Theme

1 Marcus (1969, 55-56) maintains the idea that when the verb is stative and/or the predicate precedes the subject, such a verbal form can be a finite verbal form, the simplest of all, that is third masculine singular. 2 This is one of those instances where the word divisor split the word over two adjacent lines, just for practical reasons (Segert 1987, 288).

APPENDIX 5

(18-19) 1.19 iii 8-91

knp
.


nCrm
.

b'l
.







yDbr / wings of eagles Ba‘al √break Goal Actor Process: doing Medium Agent Complement Subject Fin/ Pred ResMood -idue Topical Rheme Marked Theme

1 ^ =2 b'l
.








Dbr




diy
hmt2 Ba‘al √break their pinions Actor Process: Goal doing Agent Medium Subject Fin/ Pred Complement Mood Residue Topical MTheme Rheme

1 2

It also appears in 1.19 iii 22-23. Margalit (1989, 402) reads here hwt.

513

Dbr//Dbr

APPENDIX 6: ARCHAIC HPY COUPLETS WITH QYYQ VERBAL SEQUENCE

(a) q *f~l//y] qf{l ^l,êg>r;l. WKåTu ‘~hew> Deut 33.3bA `^yt,(roB.D;mi aF'ÞyI Deut 33.3bB

qtl // yqtl

Alê br'ä ‘wyd'y" Deut 33.7cA s `hyyI ^ßt.yrI)b.W Deut 33.9bB

qtl // yqtl

!keêv' ‘!Der.Y:h; rb,[eÛB. d['ªl.GI Judg 05.17aA tAY=nIa\ rWgày" hM'l'î !d'§w> Judg 05.17aB

qtl // yqtl

~yMiêy: @Axål. ‘bv;y" rveªa' Judg 05.17bA `!AK)v.yI wyc'Þr'p.mi l[;îw> Judg 05.17bB

qtl // yqtl

^ên>ymiäy> ‘t'yji’n" Exod 15.12A `#r,a'( Am[eÞl'b.Ti Exod 15.12B

qtl // yqtl

!v'Wk yleh\a' ytiyair' !w tmoßhoT. Exod 15.05A `!b,a'(-AmK. tl{ßAcm.bi Wdïr>y" Exod 15.05B

yqtl // qtl

APPENDIX 7: DISTRIBUTION OF COUPLETS IN P SALMS Psalm

total verses 1 6 2 13 3 8 4 7 5 15 6 10 7 18 8 9 9 20

10 20

monolines 12b

couplets 2, 4, 5, 6 1, 3, 4, 5, 6, 9, 10, 11 2b, 3, 4, 5, 6, 7, 9 3, 4, 5, 8 2, 4, 5, 6, 7, 8, 9, 10a, 10b, 11a, 11b, 12a, 12b, 13 2, 3, 4, 5, 6, 8, 9, 10, 11 2, 3, 4, 5, 8, 10a, 10b, 11, 12, 13, 14, 15, 16, 17, 18 4, 5, 6, 7, 8, 9, 10 2, 3, 4, 5, 6, 8, 9, 10, 11, 12, 13, 14, 16, 17, 18, 19, 20, 21 1, 2, 3, 4, 6, 7, 8a, 8b-9a, 9b, 10, 11, 12, 13, 14b, 15, 16, 17

517

tercets

quatrains 1, 3

2, 7, 8, 12a 8 2, 6-7, 9 3 7 6, 7, 9 2, 3 7, 15

5, 14a, 18

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

518

Psalm 11 12 13 14 15 16

total verses 8 7 6 7 6 11

17 17 18 51

19 14 20 9 21

13

22 31

23 5 24 10

monolines 6b 5b 1b

couplets

tercets

3, 4a, 4b, 6, 7 2, 3, 8, 9 2, 4, 5, 6a 3, 5, 6 1b, 2, 5a 2, 3, 5, 6, 7, 8, 9, 10 1b, 1c, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 15 2, 3, 4, 5, 6, 7, 10, 11, 13, 15, 16a, 16b, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 32, 33, 34, 35, 37, 38, 39, 40, 41, 42, 43, 44a, 44b-45a, 47, 49b, 50 2, 3, 4, 5a, 8, 9, 10, 11, 12, 13 2, 3, 4, 5, 8, 9, 10 2, 3, 4, 5, 6, 7, 8, 9, 11, 12, 13, 14 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15a, 15b, 18, 19, 20, 21, 22, 23, 26, 31, 32 5, 6 1b, 2, 3, 5, 6

1b, 2, 5 6, 7 3 1b, 2, 4, 7 3, 4 4, 11 3, 14a, 14b 8, 9, 12, 14, 31, 36, 45b46, 48-49a, 51

5b-6, 7, 14, 15 6, 7 10 16, 17, 24, 25, 27, 2829, 30 1-2, 3, 4 4, 7, 8, 9, 10

quatrains 4-5

APPENDIX 7

Psalm

total verses 25 21

monolines

26 12 27 17 28 12 29 9 30 12 31

27

32 10 33 22

34 22

35 31

4

519

couplets

tercets

3, 4, 6, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 1b, 2a, 2b, 3a, 3b, 4a, 4b, 7, 8, 10, 11, 12, 13 1b, 1c, 2, 3a, 3b, 6, 7a, 7b, 8 5, 6, 10, 11 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13 4, 5, 6, 7, 9, 10, 11a, 11b, 13, 14a, 14b, 15, 16, 17, 18, 19, 20, 21, 22, 23a, 23b, 25 3, 4, 7, 8, 10, 11 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 1b, 2, 3, 4a, 4b, 5, 6, 7, 9, 10a, 10b, 11, 12, 14, 16, 18, 19, 22, 23, 24, 25, 26a, 26b, 28

1b-2, 5, 7

quatrains

1b 5, 6, 14

9

4, 5, 9 3, 7-8, 9 2, 3, 8, 12, 24

1b-2, 5, 6, 9

8, 13, 15, 17, 20, 21, 27

1b-2

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

520

Psalm

total verses 36 12 37 42

38 22

39 15 40 24

41

13

42 11 43 5 44 28

45 17

monolines

couplets

tercets

2, 3, 4, 6, 8, 9, 10, 11, 12, 13 1b, 2, 3, 4, 5, 6, 8, 9, 10, 11, 12, 13, 14a, 14b, 15, 16, 17, 18, 19, 21, 22, 23, 24, 26, 27, 28a, 28b, 29, 30, 31, 32, 33, 35, 36, 37, 38, 39 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 2a, 2b, 8, 9, 10, 11, 13b, 14 3a, 3b, 4a, 4b, 5a, 5b, 6a, 8, 9, 12, 14, 15a, 15b, 16, 17a, 17b, 18a, 18b 2, 3, 4, 5, 6, 8, 9, 11, 12, 13 2, 3, 4, 8, 10, 11 2 2a, 2b, 3, 4a, 4b, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27 4, 7, 9, 10, 11, 12, 13, 14, 16, 17, 18

5, 7 7, 20, 25, 34, 40

13

3, 4, 5, 6, 7, 12, 13a 2, 6b, 7, 10, 11, 13 7, 10, 14 5, 6, 7, 9, 12 1, 3, 4, 5

2, 3, 5, 6, 8, 15

quatrains

APPENDIX 7

Psalm

total verses 46 11

monolines

47 10 48 14 49 21

50 24

51

21b

20

52 9

couplets

tercets

2, 3, 4, 5, 6, 7, 8, 9, 11, 12 2, 3, 4, 5, 6, 7, 8, 9, 10a, 10b 2, 4, 5, 6, 7, 8, 10, 11a, 13 2, 3, 4, 5, 6, 7, 8, 9, 10, 13, 14, 15a, 15b, 16, 17, 18, 19, 20, 21 3b, 4, 5, 6, 8, 9, 10, 11, 12, 13, 14, 15, 17, 18, 19, 20, 21a, 22 3, 4, 5, 6a, 6b, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20, 21 3, 4, 5, 6, 8, 10, 11

10

53 6 54 7 55 25

56 14 57 15 58 11

4b

521

3, 4, 6, 7, 8, 9 5, 6, 7, 8, 9, 10, 11, 12, 14, 15, 17, 19, 20a, 20b, 21, 22a, 22b, 24a 2, 3, 4, 6, 8, 10, 11, 12, 13, 14a, 14b 2a, 2b, 3, 4a, 5a, 5b, 6, 7a, 7b, 10, 11, 12 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12

3, 9, 14, 15

quatrains

11b-12

11, 12

1b, 2-3a, 7, 16, 23 19

7 2, 3, 4, 5, 6, 7 5 2-3a, 3b-4, 13, 16, 18, 23, 24b 5, 7, 9 8, 9

9

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

522

Psalm

total verses 59 19

monolines

60 12 61 8 62 16

13b

63 11 64 10 65 66 19 67 7 68 37

69 38

70 6

36b

couplets

tercets

2, 3, 5, 6a, 6b, 7, 9, 10, 11, 15, 16, 17a, 17b 3, 4, 5, 6, 7, 11, 12, 13, 14 2, 4, 5, 6, 7, 8, 9 2, 3, 5a, 5b, 6, 7, 8, 10a, 10b, 11a, 11b, 12a, 12b13a 3, 4, 5, 6, 7, 8, 9, 10, 11 2, 3, 4, 5, 8, 9 2, 3, 4, 5, 6, 7, 8, 9, 10a, 10b, 11, 12, 13 3, 4, 5, 8, 9, 10, 11, 13, 14, 17, 18, 19 2, 3, 4, 6, 7, 8 2, 6, 8, 10, 11, 12, 13, 15, 16, 18, 20, 21, 22, 23, 24, 25, 26, 27, 30, 31a, 31b, 32, 33, 34, 2, 3, 4, 6, 7a, 7b, 8, 9, 10, 11, 12, 13, 14a, 14b, 15, 17, 18, 19, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 37 2, 3, 4, 5, 6a, 6b

4, 8, 12, 13, 14, 18

quatrains

8, 9, 10 3 9

4

12

2

6, 7, 10, 11 14 1b-2, 6, 7, 12, 15, 16, 20 5 3, 4, 5, 7, 9, 14, 17, 19, 28, 29, 35, 36a 5, 16, 20, 21, 36

APPENDIX 7

Psalm

total verses

71

72

73 28

74 23

75 11 76 12 77 20

monolines

523

couplets

tercets

1, 2, 4, 5, 7, 8, 9, 10, 11, 12, 14, 16, 17, 18a, 18b, 23, 24 1b, 2, 3, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 17a, 17b, 18 1b, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27 1b, 3, 4, 5, 6, 7, 8, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 3, 4, 5, 6, 7, 8, 9b, 10, 11 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13 2, 4, 5, 6, 8, 9, 10, 11, 12, 13, 14, 15, 16, 21

3, 6, 13, 15, 19, 20, 2122 4, 15, 16, 19

28

2, 9

2, 9a 6 3, 7, 17, 18, 19, 20

quatrains

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

524

Psalm

total verses

monolines

78

79 15 80 20

81

16

82 8 83

84 11 85 13

13

couplets

tercets

1b, 2, 3, 5a, 5b, 8a, 8b, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20a, 20b, 22, 23, 24, 25, 26, 27, 28, 29, 30, 32, 33, 34, 35, 36, 37, 38a, 38b, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 51, 52, 53, 54, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 72 3, 4, 5, 7, 9a, 9b, 10a, 10b, 11, 12 2a, 2b-3a, 3b, 4, 5, 6, 7, 8, 9, 11, 12, 13, 14, 15a, 17, 18, 19, 20 2, 3, 4, 5, 7, 9, 10, 12, 13, 14, 15, 16, 17 1b, 2, 3, 4, 6, 7, 8 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 5, 6, 7, 8, 9, 10

4, 6, 7, 19, 21, 31, 49, 50, 55, 71

2, 3, 4, 5, 6, 7, 8, 10, 11, 12, 13, 14

1b, 2, 8, 13 10, 15b-16

6, 8, 11 5

2-3, 4, 11, 12 9

quatrains

6

APPENDIX 7

Psalm

total verses 86 16

monolines

87 5 88 19

7

89 52

53

90 18 91

16

92 14 93 5 94 23

95 11

525

couplets

tercets

1b, 3, 4, 5, 8, 10, 11, 12, 13, 15 6 2, 3, 4, 5, 6a, 6b, 7, 8, 11, 12, 13, 14, 15, 16, 17, 18, 19 2, 3, 4, 5, 6, 7, 8, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52 1b, 3, 5, 6, 7, 8, 9, 10a, 10b, 11, 12, 13, 14, 15, 16 1, 3, 5, 6, 8, 9, 10, 11, 12, 13, 14, 16 2, 3, 4, 5, 6, 7, 11, 13, 14, 15, 16 2 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 1, 2, 3, 4, 5, 6, 7a, 9, 11

2, 6-7, 9, 14, 16, 17 1b-3, 4, 5 9, 10

quatrains

9, 20

2, 4, 17 2, 4, 7, 15 10, 12 1, 3, 4, 5

7b-8, 10

8-9

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

526

Psalm

total verses 96 12

monolines

97 12 98 11 99 10 100 101 14

102 29

103 23

22b

104 39

1a, 1b, 35b, 35c

couplets

tercets

1, 2, 3, 4, 5, 6, 13b

7-8a, 8b-9, 10, 11-12a, 12b-13 7, 8, 9, 10

1, 2, 3, 4, 5, 6, 11, 12 1a, 1b, 2, 3a, 3b, 6, 7, 8, 9a, 9b 1, 2, 3, 4a, 4b, 6a 5 1b, 2a, 2b, 3a, 3b, 4, 5a, 5b, 6a, 6b, 7a, 7b, 8a, 8b 2, 3a, 3b, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 26, 28, 29 1b, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 18, 19, 21, 22a, 1c-2a, 2b-3a, 3b, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 16, 17, 18, 19, 20, 21, 22, 23, 26, 27, 28, 30, 31, 32, 33, 34, 35a

quatrains

4-5 5, 6b-7, 8, 9 1b-2, 3, 4

25, 27

17, 20

14, 15, 24, 25, 29

APPENDIX 7

Psalm

total verses 105 46

monolines 45b

106 53

1a

107 42

108 13

couplets 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45a 1b, 2, 3, 4, 6, 7a, 7b, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23a, 23b, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 39, 40, 41, 42, 44, 45, 46, 47a, 47b, 48a, 48b 1, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 35, 36, 37, 38, 39, 40, 41, 42, 43 2, 3, 4, 5, 6, 7, 11, 12, 13, 14

527

tercets

quatrains

5, 43

38

3

33-34

8, 9, 10

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

528

Psalm

total verses 109 31

monolines 1b

110 7 111 11

1a

112 11

1a

113 11

1a, 9b

114 8 115 19

18b

116 20

19b

117 3 118 29

2b

119 176 120 7 121

8

122 8 123 4

couplets

tercets

3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 17, 19, 20, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31 1b, 2, 5, 6, 7 1b, 2, 3, 4, 5, 6, 7, 8 1b, 2, 3, 4, 5, 6, 7, 8 1b, 2, 3, 4, 5, 6, 7, 8, 9a 1, 2, 3, 4, 5, 6, 7, 8 2, 3, 4, 5, 6, 8, 9, 10, 11, 13, 14, 15, 16, 17, 18a 1, 2, 4, 5, 6, 7, 9, 10, 11, 12, 13, 14, 15, 17, 18, 19a 1, 2a 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15a, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 28, 29 all 176 verses 1b, 2, 3, 4, 5, 6, 7 1b, 2, 3, 4, 5, 6, 7, 8 1b, 5, 6, 7, 8, 9 1b, 3

2, 16, 18, 21

quatrains

3, 4 9, 10 9, 10

1, 7, 12 3, 8, 16

12, 15b-16, 27

4 4

2-3 2

APPENDIX 7

Psalm

total verses 124 8

monolines

125 7 126 8

5b

127 8 128 7 129 9

6b 8b

130 7 131 5 132 19

133 3 134 3 135 23

136 26 137 11 138 8 139 23

140 13 141

10

21b

couplets 3, 4, 5, 6, 7a, 7b, 8 3a, 3b, 4, 5a 1b, 2a, 2b, 3, 4, 5, 6a, 6b 1b, 1c, 2a, 2b, 3, 4, 5a, 5b 1b, 2, 3a, 3b, 4 1b, 2, 3, 4, 5, 6, 7, 8a 1b-2a, 2b, 3, 4 1b, 1c, 2a, 2b, 3 1b, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11a, 11b, 13, 14, 15, 16, 17, 18 1b 2, 3 1b, 2, 3, 4, 8, 10, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21a, 21b all 26 verses 2, 4, 5, 6a, 6b, 7a, 7b, 9 1b, 3, 4, 5, 6 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 17, 18, 19, 20, 21, 22, 23, 24 2, 3, 4, 7, 8, 9, 10, 12, 13, 14 1b, 2, 3, 6, 7, 8, 9, 10

529

tercets

quatrains 1b-2

1b, 2

5-6a 5, 6

7 12

2, 3 1b 5, 6, 7, 9, 11

1, 3, 8 2, 7, 8 1b-2, 12, 14, 15, 16 5, 6, 11 4, 5

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

530

Psalm

total verses 142 11

monolines

143 13 144 18

145 21

146 10

10b

147 22

1a, 20b

148 16

1a, 14b

149 11

1a, 9b

150 7 2529

1a 39

TOTAL

couplets

tercets

2, 3, 4a, 4b, 5a, 5b, 7a, 7b, 8a, 8b 2, 4, 6, 8a, 8b, 9, 11 2a, 2b, 3, 4, 5, 6, 8, 9, 10, 12a, 12b, 13a, 13b, 15 1b, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 1, 2, 3, 4, 5, 9b, 10a

6

1b, 2, 3, 4, 5, 6, 7, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20a 1b, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 1b, 2, 3, 4, 5, 6, 7, 8, 9a 1b, 2, 3, 4, 5, 6 2126 (= 84%)

8

quatrains

1b, 3, 5, 7, 10, 12 1b, 7, 11, 14

6-7a (pentet), 7b-9a (pentet)

13, 14a

381 (=15%)

23

APPENDIX 8: VARIATIONS OF VERSES WITH THE QYYQ VERBAL SEQUENCE

Distributio n of couplets with a QYY Q verbal sequence on one of its lines q*f~l//y]qf{l on line A y]qf{l//q*f~l on line A 10:6; 13.5; 17.6; 18.44; 38.17; 39.10; 16.7; 31.5; 58.11; 90.6; 110.6 56.12; 68.23; 69.33; 90.5; 104.9; 116.1, 10; 119.73, 163; 140.13 q*f~l//y]qf{l on line B y]qf{l//q*f~l on line B 3.6; 7.13; 10.13; 21.12; 35.11; 41.9; 10.2; 18.20; 41.6; 51.10; 90.6; 132.14 46.7; 60.3; 68.31b Distributio n of tercets with a QYYQ verbal sequence base q*f~l//y]qf{l//X X//q*f~l//y]qf{l X=qtl 18.9; 39.4; 40.10; 137.8

X=qtl 77.18; 88.9; 93.3; 116.3; 118.12; 143.5

X=yqtl 6.7; 66.6

X/X//q*f~l//y]qf{l

X=vol 11.1b

X=yqtl 64.6

X=nom 59.4

X=wayqtl 38.13

X=infc 10.14a; 11.2

X=vol 140.5 531

532

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

q*f~l//y]qf{l//X/X X=yqtl 65.14

X=Ø 24.4 X/Y//q*f~l//y]qf{l X=infc; Y=nom 35.13

y]qf{l//q*f~l//X

X=qtl; Y=yqtl 82.5 X//y]qf{l//q*f~l

X=qtl 2.7

X=yqtl 69.36

X=yqtl 50.23; 68.17; 87.5

X/X//y]qf{l//q*f~l

X=Ø 2.2; 64.7

X=yqtl 49.11; 52.7; 55.13; 56.7 X/Y//y]qf{l//q*f~l X=qtl; Y=wayqtl 22.30; 78.21

APPENDIX 9: AN OVERVIEW OF VERSE-LINE DELIMITATION IN P SALMS AS PRESERVED BY THE MASORETES AND COMPARED WITH ANCIENT, MEDIEVAL AND MODERN AUTHORITIES TEXT hw"+hyl; rv"ï-rv,a] dwIïd'ñl. !Ay©G"vi `ynI)ymiy>-!B, vWk©÷-yreb.DI-l[; ytiysi_x' ^åB. yh;l{a/â hw"åhy> `ynIlE)yCih;w> yp;ªd>ro÷-lK'mi ynI[EïyviAh yvi_p.n: hyEår>a;K. @roæj.yI-!P, `lyCi(m; !yaeäw> qreªPo÷ tazO= ytiyfiä['-~ai yh;l{a/â hw"åhy> `yP'(k;B. lw yvi‡p.n: ŸbyE“Aa @Doðr;yI) yY"+x; #r,a'äl' smoår>yIw> `hl's,( !KEåv.y: rp"ß['l, ŸydI¦Abk.W

Reference

Verb usage

07.01

Superscript

07.02A 07.02B1 07.03A2 07.03B 07.04A 07.04B 07.05A 07.05B3 07.06A4 07.06B5 07.06C 6

qtl // vol / vol yqtl // nom / nom qtl // nom qtl // wayqtl vol / vol // vol // vol

A renders this verse as a ‘log’. Dahood (1965b, 40) splits this line into two (2+2), resulting a new couplet, without the support of line 07.03B, which stays by itself as a monoline. 3 Wellhausen (1895) suspected two lines being missing between the current A and B lines. 4 Wellhausen (1895) transposes gvyw to the next line. 5 GHB and Langa (2000, 34) prefer to merge line A with line B. Langa also erroneously places the Sela mark after the first line. 6 Dahood (1965b, 40) prefers this lines as three monolines. 533 1 2

534

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

TEXT ^P,ªa;B. Ÿhw"“hy> hm'ÛWq« yr'_r>Ac tAråb.[;B. afeN"hiâ `t'yWI)ci jP'îv.mi yl;ªae÷ hr'W[ïw> &'b÷ ~yMiî[;ñ !ydIçy" éhw"hy> yqIßd>ciK. hw"+hy> ynIjEïp.v' `yl'([' yMiätuk.W é~y[iv'r> Ÿ[r;’ an"í-rm'g>yI qyDIîc;ñ !nEáAkt.W tAy©l'k.W tABªliâ !xEåboW `qyDI¥c; ~yhiîl{a/ ~yhi_l{a/-l[; yNIïgIm") `ble(-yrev.yI [;yviªAm÷ qyDI_c; jpeäAv ~yhil{a/â `~Ay*-lk'B. ~[eîzO laeªw>÷ vAj+l.yI ABær>x; bWvy"â al{å-~ai `h'nAky>w:) %r;ªd'÷ ATïv.q;

Reference 07.07A1 07.07B 07.07C 2 07.08A 07.08B 07.09A 07.09B3 07.09C 4 07.10aA5 07.10aB 07.10bA 07.10bB6 07.11A 07.11B7 07.12A8 07.12B9 07.13A10 07.13B11

Verb usage vol // vol // vol vol // vol vol // vol // nom vol // yqtl nom // nom nom // nom nom // nom yqtl / yqtl // qtl / yqtl

A transposes $pab to the next line. A, Dahood (1965b, 40), NJB, and Craigie (1983, 97) split this line into two (2+2), resulting in a new couplet. 3 Wellhausen (1895) and Langa (2000, 34) transposes yqdck o the next line. 4 GHB renders this verset as a couplet instead, crediting the Masoretic markers to divide the lines (5+3). 5 With the exception of ~y[vr, which joins the next line, A merges line 10aA with line 09bB and 09bA. 6 a and ç transpose ~yhla to the previous line and qydc to the next line. GHB renders this verset as a couplet instead (5+5). Wellhausen (1895) inverted the lines of this verse. 7 a, ç, and A render this verset as a monoline. 8 a adds kai. iv scuro.j kai. makroqumoj to the line. 9 ç merges line B with line A. 10 Dahood (1965b, 41), NJB, and Langa (2000, 35) split this line into two, developing the verse into a tercet (2+2+3). Craigie (1983, 97) promotes the discrimination of the first half against the other as an individual unit (monoline). 11 A scans this verse as a ‘log’. 1 2

APPENDIX 9

TEXT tw hNEïhi `rq,v'( dl;y"åw> lm'ª['÷ hr'îh'w> Whre_P.x.Y:w:) hr'(K'â rABæ `l['(p.yI tx;v;äB. lPoªYiw:÷ Av=arob. Alåm'[] bWvåy" `dre(yE Asðm'x] Adªq\d>q'÷ l[;îw> Aq+d>ciK. hw"åhy> hd,äAa `!Ay*l.[, hw"ïhy>-~ve( hr'ªM.z:a]w:÷ ŸdwI“d'l. hw"©hy> ynIjEÜp.v' yTik.l;_h' yMiätuB. ynIa]â-yKi( `d['(m.a, al{å yTix.j;ªB'÷ hw"ïhyb;W ynISE+n:w> hw"åhy> ynInEåx'B. `yBi(liw> yt;äAyl.ki hp'Þr>c' yn"+y[e dgv"+-ytem.-~[i yTib.v;y"â-al{ `aAb)a' al{å ~ymiªl'[]n:÷ ~[iîw> ~y[i_rem. lh;äq. ytia nEf'â `bve(ae al{å ~y[iªv'r>÷-~[iw> yP'_K; !AyæQ'nIB. #x;är>a, `hw")hy> ^åx]B;z>mi-ta, hb'Þb.soa]w:

Reference 07.14A 07.14B 07.15A 07.15B1 07.16A 07.16B 07.17A 07.17B 07.18A 07.18B2 26.01a 26.01bA3 26.01bB 26.01bC 4 26.02A5 26.02B 26.03A 26.03B 26.04A6 26.04B 26.05A 26.05B 26.06A 26.06B7

535

Verb usage qtl // yqtl yqtl // qtl / qtl qtl / wayqtl // wayqtl / yqtl yqtl // yqtl vol // vol Superscript vol // qtl // qtl / yqtl vol / vol // vol nom // qtl qtl // yqtl qtl // yqtl yqtl // yqtl

1 ç merges line B with line A. Dahood (1965b, 41) and Craigie (1983, 97) render verse 8.15 as a tercet, with this line split into two (2+2). Watson (1989c, 168) supports our delimitation on grounds of internal parallelism. 2 Kraus (1988, 168, note u) proposes that YHWH should be deleted since its presence is made absolete by its appearance in the first line and for metrical reasons. 3 This division is supported by A, NJB, Kraus (1988, 324) and Langa (2000, 74). 4 Dahood (1965b, 160) renders this verset as two monolines and a sandwiched couplet (2, 3+2, 2). Fokkelman (2000, 405) renders this verset as a double couplet (2+3, 2+2). 5 A merges line 26.02A with line 26.01bB, forming a ‘log’. 6 A merges line 26.04A with line 26.03B, forming a ‘log’. 7 Kraus (1988, 325, note d) views hwhy as an addition.

536

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

TEXT hd'_AT lAqåB. [;miv.l;â `^yt,(Aal.p.nI-lK' rPeªs;l.W÷ ^th;l. dwIåd'l. rAmàz>mi ynIxE+ykiAt ^ïP.c.q,B.-la; hw"©hy>) `ynIrE)S.y:t. ^ït.m'x]b;W¥ ybi_ Wtx]nIå ^yC,xiâ-yKi( `^dw: tAW=h; WråB.DI yti['r'â yveär>dow> `WG*h.y< ~AYðh;-lK' tAmªr>miW÷ [m'_v.a, al{å vrexek.â ynIåa]w: `wyPi(-xT;p.yI al{å ~Leªaik.W÷ [;me_vo-al{ rv ^ål.-yKi( `yh'(l{a/ yn"ïdoa] hnhi yl;î[' yliªg>r;÷ jAmïB. !Ak+n" [l;c,äl. ynIa]â-yKi( `dymi(t' yDIäg>n< ybiÞAak.m;W dyGI+a; ynIïwO[]-yKi( `yti(a J'x;me( ga;ªd>a,÷ Wmce_[' ~yYIåx; yb;y>aow>â) `rq,v'( ya;än>fo WBßr;w> hb'_Aj tx;T;ä h['r'â ymeäL.v;m.W `bAj)-ypid>r'( tx;T;ä ynIWn©j.f.yI÷ hw"+hy> ynIbEïz>[;T;(-la; `yNIMTi-la; yh;ªl{a/÷ yti_r'z>[,l. hv'Wxï `yti([‚WvT. Yn”©doa]÷ dwI+d'l. rAmðz>mi `hd'(Why> rB:ïd>miB. AtªAyh.Bi÷ &'r hY"ßci-#r,a,(B. 1

Reference

Verb usage

38.16A 38.16B1 38.17A 38.17B2 38.18A 38.18B 38.19A 38.19B 38.20A 38.20B 38.21A 38.21B 38.22A 38.22B3 38.23A 38.23B4 63.01

qtl // yqtl

63.02A 63.02B5 63.02C 63.02D6

nom / yqtl // qtl // qtl // ø

a scans this verset as a monoline.

qtl / yqtl // qtl nom // nom yqtl // yqtl qtl // qtl ø // yqtl vol // vol vol // ø Superscript

A displays RFSO (2:2). Dahood (1965b, 232) segregates the last two verseunits against the rest of the line as a distinct monoline. 3 ç scans this line as a monoline. Wellhausen (1895) prefers yhla in the previous line. 4 a and ç scan this verset as a monoline. A merges this verse with line 38.22B as one ‘log’. 5 A prefers a different arrangement of the verse-units in lines B and C (2+4). a, NJB, Kraus (1989, 17), Tate (1990, 123), and Langa (2000, 164) read this verse as a double couplet (3+3, 3+3). Dahood (1968, 95) prefers a tercet and monoline. Tate would place the introductory vocative as anacrusis. This line knows a different rendering at Fokkelman (2003, 347) who prefers this verset as a double couplet with shorter lines (3+4, 3+3). 6 A splits this line into two (1+2). Kraus (1989, 18, note d) deletes ~ym-ylb as accretion. 2

APPENDIX 9

TEXT ^yti_yzIx] vd,QOåB; !Keâ `^d[u÷ tAaïr>li ~yYI©x;me( ^D>s.x;â bAjå-yKi `^n>Wx)B.v;y> yt;îp'f. yY"+x;b. ^åk.r,b'a] !KEå `yP'(k; aF'îa, ^m.viB.÷ yvi_p.n: [B;äf.Ti !v,d,w"â bl,xeä AmÜK. `yPi(-lL,h;y> tAn©n"r>÷ yteîp.fiw> y['_Wcy>-l[; ^yTiîr.k;z>-~ai `%B")-hG[, t'yyIåh'-yKi( `!NE)r;a] ^yp,än"K. lceÞb.W ^yr,_x]a; yviäp.n: hq"åb.D' `^n hk'îm.T' yBi÷ª yvi_p.n: Wvåq.b;y> ha'Avl.â hM'heªw> `#r,a'(h' tAYðTix.t;B.( Waboªy"÷ br,x'_-ydey>-l[; WhruîyGIy: `Wy*h.yI ~yliä['vu tn"ßm.

Reference 63.03A 63.03B1 63.04A2 63.04B 63.05A3 63.05B 63.06A4 63.06B5 63.07A 63.07B 63.08A 63.08B6 63.09A 63.09B 63.10A 63.10B7 63.11A8 63.11B9

539

Verb usage qtl // ø nom // yqtl yqtl // yqtl yqtl // yqtl qtl // yqtl qtl // yqtl qtl // qtl yqtl // yqtl yqtl // yqtl

ç merges line B with line A. A transposes ~yyxm to the next ‘log’. 3 A merges line 63.05A with line 63.04B as one ‘log’. 4 A displays RFSO (3:2). 5 Dahood (1968, 95) scans this verse as a double couplet (3+2, 2+1). 6 Dahood (1968, 95) extends the current couplet into a tercet, by splitting this line into two (3+2+1). 7 A displays RFSO (1:3). 8 A displays RFSO, transposing brx to the next ‘brick’. Dahood (1968, 95) discriminates the first vrse-unit against the rest of the line, producing a new couplet. 9 ç scans this verse as a monoline. Dahood (1968, 95) renders this line as a monoline. 1 2

540

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

TEXT

Reference

~yhiîl{ña Be xm;çf.yI é%l,M,h;w> 63.12A1 AB+ [B'äv.NIh;-lK' lLeh;t.yIâ 63.12B2 `rq,v'(-yreb.Ad) yPiä rkeªS'yI÷ yKiî 63.12C 3

Verb usage yqtl // yqtl // yqtl

A displays RFSO (1:2). A transposes wb to the next ‘brick’. 3 After adopting wb from the previous ‘brick’, A splits into two ‘bricks’ (3+3). Dahood (1968, 95) renders this line as a monoline. Fokkelman (2003, 347) renders this line as a monoline. 1 2

INDEX OF SUBJECTS Complement, 89, 90, 100, 101, 102, 103, 104, 107, 108, 109, 112, 120, 130, 131, 156, 160, 161, 162, 169, 183, 184, 187, 189, 191, 192, 193, 194, 196, 197, 198, 199, 201, 202, 208, 209, 216, 217, 218, 235, 236, 239, 284, 286, 292, 295, 296, 301, 335, 342 constituency, 77 couplet, 7, 11, 16, 23, 26, 34, 41, 45, 59, 64, 67, 273, 277, 278, 279, 280, 284, 285, 286, 287, 288, 289, 290, 291, 293, 295, 296, 298, 302, 303, 304, 305, 306, 307, 309, 313, 314, 315, 316, 320, 326, 327, 328, 329, 330, 331, 332, 335, 337, 342

A accents, 6, 7, 18, 47, 48, 49, 50, 51, 52, 53, 54, 55, 57, 58, 59, 62, 64, 66, 74, 153, 343 Adjunct, 89, 91, 92, 98, 99, 100, 101, 102, 103, 104, 105, 109, 112, 113, 130, 131, 173, 184, 185, 187, 189, 190, 191, 192, 194, 197, 198, 199, 200, 201, 203, 204, 225, 235, 237, 238, 239, 292, 293, 294, 295, 296, 297, 301, 341, 342 Adjuncts of modality, 98, 185 Adjuncts of temporality, 98 Archaic Hebrew, 12, 15, 30, 35, 36, 37, 170 alternative couplet, 11 Attribute, 111, 114, 115, 117, 119, 121, 122, 130, 132, 161, 209, 212, 213, 214, 215, 216, 220, 225, 228, 235 Amarna Letters, 29, 33

D declarative, 77, 78, 89, 90, 92, 96, 97, 98, 103, 104, 105, 147, 156, 157, 161, 162, 163, 164, 165, 168, 169, 170, 172, 176, 180, 182, 185, 191, 192, 193, 200, 201, 202, 204, 205, 270, 278, 279, 280, 281, 282, 283, 284, 286, 288, 289, 292, 293, 294, 298, 319, 334 Defamiliarisation, 45 Dependent bound clauses, 94

C Carrier, 111, 117, 119, 121, 122, 129, 132, 212, 213, 214, 215, 216, 231, 234, 235 chiasmus, 31, 38, 41, 44, 46, 62, 190, 277, 293, 297, 303 circumstantial elements, 112, 230 clause complex, 23, 77, 85, 132, 133, 134, 135, 139, 152, 153, 179, 196, 240, 241, 253, 310, 312, 320, 332, 337, 341, 343 Client, 114, 115, 130, 208, 209, 230, 238 Codex Aleppo, 48, 49 Codex Sinaiticus, 48, 49 cohortative, 65, 165, 166, 189, 200

E Early Hebrew, 13 ellipsis, 7, 11, 21, 101, 179, 189, 196, 233 Elliptical clauses, 95, 106 Embedded material, 108 enumeration, 11 Equative constructions, 93, 176

541

542

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

ergativity, 23, 128, 129, 132, 152, 153, 231, 239, 301, 307, 308, 320, 336, 337, 342 Expletives, 102, 103, 180, 191

F Finite, 89, 90, 91, 93, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 113, 130, 156, 157, 162, 163, 164, 170, 173, 174, 178, 179, 183, 184, 186, 189, 191, 192, 198, 199, 200, 203, 204, 205, 236, 237, 292, 293, 294, 295, 296, 307, 326, 342 Focus, 37, 39, 40, 41, 88 Foregrounding, 39

G gapping, 38, 39, 44, 277 Given, 18, 31, 34, 38, 40, 74, 80, 85, 86, 87, 88, 89, 94, 153, 154, 155, 216, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 299, 312, 339 Given Information, 40 Goal, 37, 111, 113, 114, 115, 129, 130, 131, 205, 206, 207, 208, 209, 211, 212, 220, 222, 223, 224, 225, 227, 228, 229, 230, 234, 235, 236, 254, 302, 304, 305

imperative, 65, 80, 89, 91, 96, 105, 106, 165, 166, 174, 188, 190, 199, 200, 203, 243, 245, 274 indicative, 77, 78, 89, 96, 98, 100, 188, 310, 331 Information, 41, 85, 88 information structure, 40, 41, 45, 86, 87, 94, 151, 153, 154, 155, 277, 280, 281, 284, 288, 340, 341, 344 insertion, 38, 277 instantiation, 22, 76, 80, 341 Internal parallelism, 11 interpersonal metafunction, 81 interpersonal Theme, 92 interrogative, 12, 77, 78, 86, 89, 90, 92, 93, 96, 98, 103, 104, 118, 119, 147, 149, 156, 157, 162, 163, 164, 171, 180, 182, 183, 193, 194, 195, 196, 198, 201, 202, 215, 247, 265, 270, 289, 298 inversion, 38, 277

J Janus parallelism, 11

L Late Hebrew, 13, 22, 35, 36, 38, 158, 169, 195, 217, 344

H

M

Hebrew poetry, 3, 4, 5, 6, 8, 9, 11, 12, 13, 14, 16, 19, 21, 22, 25, 27, 29, 30, 31, 32, 33, 34, 35, 37, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 59, 60, 61, 62, 67, 69, 70, 71, 153, 154, 176, 189, 190, 205, 207, 258, 262, 264, 273, 275, 277, 286, 291, 298, 306, 310, 319, 320, 322, 325, 326, 330, 337, 338, 339, 340, 342, 344

Marked Theme, 88, 89, 90, 157, 164, 165, 166, 174, 175, 176, 177, 178, 179, 180, 215, 278, 279, 280, 281, 282, 284, 285, 286, 289, 290, 291, 296, 297, 298, 299, 301, 333, 334, 335, 341, 342 markedness, 38, 44, 46, 163, 165, 166, 171, 205, 283, 298, 319, 320, 334, 335, 337, 338, 340 Mental clauses, 115, 118 metafunction, 76, 82, 83, 84, 92, 109, 151, 152, 240 Minor clauses, 95, 107, 172 MOOD, 23, 78, 86, 89, 95, 96, 98, 99, 100, 105, 120, 130, 131, 147, 151, 153, 155, 156, 157, 160, 163, 165, 182, 183, 184, 185, 186, 190, 191, 195, 196, 204, 205,

I ideational metafunction, 81, 109 Identified, 117, 118, 119, 120, 121, 122, 215, 216, 217, 218, 229, 274 Identifier, 117, 118, 119, 120, 121, 122, 215, 216, 217, 218, 229

INDEX OF SUBJECTS 214, 225, 289, 294, 298, 319, 334, 337, 341, 342 mood Adjunct, 98

N New, 3, 14, 40, 72, 85, 86, 87, 88, 89, 94, 151, 153, 154, 155, 232, 277, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288 New Information, 40 nominal clause, 18, 257 numerical parallelism, 11

P parallelism, 3, 5, 6, 7, 8, 17, 18, 19, 21, 28, 31, 32, 34, 38, 39, 41, 43, 44, 45, 46, 47, 54, 57, 58, 61, 62, 64, 65, 69, 70, 191, 277, 279, 283, 290, 292, 293, 298, 306, 311, 313, 318, 319, 320, 321, 322, 323, 325, 326, 327, 330, 336, 337, 338, 340, 343 participant, 83, 93, 109, 110, 111, 113, 114, 115, 116, 117, 118, 122, 123, 125, 127, 144, 152, 160, 219, 221, 222, 265, 266 pivot, 38, 46, 277, 337 poetry, 1, 2, 3, 4, 5, 6, 7, 8, 11, 13, 15, 16, 20, 21, 22, 30, 33, 34, 37, 38, 42, 45, 46, 51, 54, 55, 57, 58, 59, 60, 61, 62, 63, 67, 72, 147, 153, 187, 233, 241, 261, 263, 265, 277, 298, 319, 320, 321, 322, 323, 326, 339, 344 polar interrogative, 89, 180 pragmatics, 41, 45, 46, 340 Predicated Themes, 93 Predication, 88 Predicator, 89, 91, 99, 100, 101, 102, 103, 104, 105, 106, 108, 109, 130, 131, 156, 157, 158, 159, 160, 161, 162, 163, 169, 171, 172, 177, 179, 180, 182, 183, 184, 189, 192, 193, 196, 197, 198, 199, 200, 202, 203, 204, 208, 209, 234, 235, 236, 237, 278, 279, 280, 281, 284, 285, 286, 295, 298, 314, 342 Process types, 109, 110, 111, 122, 144, 235, 245

543

proposals, 28, 96, 98, 145, 149, 150, 185, 188, 270 propositions, 63, 96, 98, 149, 150, 185 prose, 1, 3, 4, 8, 9, 11, 13, 16, 20, 21, 22, 23, 30, 35, 40, 42, 43, 49, 50, 53, 72, 153, 180, 187, 188, 207, 240, 241, 277, 319, 320, 321, 339

Q Qumran Hodayot, 32

R rank, 49, 55, 57, 77, 78, 85, 87, 142, 252 realization, 3, 78, 79, 80, 85, 86, 97, 104, 105, 109, 136, 139, 159, 163, 173, 204, 211, 213, 214, 215, 219, 224, 226, 230, 241, 249, 275, 308, 309, 313, 320, 338, 342 Recipient, 37, 114, 115, 130, 208, 220, 228, 229, 230, 235 Relational clauses, 116, 118 repetition, 6, 11, 38, 39, 61, 277, 280, 322, 326 Residue, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 113, 130, 162, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 199, 200, 201, 202, 203, 205, 234, 235, 236, 237, 238, 239, 291, 292, 295 Rheme, 40, 87, 88, 89, 90, 91, 93, 94, 95, 109, 112, 113, 153, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 234, 235, 236, 237, 238, 239, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288 rhythm, 2, 5, 6, 7, 21, 48, 69, 85, 325, 326

S Scope, 114, 115, 123, 130, 207, 208, 305 Secondary lines, 44 Senser, 111, 112, 116, 124, 125, 129, 130, 209, 210, 211, 228, 230, 231, 306 split-couplet, 11

544

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

staircase, 39, 46, 64, 130, 136, 277, 326 Standard Hebrew, 13, 23, 153 stratification, 76, 79 structure, 6, 12, 16, 17, 20, 23, 26, 31, 32, 33, 34, 38, 39, 41, 43, 44, 45, 47, 48, 50, 51, 54, 55, 58, 59, 60, 63, 64, 65, 67, 71, 76, 77, 78, 81, 83, 85, 86, 88, 89, 91, 94, 95, 96, 98, 100, 102, 104, 107, 108, 109, 111, 112, 115, 116, 120, 122, 125, 129, 132, 135, 138, 142, 151, 152, 153, 154, 155, 157, 168, 170, 171, 174, 176, 178, 180, 181, 182, 183, 193, 200, 203, 204, 205, 209, 211, 212, 218, 227, 234, 235, 240, 254, 256, 273, 277, 278, 280, 281, 283, 284, 285, 286, 288, 290, 291, 294, 296, 297, 299, 301, 302, 303, 304, 305, 306, 307, 309, 310, 313, 319, 320, 321, 324, 327, 333, 334, 335, 336, 337, 338, 341, 342, 343, 344 stylistics, 3, 41, 45, 76 Subject, 21, 35, 37, 41, 82, 83, 89, 90, 91, 92, 94, 96, 97, 98, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 112, 113, 114, 120, 125, 130, 131, 136, 151, 153, 156, 157, 158, 159, 160, 161, 162, 163, 164, 168, 169, 177, 179, 180, 183, 184, 185, 186, 187, 189, 190, 191, 192, 193, 194, 195, 196, 197, 198, 200, 201, 203, 204, 205, 208, 209, 211, 213, 215, 216, 217, 218, 234, 235, 236, 237, 238, 239, 245, 248, 278, 279, 280, 281, 282, 283, 284, 285, 286, 291, 292, 293, 294, 295, 296, 297, 298, 299, 300, 301, 319, 321, 334, 335, 336, 337, 342, 343 swapping, 38, 277 system, 2, 6, 8, 9, 11, 36, 37, 51, 58, 68, 69, 75, 76, 77, 78, 80, 83, 85, 86, 87, 92, 96, 109, 118, 128, 134, 153, 155, 188, 205, 231, 232, 236, 242, 249, 275, 317, 332, 342 system network, 78

T terrace, 39, 46, 277, 335, 337 textual metafunction, 81 textual Theme, 92, 102

Theme, 21, 38, 40, 82, 83, 87, 88, 89, 90, 91, 92, 93, 94, 95, 98, 100, 102, 109, 112, 113, 151, 153, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 184, 186, 192, 211, 213, 215, 218, 221, 234, 235, 236, 237, 238, 239, 278, 279, 280, 281, 282, 283, 284, 285, 286, 287, 288, 289, 290, 291, 292, 293, 294, 295, 296, 298, 299, 301, 319, 333, 334, 335, 341, 342 Topic, 37, 40, 41, 87, 88 transitivity, 23, 42, 95, 109, 111, 128, 129, 130, 132, 134, 135, 152, 153, 205, 231, 236, 301, 307, 320, 336, 337, 342 Transitivity structure, 113 translation work, 9

U Ugaritic poetry, 6, 12, 21, 25, 26, 27, 31, 32, 33, 34, 47, 61, 320, 321, 323, 324, 325, 326, 333, 337, 338, 342

V verbal ellipsis, 47 vertical parallelism, 11 Vocatives, 38, 92, 102, 171, 172, 180, 342 Vulgate, 48, 49, 59WH- interrogative, 89, 90, 98, 104

W word order, 7, 30, 31, 35, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 57, 103, 105, 159, 160, 161, 163, 193, 194, 213, 321, 322, 326, 340, 341 word-pairs, 6, 7, 28, 38, 277, 312, 315, 322

INDEX OF AUTHORITIES

A

D

Abercrombie, 75, 85 Aistleitner, 190 Aitken, 323 Albright, 30, 67, 190, 325, 338 Allen, 69, 70, 75, 84 Alter, 3, 5 Andersen, 11, 160, 161 Auffret, 71, 72 Aviram, 1, 2 Avishur, 71, 338

Dahood, 27, 28, 31, 33, 46, 67, 69, 70, 170, 190, 198, 340, 343 Delekat, 33, 34 Dik, 37, 76, 88 Dotan, 49, 50, 51, 52 Dresher, 50, 51, 52, 53, 55, 154 Driver, G.R., 323

E Eaton, 51, 356 Emerton, 70 Eskhult, 11 Ewald, 9, 10, 14, 356, 368

B Berlin, 5, 16, 31, 340, 349, 350, 351, 361, 362, 367, 368 Brongers, 163, 168, 169, 170 Buth, 30, 31, 32, 33, 38, 43, 44, 154, 320, 340 Butler, 76, 80, 84, 88

F Fassberg, 188 Fensham, 29, 30 Fenton, 33, 34, 37 Firth, 75 Floor, 88, 154, 156 Fokkelman, 17, 57, 60-64 Freedman, 7, 13, 30, 61, 189, 325

C Cassuto, 28, 33, 322, 323, 327, 328, 329, 331, 333 Ceresko, 67, 71, 72 Christensen, 6, 7, 51, 72 Collins, 62 Cook, 187 Coppens, 33 Craigie, 69, 70, 322 Creach, 70 Cross, 13, 30, 61 Culley, 72

G Garr, 168, 174 Geller, 5, 14, 178 Gevirtz, 29, 32, 33, 319 Gianto, 174 Gibson, 323, 331 Ginsberg, 27 Givón, 31, 35, 36, 37, 45, 63, 153 Goldfajn, 11

545

546

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Gordon, 27, 320, 321, 323, 328, 329 Goshen-Gottstein, 49 Gray, John, 6, 11, 69, 322, 323, 330 Gray, George B., 5, 69 Greenberg, 325, 326 Greenstein, 154, 326 Gross, 35, 40, 41, 42, 45, 167, 169, 340 Gunkel, 68

H Habermann, 6 Hadas-Lebel, 14 Halliday, 20, 75, 76, 77, 78, 79, 80, 81, 83, 84, 85, 86, 88, 89, 114, 120, 128, 131, 132, 134, 145 Hatav, 11, 242, 243 Heimerdinger, 154, 242 Held, 12, 25, 26, 31, 33, 329, 330, 331, 332, 333, 340 de Hoop, 50, 58 Horwitz, 324, 325 Hrushovski, 5 Huehnergard, 344 Hurvitz, 13, 14

I Isaksson, 10, 188, 363

J Jakobson, 1, 2 Jongeling, 167, 169 Joosten, 9, 10, 14, 188 Joüon, 25, 164, 184, 188, 193, 194, 195, 196, 198, 200, 204, 232, 246, 252, 253, 255, 258, 259, 260, 261, 262, 270

K Kaddari, 17, 18 Kennedy, 70, 84 Kloos, 338 Koopmans, 58 Korpel, 6, 7, 11, 17, 56, 57, 60, 61, 63, 324 Kosmala, 325 Kraus, 67, 68

Kugel, 5, 6, 18, 31, 70, 340 Kutscher, 13, 366

L Labuschagne, 168 Lambrecht, 40, 41, 42, 43 Landy, 5, 366 Langa, 73 Launderville, 69, 71, 366 Leys, 68 Loewinger, 49 Longacre, 11, 42, 241, 242, 243 Longman III, 3 Loretz, 67, 323, 325 Lowth, 5, 44 Lugt, 17, 66, 71 Lundbom, 72 Lunn, 30, 35, 38, 39, 40, 41, 42, 43, 44, 45, 154, 277, 340, 341

M Mabie, 324 Maori, 56 March, 70 Margalit, 16, 18, 57, 58, 61, 323, 325 Martin, xiii, 78, 348, 351, 368 Mays, 68 McCann, 70 McFall, 9, 29 Meier, 39, 219 van der Merwe, 40, 256, 257 Meschonnic, 5, 368 Michel, 9, 10 Miller, C.L., 5, 70, 190, 219, 262, 265, 266, 267, 268, 270, 272 Miller, P.D., 5 Morgenstern, 338 de Moor, 4, 7, 11, 17, 56, 57, 60, 61 Mowinckel, 68, 338 Mulder, 169 Mullen, 338 Muraoka, 40, 184, 188, 232 Murphy, 67, 68

INDEX OF AUTHORITIES N Niccacci, 8, 11, 186, 241, 242, 243, 244, 245, 246, 247, 256, 257, 260, 266, 310, 330

O O’Connor, 3, 5, 62 Oesch, 56, 57 Olley, 56 del Olmo Lete, 323, 329, 331

Sivan, 323, 335, 337 Smith, 323, 328, 331, 333 Stuart, 325

T Talstra, 19, 39, 241 Tate, 69, 70 Thibault, 76, 81, 375 Tournay, 72, 73 Tropper, 190, 323, 327, 329, 332, 333 Tsevat, 15

P Pardee, 171, 323, 325, 326 Parker, 323, 325, 326, 329 Pike, 75, 85 Porte, 51 Price, 49, 50, 51, 52, 53, 55

R Rata, 12, 187, 371 Revell, 49, 56, 58, 59, 64, 154 Rosenbaum, 35, 37, 38, 39, 42, 45, 154, 172, 277, 340

S Sáenz-Badillos, 14 Schneider, 11 Schoors, 169, 170 Segert, 187, 190, 324, 325, 326 Shimasaki, 43, 154 Sievers, 6, 68, 69, 325 Siewierska, 75

547

V Vantoura, 51, 375 Vogels, 60, 375

W Waltke, 49, 50 Watson, 4, 7, 11, 31, 32, 33, 51, 60, 241, 311, 312, 313, 323, 324, 326, 327, 330, 331, 332, 333, 335, 337, 340 Watts, 72 Wellhausen, 66 Wendland, 9, 320 Wickes, 49, 53, 54, 55 Widengren, 338 Wutz, 66 Wyatt, 323, 328, 331

Y Yeivin, 49, 50, 51 Young, 324, 325

INDEX OF BIBLICAL REFERENCES

Gen 18.25, 162 Gen 18.30, 199, 200 Gen 18.32, 199 Gen 19.2, 171, 179 Gen 19.7, 199 Gen 19.18, 199 Gen 19.20, 164, 189 Gen 20.4, 171 Gen 20.9, 165 Gen 20.15, 169 Gen 21.17, 171 Gen 22.1, 173 Gen 22.7-8, 182 Gen 22.8, 172 Gen 22.11, 173 Gen 22.12, 166 Gen 23.4, 199 Gen 23.6, 171 Gen 23.11, 173 Gen 24.12, 172 Gen 24.14, 176 Gen 24.15, 168 Gen 24.16, 193 Gen 24.18, 171 Gen 24.23-24, 182 Gen 24.56, 200 Gen 24.60, 166 Gen 25.30, 161 Gen 27.4, 200 Gen 27.18, 171 Gen 27.20, 171, 172 Gen 27.25, 200 Gen 27.26, 171 Gen 27.31, 171 Gen 27.34, 171

Genesis Genesis, 36 Gen 1.1, 155, 206 Gen 2.17, 184 Gen 2.18, 184 Gen 2.19, 160 Gen 3.5, 158 Gen 3.11, 163 Gen 3.16, 177 Gen 3.22, 167 Gen 4.9, 163 Gen 4.23-24, 14 Gen 5.1, 161 Gen 6.5, 161 Gen 8.21, 185 Gen 10.12, 176 Gen 11.3, 199 Gen 11.4, 198 Gen 11.7, 198 Gen 12.18, 197 Gen 12.2, 200 Gen 12.4, 178 Gen 13.8, 199 Gen 13.9, 168 Gen 14.24, 165 Gen 16.6, 157 Gen 16.8, 172 Gen 17.15, 160 Gen 17.19, 171 Gen 18.3, 199 Gen 18.14, 162 Gen 18.15, 186 Gen 18.20, 170

549

550

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Gen 27.36, 195 Gen 27.37, 171 Gen 27.38, 171 Gen 27.43, 171 Gen 27.45, 175 Gen 27.8, 171 Gen 27.9, 200 Gen 28.13, 203 Gen 28.17, 157 Gen 29.21, 200 Gen 29.25, 164 Gen 29.4, 171, 201 Gen 29.5, 201 Gen 29.6, 203 Gen 30.21, 160 Gen 30.25, 200 Gen 30.26, 200 Gen 31.39, 176 Gen 32.28, 201 Gen 33.9, 159, 172 Gen 33.10, 199 Gen 33.17, 186 Gen 37.1, 157 Gen 37.3, 157 Gen 37.4, 174 Gen 37.15, 174 Gen 37.16, 197 Gen 37.23, 178 Gen 38.1, 160 Gen 38.6, 191 Gen 38.13, 174, 191 Gen 38.16, 200 Gen 38.18, 201 Gen 39.4, 192 Gen 40.20, 155 Gen 41.28, 176 Gen 41.43, 158 Gen 42.10, 173 Gen 42.34, 200 Gen 42.37, 175 Gen 43.20, 172 Gen 43.27, 203 Gen 44.2, 181 Gen 44.5, 170 Gen 44.18, 172 Gen 47.10, 183 Gen 47.15, 198 Gen 47.29, 199

Gen 48.21, 183 Gen 49, 14, 264 Gen 49.1, 200

Exodus Exod 1.10, 198 Exod 2.7-8, 181 Exod 2.14, 186 Exod 3.3, 200 Exod 3.4, 172 Exod 3.15, 160 Exod 4.10, 172 Exod 4.11, 164 Exod 4.13, 172 Exod 5.11, 174 Exod 6.2, 160 Exod 8.11, 158, 159 Exod 11.9, 170 Exod 15, 14, 32, 223 Exod 15.1-11, 14 Exod 15.3, 160 Exod 16.15, 177 Exod 20.8, 167 Exod 29.14, 161 Exod 34.14, 160 Exod 40.9, 187 Exod 40.33, 189

Leviticus Lev 1.10, 190 Lev 13.36, 161 Lev 18.15, 160 Lev 23.4, 193

Numbers Num 10.31, 199 Num 10.35, 14 Num 12.11, 172, 199 Num 12.12, 199 Num 12.14, 164 Num 12.2, 165 Num 13.16, 160 Num 13.18, 196 Num 13.28, 161, 170 Num 14.28, 179

INDEX OF BIBLICAL REFERENCES Num 16.7, 176 Num 17.3, 165 Num 21.16, 176 Num 22.16, 199 Num 22.37, 164 Num 23-24, 14 Num 23.7-10, 43 Num 23.12, 165 Num 23.26, 164 Num 24.11, 188 Num 24.12, 165 Num 31.15, 193

Deuteronomy Deut 1.8, 196 Deut 4.39, 178 Deut 5.1, 166 Deut 7.20, 192 Deut 8.18, 177 Deut 18.22, 176 Deut 24.18, 192 Deut 32, 14, 40, 163 Deut 32.1, 200 Deut 33, 12, 14, 29, 32, 40, 54, 56, 62, 84, 86, 163, 169, 209, 211, 234, 251, 257, 259, 264, 272, 274, 289, 290

Joshua Josh 1.3, 175 Josh 5.13, 196 Josh 7.7, 198 Josh 7.8, 172 Josh 7.19, 171 Josh 9.20, 158 Josh 10.12, 171 Josh 16.28a, 171 Josh 19.23, 171 Josh 22.20, 170

551

Judg 5.3, 171 Judg 5.10, 171 Judg 5.12, 171 Judg 5.21, 171 Judg 5.30, 170 Judg 5.31, 171 Judg 6.12, 172, 202 Judg 6.13, 172 Judg 6.14, 164 Judg 6.14-15, 181 Judg 6.15, 172 Judg 6.18, 199 Judg 6.22, 191 Judg 6.22-23, 182 Judg 6.29, 197 Judg 6.31-32, 181 Judg 7.19, 158 Judg 8.31, 160 Judg 9.7, 171 Judg 11.1, 154 Judg 11.23, 195 Judg 11.24, 165 Judg 11.36, 172 Judg 13.24, 160 Judg 14.16, 195 Judg 16.12, 172 Judg 16.20, 172 Judg 16.28, 171 Judg 16.9, 172 Judg 18.19, 196 Judg 19.6, 189 Judg 19.19, 159 Judg 20.7, 172 Judg 20.39, 173 Judg 21.3, 171

Ruth Ruth 2.14, 199 Ruth 3.1, 164 Ruth 3.11, 171

Judges Judg 3.12, 189 Judg 3.19, 172, 179 Judg 4.14, 168 Judg 4.6, 195 Judg 5, 3, 4, 14

1 Samuel 1 Sam 1.15, 173 1 Sam 1.26, 172 1 Sam 2.1-10, 14 1 Sam 2.24, 173

552

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

1 Sam 2.27-28, 158 1 Sam 3.6, 190 1 Sam 3.9, 171 1 Sam 3.10, 172, 202 1 Sam 3.17, 199 1 Sam 4.9, 171 1 Sam 9.3, 198 1 Sam 9.6, 173 1 Sam 9.21, 170 1 Sam 10.1, 168 1 Sam 11.6, 159 1 Sam 13.19, 175 1 Sam 14.41, 171 1 Sam 14.44, 172 1 Sam 16.4, 194 1 Sam 17.55, 171 1 Sam 17.58, 171 1 Sam 19.24, 157 1 Sam 20.3, 202 1 Sam 20.37, 168 1 Sam 21.10, 176 1 Sam 21.12, 170 1 Sam 24.12, 171 1 Sam 24.17, 171 1 Sam 25.25, 199 1 Sam 25.26, 172 1 Sam 27.5, 200 1 Sam 30.13, 197

2 Samuel 2 Sam 1.18, 194 2 Sam 1.24, 171 2 Sam 1.25, 194 2 Sam 1.26, 172 2 Sam 1.27, 194 2 Sam 3.38, 170 2 Sam 4.11, 163 2 Sam 7.18, 171 2 Sam 7.19, 172 2 Sam 7.20, 172 2 Sam 7.22, 172 2 Sam 7.24, 172 2 Sam 7.25, 171 2 Sam 7.27, 172 2 Sam 7.28, 172 2 Sam 7.29, 172 2 Sam 9.13, 185

2 Sam 11.10, 170 2 Sam 11.11, 195 2 Sam 11.3, 194 2 Sam 12.19, 185 2 Sam 13.25, 199 2 Sam 14.4, 171 2 Sam 14.18, 199 2 Sam 14.19, 172 2 Sam 15.4, 158, 194 2 Sam 16.7, 171 2 Sam 18.23, 203 2 Sam 20.9, 203 2 Sam 22, 3 2 Sam 22.39, 175 2 Sam 23.5, 170 2 Sam 24.14, 167

1 Kings 1 Kgs 1.13, 171 1 Kgs 1.17, 172 1 Kgs 1.18, 172 1 Kgs 1.20, 172 1 Kgs 1.24, 172 1 Kgs 1.43, 169 1 Kgs 1.45, 176 1 Kgs 2.13, 194, 203 1 Kgs 3.17, 172 1 Kgs 3.26, 172 1 Kgs 7.21, 160 1 Kgs 8.23, 172 1 Kgs 8.25, 26, 171 1 Kgs 8.28, 172 1 Kgs 8.53, 172 1 Kgs 9.25, 158 1 Kgs 12.16, 172 1 Kgs 12.16b, 171 1 Kgs 13.7, 200 1 Kgs 13.30, 194 1 Kgs 14.19, 194 1 Kgs 14.30, 172 1 Kgs 18.13, 164 1 Kgs 18.16, 171 1 Kgs 18.17, 163 1 Kgs 18.24, 176 1 Kgs 18.36, 171 1 Kgs 18.41, 194 1 Kgs 19.4, 172

INDEX OF BIBLICAL REFERENCES 2 Chronicles

1 Kgs 19.9, 171 1 Kgs 19.20, 200 1 Kgs 20.4, 172 1 Kgs 21.20, 171 1 Kgs 22.15, 196

2 Chr 28.19, 158

Ezrah 2 Kings

2 Kgs, 36 2 Kgs 1.9, 172 2 Kgs 1.11, 172 2 Kgs 1.13, 172 2 Kgs 2.12, 172 2 Kgs 2.23, 171 2 Kgs 3.11, 187 2 Kgs 3.23, 172 2 Kgs 4.2, 162 2 Kgs 4.18, 172 2 Kgs 4.19, 179, 203 2 Kgs 4.26, 203 2 Kgs 5.21-22, 202 2 Kgs 6.5, 172 2 Kgs 6.12, 173 2 Kgs 6.21-22, 181 2 Kgs 6.27, 196 2 Kgs 9.5, 172 2 Kgs 9.11, 194, 203 2 Kgs 9.17, 203 2 Kgs 9.18, 203 2 Kgs 9.22, 203 2 Kgs 9.31, 203 2 Kgs 11.14, 172 2 Kgs 13.14, 172 2 Kgs 15.11, 194 2 Kgs 19.15, 172 2 Kgs 19.16, 171 2 Kgs 19.17, 172 2 Kgs 19.19, 171, 172 2 Kgs 19.25, 164 2 Kgs 21.17, 164 2 Kgs 23.34, 160 2 Kgs 24.17, 160

1 Chronicles 1 Chr 16.36, 158 1 Chr 5.20, 158

553

Ezrah 1.3, 176

Nehemiah Neh 7.3, 158 Neh 8. 8, 158 Neh 9.13, 158 Neh 9.7, 160 Neh 9.8, 158

Esther Esther, 36 Esth 2.3, 158 Esth 3.13, 158 Esth 6.9, 158 Esth 8.8, 158 Esth 9.12, 158 Esth 9.16-18, 158 Esth 9.6, 158

Job Job, 4, 12, 14, 43, 49, 54, 59, 187, 195, 196, 234, 267, 272 Job 1.1, 160 Job 3-14, 154 Job 3.3, 12 Job 4.3, 12, 82, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 100, 102, 103, 104, 105, 106, 108, 115, 116, 126, 185, 231, 262, 265, 266, 268, 273 Job 5.2, 170 Job 6.25, 12 Job 7.2, 174 Job 9.2, 12 Job 15.18, 12 Job 15.7, 12 Job 15.9, 12 Job 16.3, 196 Job 21.29, 12 Job 22.12, 170 Job 23.12, 12

554

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Job 28.1, 170 Job 33.13, 12 Job 33.13b, 12 Job 35.6c, 12 Job 37.15, 12 Job 38.17, 12 Job 38.22, 12, 14, 29, 32, 45, 54, 86, 119, 195, 220, 223, 227, 228, 233, 234, 238, 245, 254, 261, 262, 264, 267, 270, 274, 287, 288, 289, 317, 329 Job 38.28, 196 Job 38.31, 196

Psalms Psalms, 3, 4, 9, 10, 11, 12, 13, 14, 15, 16, 17, 20, 22, 25, 27, 28, 29, 30, 31, 32, 42, 47, 48, 49, 52, 59, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 154, 170, 190, 275, 289, 290, 299, 300, 301, 307, 310, 318, 319, 338, 339, 343, 344 Ps 1, 43 Ps 2.1, 14, 25, 27, 29, 30, 31, 32, 35, 56, 85, 88, 109, 136, 144, 183, 249, 290, 299, 307, 308, 309, 316 Ps 2.1A, 298, 301 Ps 2.1B, 298, 301, 341 Ps 3.7, 31, 72, 290, 307, 308, 309, 317, 341 Ps 3.7A, 298, 299 Ps 3.7B, 301 Ps 5.6, 31, 220, 290, 308, 309, 316 Ps 5.6A, 298, 301, 308 Ps 5.6B, 300, 307 Ps 6.10, 8, 29, 290, 299, 307, 308, 309, 316 Ps 6.10A, 300 Ps 6.10B, 300, 341 Ps 7, 310 Ps 7.4-6, 10 Ps 7.8, 190 Ps 7.14, 29, 290, 307, 308, 309, 316 Ps 7.14A, 298, 300 Ps 7.14AB, 341 Ps 7.14B, 300 Ps 8.7, 29, 290, 307, 308, 309, 316 Ps 8.7A, 300

Ps 8.7B, 300 Ps 9.8, 29, 290, 299, 307, 308, 309, 316 Ps 9.8A, 300 Ps 9.8B, 300 Ps 10.14A, 307 Ps 10.14b, 290, 308, 309, 316 Ps 10.14bA, 300 Ps 10.14bB, 300 Ps 11.3, 206, 230, 290, 298, 299, 301, 307, 308, 309, 317 Ps 11.3A, 301 Ps 11.3B, 298, 300 Ps 13.6a, 290, 299, 307, 308, 309, 316 Ps 13.6aA, 299 Ps 13.6aB, 300, 341 Ps 15.5, 65 Ps 17, 14 Ps 17.3, 10, 65 Ps 17.4, 65 Ps 17.11, 228, 290, 307, 308, 310, 316, 341 Ps 17.11A, 301 Ps 17.11B, 301 Ps 18.5, 264, 290, 299, 307, 308, 309, 316 Ps 18.5A, 301 Ps 18.5B, 301, 341 Ps 18.29-30, 170 Ps 18.37, 290, 307, 308, 309, 316 Ps 18.37A, 300 Ps 18.37B, 298, 301 Ps 18.41, 290, 299, 308, 310, 316 Ps 18.41A, 301, 308 Ps 18.41B, 299, 308 Ps 19.4, 65 Ps 20.8, 65 Ps 21.12, 10 Ps 23.4, 65 Ps 23.6, 290, 299, 307, 308, 309, 316, 341 Ps 23.6A, 301 Ps 23.6B, 299, 308 Ps 24, 14, 15, 31, 40, 43, 54, 58, 86, 154, 168, 194, 195, 204, 208, 220, 229, 234, 239, 245, 256, 257, 264, 267, 270, 271, 274, 307, 325, 326 Ps 26, 26, 301, 302, 306 Ps 26.4-5, 26 Ps 26.4, 29, 32, 258, 290, 308, 309, 316 Ps 26.4A, 299

INDEX OF BIBLICAL REFERENCES Ps 26.4B, 299, 341 Ps 26.5, 29, 264, 290, 307, 308, 309, 316 Ps 26.5A, 299 Ps 26.5B, 298, 299, 341 Ps 26.12, 29, 290, 299, 308, 316 Ps 26.12A, 300 Ps 26.12B, 219, 299 Ps 27.3b, 10 Ps 27.4a, 290, 307, 308, 309, 316 Ps 27.4aA, 299 Ps 27.4aB, 299 Ps 27.10, 261, 290, 299, 308, 309, 316 Ps 27.10A, 301 Ps 27.10B, 300 Ps 28.1c, 290, 307, 308, 309, 317 Ps 28.1cA, 300 Ps 28.1cB, 299 Ps 28.5, 65 Ps 29, 13, 14, 26, 30, 32, 41, 53, 54, 56, 83, 93, 169, 178, 195, 203, 204, 211, 217, 224, 234, 237, 241, 242, 261, 264, 267, 268, 269, 270, 271, 290, 338 Ps 30.7, 31, 290, 299, 307, 308, 309, 316 Ps 30.7A, 299, 308 Ps 30.7B, 298, 299, 308 Ps 31.6, 31, 290, 307, 308, 310, 316 Ps 31.6A, 299 Ps 31.6B, 300 Ps 33.1, 190 Ps 33.21, 290, 307, 308, 310, 316, 341 Ps 33.21A, 300 Ps 33.21B, 300 Ps 37.11, 290, 299, 308, 309, 316 Ps 37.11A, 301, 307, 308 Ps 37.11B, 301, 307, 308 Ps 37.21, 10 Ps 37.23, 290, 299, 308, 309, 317, 341 Ps 37.23A, 300, 307, 308 Ps 37.23B, 300, 307, 308 Ps 38.5, 290, 299, 301, 307, 308, 309, 316 Ps 38.5aA, 308 Ps 38.5aB, 308 Ps 38.12, 26, 29, 32, 41, 290, 299, 301, 308, 309, 316 Ps 38.12A, 301 Ps 38.12B, 301 Ps 38.16, 234, 290, 299, 308, 310, 316 Ps 38.16A, 299

555

Ps 38.16B, 300 Ps 40.6, 10 Ps 41.4, 290, 299, 301, 307, 308, 309, 316 Ps 41.4A, 300 Ps 41.4B, 300 Ps 41.7, 10 Ps 44.11, 290, 299, 307, 308, 309, 316 Ps 44.11A, 300, 308 Ps 44.11B, 301, 308, 341 Ps 44.12, 290, 307, 308, 309, 316 Ps 44.12A, 300 Ps 44.12B, 300, 341 Ps 44.13, 290, 307, 308, 310, 316 Ps 44.13A, 300, 308 Ps 44.13B, 298, 300, 308 Ps 44.21-22, 10 Ps 47.5, 290, 307, 308, 309, 316 Ps 47.5A, 300 Ps 47.5B, 300, 341 Ps 48, 338 Ps 49.9, 290, 308, 309, 316 Ps 49.9A, 300, 308 Ps 49.9B, 300, 308 Ps 49.15a, 290, 299, 307, 308, 309, 316 Ps 49.15aA, 301, 308 Ps 49.15aB, 300, 308 Ps 49.16, 170 Ps 50.3, 299, 307 Ps 50.3b, 290, 308, 310, 316 Ps 50.3bA, 300 Ps 50.3bB, 300 Ps 50.19, 29, 290, 299, 307, 308, 310, 316 Ps 50.19A, 300 Ps 50.19B, 301 Ps 51.8, 290, 307, 308, 309, 316 Ps 51.8A, 300, 308 Ps 51.8B, 300, 308 Ps 55.5, 29, 31, 32, 290, 299, 301, 307, 308, 316 Ps 55.5A, 300 Ps 55.5B, 301 Ps 58.9, 31, 290, 308, 309, 316, 341 Ps 58.9A, 300, 307 Ps 58.9B, 298, 301, 307 Ps 59.4, 70 Ps 60.11, 290, 307, 308, 309, 316 Ps 60.11A, 298, 300 Ps 60.11B, 298, 300

556

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Ps 60.12, 290, 299, 307, 308, 316 Ps 60.12A, 298, 300 Ps 60.12B, 298, 300 Ps 62.5a, 290, 307, 308, 310, 316 Ps 62.5aA, 301 Ps 62.5aB, 301 Ps 63.7, 10, 29, 290, 307, 308, 309, 316 Ps 63.7A, 299 Ps 63.7B, 299, 341 Ps 63.8, 290, 308, 309, 316 Ps 63.8A, 300, 308 Ps 63.8B, 299, 308, 341 Ps 64, 338 Ps 65.4, 290, 299, 307, 308, 309, 316 Ps 65.4A, 301 Ps 65.4B, 300 Ps 65.12, 290, 299, 307, 308, 310, 316 Ps 65.12A, 300 Ps 65.12B, 301, 341 Ps 66.18, 290, 307, 308, 309, 317 Ps 66.18A, 299 Ps 66.18B, 298, 300 Ps 67.7, 290, 299, 307, 308, 309, 316 Ps 67.7A, 300 Ps 67.7B, 300 Ps 68, 14, 60 Ps 68.11, 290, 299, 307, 308, 309, 316 Ps 68.11A, 300 Ps 68.11B, 300 Ps 68.36, 170 Ps 69.33, 10 Ps 71.6, 308 Ps 71.17, 29, 290, 308, 309, 316 Ps 71.17A, 300, 307, 308 Ps 71.17B, 299, 308 Ps 71.23, 170 Ps 73.3, 29, 31, 290, 307, 308, 309, 316 Ps 73.3A, 299 Ps 73.3B, 299, 341 Ps 73.6, 290, 307, 309, 316 Ps 73.6A, 300 Ps 73.6B, 300 Ps 73.9, 29, 290, 299, 307, 308, 309, 316 Ps 73.9A, 301 Ps 73.9B, 301 Ps 73.14, 307 Ps 73.18, 29, 290, 307, 308, 309, 316 Ps 73.18A, 300

Ps 73.18B, 300, 341 Ps 73.21-22, 10 Ps 73.27, 290, 299, 307, 308, 309, 316 Ps 73.27A, 301, 308, 341 Ps 73.27B, 300, 308 Ps 74.1A, 298 Ps 74.1b, 290, 298, 308, 309, 316 Ps 74.1bA, 300 Ps 74.1bB, 300 Ps 74.14, 29, 290, 299, 308, 309, 316 Ps 74.14A, 300 Ps 74.14B, 300 Ps 76.11, 170 Ps 77, 14, 68, 90 Ps 77.12, 170 Ps 78, 3 Ps 78.20, 307, 309 Ps 78.20b, 290, 308, 316 Ps 78.20bA, 298, 300 Ps 78.20bB, 300 Ps 78.64, 29, 41, 290, 299, 308, 309, 316 Ps 78.64A, 301, 307 Ps 78.64B, 298, 301, 308 Ps 81.7, 29, 41, 290, 299, 307, 308, 309, 316 Ps 81.7A, 299, 308 Ps 81.7B, 301, 308, 341 Ps 82, 338 Ps 83.3, 29, 290, 299, 301, 308, 310, 316 Ps 83.3A, 301, 308 Ps 83.3B, 301, 308 Ps 83.6, 290, 308, 309, 316 Ps 83.6A, 301 Ps 83.6B, 301, 341 Ps 85.9, 170 Ps 85.12, 290, 299, 301, 308, 309, 316 Ps 85.12A, 300, 308 Ps 85.12B, 300, 307 Ps 87.4-5, 70 Ps 87.6, 290, 299, 301, 308, 309, 316 Ps 87.6A, 300, 307, 308 Ps 87.6B, 300, 308 Ps 88.14, 290, 299, 308, 309, 316 Ps 88.14A, 299, 308 Ps 88.14B, 300, 307, 308 Ps 89, 8, 14, 29, 31, 54, 85, 86, 299, 308, 316 Ps 89.13, 290, 309, 316

INDEX OF BIBLICAL REFERENCES Ps 89.13A, 300, 307 Ps 89.13B, 301, 308 Ps 89.36, 290, 308, 309, 316 Ps 89.36A, 299 Ps 89.36B, 299 Ps 89.44, 290, 307, 308, 310 Ps 89.44A, 300 Ps 89.44B, 298, 300 Ps 90.7, 170 Ps 92.5, 290, 308, 309, 316 Ps 92.5A, 300, 308 Ps 92.5B, 299, 308, 341 Ps 93.3, 26 Ps 94.7, 70 Ps 95.9, 70 Ps 95.11, 290, 308, 309, 316 Ps 95.11A, 299, 308 Ps 95.11B, 298, 301, 307 Ps 101.3a, 341 Ps 101.3b, 290, 298, 308, 309, 316 Ps 101.3bA, 299, 307 Ps 101.3bB, 300, 307 Ps 102.11, 170 Ps 102.15, 29, 290, 307, 308, 316 Ps 102.15A, 301 Ps 102.15B, 301 Ps 102.15b, 341 Ps 103, 14, 43 Ps 104.5, 290, 307, 308, 309, 316 Ps 104.5A, 300, 308 Ps 104.5B, 298, 300, 308 Ps 104.6, 290, 307, 308, 310, 316 Ps 104.6A, 308 Ps 104.6B, 301, 308 Ps 104.13, 170 Ps 104.16, 290, 308, 309, 316 Ps 104.16A, 301, 308 Ps 104.16B, 301, 307, 308 Ps 104.26, 290, 299, 301, 307, 308, 309, 316 Ps 104.26A, 301 Ps 104.26B, 300 Ps 105, 3 Ps 106, 3 Ps 107.2, 290, 308, 309, 317 Ps 107.2A, 301, 308 Ps 107.2B, 301, 307, 308 Ps 108.11, 290, 307, 308, 309, 316

557

Ps 108.11A, 298, 300 Ps 108.11B, 298, 300 Ps 108.12, 290, 308, 316 Ps 108.12A, 298, 300 Ps 108.12B, 298, 300 Ps 110, 338 Ps 111, 62 Ps 111.5, 29, 290, 308, 309, 316 Ps 111.5A, 300, 307 Ps 111.5B, 300, 307 Ps 112, 62 Ps 114, 14, 96, 109, 329, 332 Ps 116, 14, 29, 44, 54, 193, 270 Ps 116.2, 290, 308, 316 Ps 116.2A, 300, 308 Ps 116.2B, 299, 308 Ps 116.16, 170 Ps 118.10, 29, 170, 290, 299, 307, 308, 310, 316, 341 Ps 118.10-12, 170 Ps 118.10A, 301 Ps 118.10B, 299 Ps 119, 191 Ps 119.11, 290, 307, 308, 309, 317 Ps 119.11A, 299 Ps 119.11B, 298, 299 Ps 119.69, 290, 299, 308, 309, 316 Ps 119.69A, 301, 307, 308 Ps 119.69B, 299, 307, 308 Ps 119.82, 290, 307, 308, 309, 316 Ps 119.82A, 300, 308 Ps 119.82B, 300, 308 Ps 119.93, 290, 308, 309, 317 Ps 119.93A, 299, 307, 308 Ps 119.93B, 300, 307, 308 Ps 119.95, 290, 307, 308, 316 Ps 119.95A, 301 Ps 119.95B, 299 Ps 119.100, 290, 307, 308, 310, 317, 341 Ps 119.100A, 299 Ps 119.100B, 299 Ps 119.101, 290, 307, 308, 309, 317 Ps 119.101A, 299 Ps 119.101B, 299 Ps 119.104, 290, 307, 308, 309, 317 Ps 119.104A, 299 Ps 119.104B, 299 Ps 119.121, 290, 307, 308, 316

558

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

Ps 119.121A, 299 Ps 119.121B, 300, 310 Ps 119.126, 190 Ps 120.7, 10 Ps 120.7, 170 Ps 126.2b, 290, 308, 309, 316 Ps 126.2bA, 301, 308 Ps 126.2bB, 298, 300, 307 Ps 127.1, 10 Ps 127.1b, 290, 299, 307, 308, 309, 317 Ps 127.1bA, 298, 300 Ps 127.1bB, 300, 341 Ps 127.1c, 290, 299, 308, 309, 317 Ps 127.1cA, 298, 300 Ps 127.1cB, 300, 341 Ps 128.2, 170 Ps 132.11, 290 Ps 132.11a, 308, 316, 341 Ps 132.11aA, 300 Ps 132.11aB, 298, 300 Ps 132.17, 29, 290, 307, 308, 310, 316 Ps 132.17A, 299, 341 Ps 132.17B, 299 Ps 135, 12, 14, 69, 247 Ps 135.5, 171 Ps 139.8-11, 10 Ps 141.5, 10, 171 Ps 139.13, 29, 290, 299, 307, 308, 309, 316 Ps 142.4, 307 Ps 142.4b, 290, 308, 309, 317 Ps 142.4bA, 300, 308, 341 Ps 142.4bB, 301, 308

Proverbs Prov 1-9, 43, 154 Prov 3-14, 43 Prov 11.7, 26 Prov 26.19, 170 Prov 30.1, 170

Song of Songs Song of Songs, 4, 36, 43, 154 Song of Songs 1, 43

Ecclesiastes Ecclesiastes, 36 Eccl 2.19, 196 Eccl 7.7, 170 Eccl 8.9, 158 Eccl 9.11, 158

Isaiah Isaiah 40-55, 37, 38, 39, 154, 172 Is 1.29, 170 Is 2.6, 170 Is 26.9, 174 Is 3.1, 170 Is 5.19, 200 Is 7.9, 170 Is 8.6, 158 Is 8.19, 170 Is 15.1, 170 Is 15.9, 170 Is 21.12, 174 Is 30.18, 174 Is 32.12, 170 Is 37.19, 158 Is 40-66, 43 Is 41.16, 39 Is 44.9a, 39 Is 45.16, 170 Is 46.4, 39 Is 48.13, 39 Is 49.14-15, 39 Is 51.13, 39 Is 58.2, 174 Is 60.9, 174 Is 60.16, 26

Jeremiah Jer 3.1, 158 Jer 7.18, 158 Jer 13.16, 158, 262 Jer 14.5, 42, 158, 244 Jer 19.13, 158 Jer 22.14, 158 Jer 32.33, 158 Jer 32.44, 158 Jer 33.3, 200 Jer 36.23, 158

INDEX OF BIBLICAL REFERENCES

559 Jonah

Jer 37.21, 158 Jer. 38.15, 170

Jonah 4.2, 170

Lamentations Lamentations, 36 Lam 3.22, 170

Micah Mic 1.5, 170

Habakkuk

Ezekiel Hab 2.20, 178 Hab 3, 14

Ezek 23.36, 158 Ezek 24.25, 170 Ezek 38.14, 170 Ezek 39.8, 176 Ezek. 23.47, 158

Zephaniah Daniel

Zep 1.7, 178 Zep 2.17, 178

Haggai

Dan 9.5, 158 Dan 9.11, 158

Hag 1.6, 158

Joel Joel 2, 3

Amos Amos 3.7, 170 Amos 4.5, 158 Amos 5.20, 170 Amos 7.4, 26, 32

Ps 139.13A, 300 Ps 139.13B, 300 Ps 139.20, 290, 308, 309, 316 Ps 139.20A, 301 Ps 139.20B, 301 Ps 139.4, 171 Ps 140.3, 29, 290, 307, 308, 309, 316

Zechariah Zech 3.4, 158 Zech 7.5, 158 Zech 12.10, 158

Malachi Mal 1.8, 196 Mal 3.19, 176 Mal 3.22, 176 Mal 3.7, 200 Ps 140.3A, 301 Ps 140.3B, 301

INDEX OF UGARITIC REFERENCES KTU 1.1-6, 324 KTU 1.2 iv 32-34, 331 KTU 1.3 ii 3-5, 332, 333 KTU 1.3 ii 20-21, 327, 330, 331 KTU 1.3 ii 38-41, 329 KTU 1.3 ii 40-41, 329, 331, 332, 333 KTU 1.3 iii 34-35, 330, 333 KTU 1.3 iii 38-45, 325 KTU 1.3 iv 43-44, 329, 332 KTU 1.4 i 1-23, 327 KTU 1.4 ii 17-20, 330 KTU 1.4 ii 43-45, 332 KTU 1.4 iii 13-16, 327 KTU 1.4 iii 14-16, 328, 329, 331 KTU 1.4 iii 15-16, 333 KTU 1.4 iv 8-9, 328, 329 KTU 1.4 v 15-17, 332 KTU 1.4 v 25-26, 328, 331 KTU 1.4 v 26-27, 331, 333 KTU 1.4 vi 38-40, 329, 331, 332, 333 KTU 1.4 vi 47-48, 327, 328 KTU 1.4 vi 47-54, 331 KTU 1.4 vii 21-22, 328, 331 KTU 1.4 vii 23-25, 332, 333 KTU 1.5 i 16-18, 328, 329, 333 KTU 1.5 i 16-17, 327, 329 KTU 1.5 i 17, 331 KTU 1.5 ii 6-7, 330, 333 KTU 1.5 ii 7, 331 KTU 1.5 v 19-21, 332 KTU 1.5 vi 12-13, 332 KTU 1.6 vi 30-31, 330, 331, 333 KTU 1.10 iii 7-8, 332 KTU 1.13, 324 KTU 1.14 ii 39-41, 332 KTU 1.14 ii 32-34, 332 KTU 1.14 ii 43-45, 330, 331, 333

KTU 1.14 iv 21-23, 332 KTU 1.16 vi 20-21, 328 KTU 1.17-19, 324 KTU 1.19 ii 8-9, 332 KTU 1.19 ii 45-49, 330 KTU 1.19 iii 6-7, 327, 328, 329 KTU 1.19 iii 8-9, 331, 333 KTU 1.19 iii 20-21, 327, 329 KTU 1.19 iii 22-23, 332 KTU 1.19 iii 34-35, 329 KTU 1.19 iii 53-54, 332 KTU 1.19 iv 58-59, 330, 332, 333 KTU 1.20, 256, 324, 331 KTU 1.20 ii 2-3, 332 KTU 1.20 ii 3, 331 KTU 1.21, 324 KTU 1.22 ii 22, 331 KTU 1.23, 324 33-35, 330, 331 KTU 1.24, 324, 338 30-31, 331 30-32, 331 KTU 1.40, 324 KTU 1.43, 324 KTU 1.65, 338 KTU 1.86, 324 KTU 1.100, 324, 325 67-68, 331, 332, 333 KTU 1.101, 324, 338 KTU 1.103, 324 KTU 1.104, 324 KTU 1.105, 324 KTU 1.108, 338 KTU 1.111, 324 KTU 1.114 11-12, 332 14-15, 332

561

562

THE QYYQ VERBAL S EQUENCE IN S EMITIC COUPLETS

KTU 1.114 17-18, 331, 333 19-20, 332 KTU 1.119, 338 37, 332

KTU 101, 338 KTU 108, 338