The Religious History of the Roman Empire: The Republican Centuries (Oxford Readings in Classical Studies) 0199644063, 9780199644063

Table of contents :
Cover
The Religious History of the Roman Empire: The Republican Centuries
Copyright
Dedication
Acknowledgements
Contents
List of Abbreviations
List of Illustrations
List of Contributors
Translators
Introduction
I: Early Rome
1: Dead Dogs and Rattles: Time, Space, and Ritual Sacrifice in Iron Age Latium
INTRODUCTION
THEORETICAL APPROACHES
New Archaeology and the Identification of Cult Places
Understanding Ritual
THE EVIDENCE
Burials at Osteria dell’Osa
Sacrifice
Votive Deposit
Votive Deposits at Rome
ANALYSIS
The Robigalia and Other Dog Sacrifices
Knives
Rattles
TOWARDS AN INTERPRETATION
AFTERWORD
BIBLIOGRAPHY
2: The Priest and the Magistrate: Reflections on the Priesthoods and Public Law at the End of the Republic
TO WHAT ORDER DID THE ROMAN PRIESTS BELONG?
A. Pontifical College
B. Augurs
C. X(V)uiri s.f.
D. Epulones
E. Curio maximus
F. Luperci
A. Pontifical college
B. Augurs
C. X(V)uiri s.f.
D. Salii
PRIESTLY RECRUITMENT
FROM REPUBLIC TO PRINCIPATE
AFTERWORD
BIBLIOGRAPHY
II: Republican Practices and Ideas
3: Acca Larentia Gains a Son: Myths and Priesthood at Rome
THE MYTHOLOGY OF ROMAN PRIESTHOOD
PRIESTS AT ROME: ACTIVITY, FUNCTION, AND HISTORY
SOME PROBLEMS OF METHOD
NUMA MEETS EGERIA: THE PROBLEM RAISED
THE SALII AND THE ARVAL BRETHREN: THE AMBIGUITY OF ORIGIN
ATTUS NAEVIUS: THE ARCHETYPAL AUGUR
THE MYTHOLOGY OF PRIESTHOOD: FUNCTIONS AND IMPLICATIONS
THE SCEPTICAL APPROACH: THE DISJUNCTION ERODED
THE PARTISAN MYTHOLOGY
PRIESTLY MYTHOLOGY AS DEBATE
BIBLIOGRAPHY
4: Religion and War: On the Relationship of a Society’s Religious and Political Systems
INTRODUCTION
‘HOLY WAR’
WAR, CAUSES, INFLUENCES, AND MOTIVES
THE RELIGIOUS ELEMENTS OF WAR: ROME
Preparation
The Military Expedition (Kriegzug)
The Return
RITUALS WITH OR WITHOUT SPECTATORS: FOR WHOM OR WHAT IS ‘STATE RELIGION’?
WHAT SORT OF POLITICS? WHAT SORT OF RELIGION?
CONCLUSION
AFTERWORD
BIBLIOGRAPHY
5: Not the One nor the Many; A Pragmatic Approach to Religious Behaviour in a Polytheistic Society: The Example of Rome
THE LIMITS OF DESCRIPTION: POLYTHEISM IN ROME
PANTHEON OR PANTHEA? LOCAL RELIGION IN THE CITY OF ROME
PLURALITY AND UNITY
THE ETHNOLOGIST’S GAZE
AFTERWORD
BIBLIOGRAPHY
6: Action and Ritual in Roman Historians: Or How Horatius Held the Door-Post
INTRODUCTION
THE CAPITOLINE DEDICATION
WHO HELD THE DOOR-POST?
THE CHARACTER OF LIVY’S REPORT
MYTHICAL PARALLELS
CONCLUSION
AFTERWORD
BIBLIOGRAPHY
7: Rites and Practices of Warfare in Italy between Romans and Samnites: Going under the Yoke and the Samnite Legio Linteata
SENDING UNDER THE YOKE
THE LEGIO LINTEATA
BIBLIOGRAPHY
III: Sacrifice
8: Festus and the Role of Women in Roman Religion
FESTUS, PLUTARCH, AND FEMALE SACRIFICIAL ACTIVITIES
WINE, WOMEN, AND THE GODS
GREEKS AND WOMEN: THE CULT OF CERES AT ROME
SOME (MORE CONSTRUCTIVE) CONCLUSIONS
AFTERWORD
BIBLIOGRAPHY
9: Interpreting Sacrificial Ritual in Roman Poetry: Disciplines and Their Models
DISCIPLINES AND MODELS
The Need for Models
Models of Sacrifice
Models of Literature
VIRGIL’S GEORGICS
Walter Burkert in the Georgics
Patterns of Sacrifice in the Georgics
OVID’S FASTI
The Two Faces of Ceres
The One Face of Ceres
The Specificities of Generic Preference
Resisting Holism
BIBLIOGRAPHY
IV: Rome and Italy
10: The Ritual of Centuriation: Frontinus, Hyginus Gromaticus, and Centuriation
FRONTINUS, HYGINUS GROMATICUS, AND CENTURIATION
CENTURIATION AND AUGURY
MAGISTRATES, RITUAL, AND THE CREATION OF BOUNDARIES
FRONTINUS AND HYGINUS GROMATICUS ON LIMITES
FRONTINUS, HYGINUS GROMATICUS, AND ANTIQUARIANISM
BIBLIOGRAPHY
11: The Lucus Pisaurensis and the Romanization of the Ager Gallicus
AFTERWORD
BIBLIOGRAPHY
12: Rome and the Great Places of Worship in Italy
BIBLIOGRAPHY
V: Late Republican Transformations
13: Cicero and Divination: The Formation of a Latin Discourse
PHILOSOPHY AT ROME: THE PLACE OF CICERO
Philosophy before Cicero
Cicero’s Innovation
THE PROBLEMS OF CULTURAL INTEGRATION: CICERO’S SUSPENSION OF JUDGEMENT
The Position of Priest and Philosopher
Dialogue Form and the Suspension of Judgement
The Choice of Interlocutors and Their Roles
DE DIVINATIONE II: THE CASE AGAINST SCEPTICISM
BIBLIOGRAPHY
14: The Peace of Augustus, the Equinox, and the Centre of the World
AFTERWORD
BIBLIOGRAPHY
Details of Original Publication
Suggestions for Further Reading
Index

Citation preview

THE RELIGIOUS HISTORY OF THE ROMAN EMPIRE The series provides students and scholars with a representative selection of the best and most influential articles on a particular author, work, or subject. No single school or style of approach is privileged: the aim is to offer a broad overview of scholarship, to cover a wide variety of topics, and to illustrate a diversity of critical methods. The collections are particularly valuable for their inclusion of many important essays which are normally difficult to obtain and for the first-ever translations of some of the pieces. Many articles are thoroughly revised and updated by their authors or are provided with addenda taking account of recent work. Each volume includes an authoritative and wide-ranging introduction by the editor surveying the scholarly tradition and considering alternative approaches. This pulls the individual articles together, setting all the pieces included in their historical and cultural contexts and exploring significant connections between them from the perspective of contemporary scholarship. All foreign languages (including Greek and Latin) are translated to make the texts easily accessible to those without detailed linguistic knowledge.

OXFORD READINGS IN CLASSICAL ST UDI ES Sallust Edited by William W. Batstone and Andrew Feldherr Greek Lyric Edited by Ian Rutherford The Epistles of Pliny Edited by Roy Gibson and Christopher Whitton Flavian Epic Edited by Antony Augoustakis The Roman Historical Tradition Regal and Republican Rome Edited by James H. Richardson and Federico Santangelo Sport in the Greek and Roman Worlds: Volume 1 Early Greece, the Olympics, and Contests Edited by Thomas F. Scanlon Sport in the Greek and Roman Worlds: Volume 2 Greek Athletic Identities and Roman Sports and Spectacle Edited by Thomas F. Scanlon Herodotus: Volume 1 Herodotus and the Narrative of the Past Edited by Rosaria Vignolo Munson Herodotus: Volume 2 Herodotus and the World Edited by Rosaria Vignolo Munson Propertius Edited by Ellen Greene and Tara S. Welch Latin Panegyric Edited by Roger Rees Tacitus Edited by Rhiannon Ash

The Religious History of the Roman Empire The Republican Centuries

Edited by J. A. NORTH

Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © Oxford University Press 2023 The moral rights of the author have been asserted First Edition published in 2023 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this work in any other form and you must impose this same condition on any acquirer Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2022943459 ISBN 978–0–19–964406–3 DOI: 10.1093/oso/9780199644063.001.0001 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.

In affectionate memory of Simon Price (1954–2011)

Acknowledgements The Editor wishes record his debt to all those who have helped in many ways with the production of this book, especially to the unendingly patient staff of Oxford University Press, Karen Raith, Charlotte Loveridge, and Henry Clarke; to all the very helpful Librarians of the Institute of Classical Studies, including Sue Willets and Joanna Ashe; and not least to my wife, Nicola Miller, constant provider of advice and support.

Contents List of Abbreviations List of Illustrations List of Contributors Translators

xi xiii xv xix

Introduction John North

1

I. EARLY ROME 1. Dead Dogs and Rattles: Time, Space, and Ritual Sacrifice in Iron Age Latium Christopher Smith 2. The Priest and the Magistrate: Reflections on the Priesthoods and Public Law at the End of the Republic John Scheid

7

33

I I . R E P U B L I C A N P R A C T I C E S AN D I D E A S 3. Acca Larentia Gains a Son: Myths and Priesthood at Rome Mary Beard 4. Religion and War: On the Relationship of a Society’s Religious and Political Systems Jörg Rüpke 5. Not the One nor the Many; A Pragmatic Approach to Religious Behaviour in a Polytheistic Society: The Example of Rome Andreas Bendlin 6. Action and Ritual in Roman Historians: Or How Horatius Held the Door-Post John North 7. Rites and Practices of Warfare in Italy between Romans and Samnites: Going under the Yoke and the Samnite Legio Linteata Olivier de Cazanove

77

101

125

162

185

x

Contents I I I . S A C R I F IC E 8. Festus and the Role of Women in Roman Religion Rebecca Flemming 9. Interpreting Sacrificial Ritual in Roman Poetry: Disciplines and Their Models Denis Feeney

205

234

IV. ROME AND ITALY 10. The Ritual of Centuriation: Frontinus, Hyginus Gromaticus, and Centuriation Daniel J. Gargola

263

11. The Lucus Pisaurensis and the Romanization of the Ager Gallicus Filippo Coarelli

295

12. Rome and the Great Places of Worship in Italy John Scheid

311

V . L A T E RE P U B L I C A N T RA N S F OR M A T IO N S 13. Cicero and Divination: The Formation of a Latin Discourse Mary Beard

325

14. The Peace of Augustus, the Equinox, and the Centre of the World Alfred Schmid

350

Details of Original Publication Suggestions for Further Reading Index

389 391 395

List of Abbreviations AA AE AJPh ANRW Arch. Laz. ARG ASNP A&R BAB BCH BÉFAR BICS BSAF CCG CÉFR CIL Cl. Ant. CPh CQ CRAI Dial.Arch DNP Enc.Virg. ÉPRO G&R HRRel HrwG HSCPh HTR HWdPh ICS IGRR II ILLRP ILS JDAI JHS

Archäologischer Anzeiger L’Année épigraphique American Journal of Philology Aufstieg und Niedergang der römischen Welt Archeologia Laziale Archiv für Religionsgeschichte Annali della Scuola Normale Superiore di Pisa Atene e Roma BABesch Bulletin de Correspondance Hellénique Bulletin Épigraphique Français d’Athènes et de Rome Bulletin of the Institute of Classical Studies Bulletin de la Société Nationale des Antiquaires de France Cahiers du Centre Gustave Glotz Collections de l’École française de Rome Corpus Inscriptionum Latinarum Classical Antiquity Classical Philology Classical Quarterly Comptes rendus de l’Académie des inscriptions et belles-lettres Dialoghi di archeologia Der Neue Pauly Encyclopedia Virgiliana Études préliminaires aux religions orientales Greece and Rome Historicorum Romanorum Reliquiae (ed. H. Peter) Handwörterbuch religionswissenschaftlicher Grundbegriffe Harvard Studies in Classical Philology Harvard Theological Review Historisches Wörterbuch der Philosophie Illinois Classical Studies Inscriptiones Graecae ad res Romanas pertinentes Inscriptiones Italiae Inscriptiones Latinae Liberae Rei Publicae (ed. A. Degrassi) Inscriptiones Latinae Selectae (ed. H. Dessau) Jahrbuch des deutschen archäologischen Instituts Journal of Hellenic Studies

xii JRA JHS JRS LIMC LTUR MAAR MDAI MEFRA

List of Abbreviations

Journal of Roman Archaeology Journal of Hellenic Studies Journal of Roman Studies Lexicon Iconographicum Mythologiae Classicae Lexicon Topographicum Urbis Romae (ed. M. Steinby) Memoirs of the American Academy in Rome Mitteilungen des deutschen archäologischen Instituts Mélanges d’archéologie et d’histoire de L’École française de Rome MRR The Magistrates of the Roman Republic (ed. T. R. S. Broughton) MusHelv Museum Helveticum OGIS Orientis Graecae Inscriptiones Selectae (ed. W. Dittenberger) OpAth Opera Atheniensia PaWB Potsdamer Altertumswissenschaftliche Beiträge PBSR Papers of the British School at Rome PCPhS Proceedings of the Cambridge Philological Society PP Parola del Passato RA Revue archéologique RAC Reallexicon für Antike und Christentum RAL Atti della Accademia Nazionale dei Lincei RE Real-Encylopädie der classischen Altertumswissenschaft (ed. A. Pauly, G. Wissowa, W. Kroll) RÉL Révue des Études Latines RGDA Res Gestae Divi Augusti RGG4 Religion in Geschichte und Gegenwart, 4th edn RGRW Religions in the Graeco-Roman World Rh.Mus. Rheinisches Museum für Philologie RHR Revue de l’Histoire des Religions RIC Roman Imperial Coinage (ed. H. Mattingly, E. A. Sydenham, et al.) RIL Rendiconti Istituto Lombardo RPAA (= rend. pont.) Rendiconti della pontificia accademia romana di archeologia RRC Roman Republican Coinage (ed. M. H. Crawford) RSI Rivista Storica Italiana SCI Scripta Classica Israelica SEG Supplementum epigraphicum Graecum SMSR Studi e Materiali di storia delle Religioni SOliv Studia Oliveriana—Pesaro YCIS Yale Classical Studies ZAC Zeitschrift für Antikes Christentum ZPE Zeitschrift für Papyrologie und Epigraphik

List of Illustrations 1.1

Graves with associated ritual or cultic objects (from Bietti Sestieri, de Santis, and La Regina 1989–90).

12

1.2

Age classes at Osteria dell’Osa.

14

1.3

Latin chronology.

14

Map of Latium.

20

1.4 11.1

The area S. of Pesaro (block IGM, Fo 109 1 SE Pesaro). Within the circle is the approximate position of the lucus Pisaurensis.

297

11.2

Facsimile of the inscriptions of the lucus, taken from CIL I2 368–75.

298

Facsimile of the inscriptions of the lucus, taken from CIL I2 376–81.

299

11.3 11.4

The ‘sors’ (the lot), preserved in the Museum of Fiesole.

303

14.1

The solarium with the Ara Pacis, after E. Buchner.

352

14.2

Augustus’ horoscope in Rome on 23 September 63 , c.5”15 (System B).

357

14.3

Augustus’ horoscope in Rome on 23 September 63 , c.5”30 (Hipparchic system).

358

14.4

For discussion of Fig 14.4 see: page 360.

361

14.5

Excavated fragment of the animal-circuit pavement, after E. Buchner.

362

14.6

For discussion of Fig. 14.6 see: page 369.

370

14.7

For discussion Fig. 14.7 see: page 369.

370

14.8

Seated Venus/Tellus/Italia, with veiled maidens, from the Ara Pacis. The so-called ‘Panel of Tellus’.

375

14.9

Denarius issued by moneyer Cordius Rufus, 42  (RRC 463/1a (see pl. LIV). Obv., showing Venus, carrying Cupid on her back; for discussion, see vol. 1, pp. 473–5.

376

14.10 Augustus’ horoscope in Rome on 22 September 63 , c.5”35 (System B).

377

List of Contributors Mary Beard is professor of Classics at the University of Cambridge, and a fellow of Newnham College. She has written widely on Roman religion and on other aspects of Roman cultural history. Her most recent book is Twelve Caesars: Images of Power from the Ancient World to the Modern. Andreas Bendlin is Associate Professor of Classics at the University of Toronto. He is interested in and has published on ancient Mediterranean migration, Roman associations and law, and Graeco-Roman cultural history more generally, but the main focus of his research is religion in the Roman Mediterranean, from the city of Rome and Italy to the Imperial Greek East, and from the archaic period to Roman religion’s Nachleben in the modern world. In his work, he attempts to combine the use of theoretical categories and perspectives with a careful contextualization of the historical evidence. Olivier de Cazanove studied at the École normale supérieure in Paris and as member of French School at Rome. For six years (1985–91) he was director of the Centre Jean Bérard (Naples). He taught ancient history and classical archaeology at the universities of Ottawa, Paris 1, and Dijon. He is currently professor of Roman archaeology at the University of Paris 1 Panthéon-Sorbonne. His work focuses on preRoman Italy, as well as on Roman Gaul, in particular on sanctuaries and offerings. He is conducting excavations in southern Italy on the Lucanian settlement of Civita di Tricarico. In France, he directed a programme on the cult places of Alesia, including eleven years of excavations of the large sanctuary of Apollo Moritasgus. Filippo Coarelli was born in Rome on 9 June 1936. He taught in the Universities of Rome, Siena, Cosenza, and Perugia. He is currently Professor Emeritus of the University of Perugia, where he was Professor Ordinario of Roman History, from 1981 to 2008, teaching ‘Greek and Roman Antiquities’ and ‘The Religions of the Classical world’. From 1978, he directed the excavation of the Latin Colony of Fregellae (Ceprano, Frosinone), and subsequently of Falacrinae (Città Reale, Rieti). From 2003, in collaboration with Giuseppina Ghini, he directed the excavation of the sanctuary of Diana at Nemi. He is the author of more than five hundred publications, both scientific and popular in character. Central to his interests are the topography and history of

xvi

List of Contributors

Rome and of Ancient Italy and also Art, both Hellenistic and Republican. He directed and partly wrote the Laterza series of Archaeological Guidebooks. He was the inspirer of the exhibitions on mid-Republican Rome (Musei Capitolini, 1973) and of that on the bimillennium of the Flavians—Divus Vespasianus (Colosseum and the Curia of the Senate, 2009). He was the Editor for the re-foundation of the Review, Dialoghi di Archeologia, originally created by Ranuccio Bianchi Bandinelli. He has received the laurea honoris causa of the Universities of Tours, of Clermont-Ferrand, of the Sorbonne, of Oulu (Finlandia), and of Alicante. He is a member of the Academia Europaea, of the Accademia Nazionale Dei Lincei, of the Académie des Inscriptions et de Belles Lettres, of the British Academy, of the Société des Antiquaires de France and of the Istituto di Studi Etruschi. He has received the Gold Medal of the Republic of Italy for achievements in ‘Education, Culture and Art’. Denis Feeney is Giger Professor of Latin at Princeton University. He is the author of numerous articles on Latin literature and Roman culture, especially religion, together with four books: The Gods in Epic: Poets and Critics of the Classical Tradition (1991); Literature and Religion at Rome: Cultures, Contexts, and Beliefs (1998); Caesar’s Calendar: Ancient Time and the Beginning of History (2007); Beyond Greek: The Beginnings of Latin Literature (2016). Rebecca Flemming has been senior lecturer in ancient history in the Classics Faculty of the University of Cambridge and a Fellow of Jesus College. She has recently moved to the University of Exeter, where she has been appointed to the A.G. Leventis Chair in Ancient Greek scientific and technological thought. Her research focuses on the society and culture of the Roman world, particularly on women and gender, and often using medical evidence. Her monograph Medicine and the Making of Roman Women: Gender, Nature and Authority from Celsus to Galen came out from Oxford University Press in 2000, and the volume she coedited with Nick Hopwood and Lauren Kassell, in 2018. She is currently working on a book about medicine and empire in the Roman world. Daniel J. Gargola is Professor of History at the University of Kentucky. His books are Lands, Laws, & Gods: Magistrates and Ceremony in the Regulation of Public Lands in Republican Rome (1995) and The Shape of the Roman Order: The Republic and Its Spaces (2017). John North taught ancient history in the History Department of UCL from 1963 until 2003, when he retired after having been Professor of History since 1992. On retirement, he became Emeritus Professor and an

List of Contributors

xvii

Honorary Fellow of UCL. He was Director of London University’s Institute of Classical Studies from 2012–14, and subsequently edited a series of the Institute’s Reception studies together with Peter Mack. His research interest has been primarily in the religious history of Rome in the republican period and in the changing character of religious life in the Roman Empire down to the fourth century . More recently, he has worked on the significance of religious activities in the lives of Roman slaves. Jörg Rüpke is Fellow in Religious Studies and Vice-director of the Max Weber Centre for Advanced Cultural and Social Studies of the University of Erfurt, Germany. Before, he was professor for Classical Philology at the University of Potsdam (1995–9) and for Comparative Religion at the University of Erfurt (1999–2008). He was director of the ERC-Advanced Grant Project ‘Lived Ancient Religion: Questioning “Cults” and “Polis Religion”’, co-director of the DFG-Kolleg-Forschungsgruppe ‘Religious Individualisation in Historical Perspectives’ and of the DFG-Research Programme ‘Roman Imperial and Provincial Religion: Globalisation and Regionalization in the Ancient History of Religion’ and is now Co-director of the Kolleg-Forschergruppe ‘Urbanity and Religion: Reciprocal Formations’ (with Susanne Rau). He has published widely on Roman culture and religion. John Scheid was born in Luxembourg in 1946. He achieved his PhD (doctorat de IIIe cycle) in 1972 (Strasbourg). Member of the French School in Rome (1974–7). Director of the excavations at La Magliana (1975–88; 1997–8). Assistant for Roman History at the University of Lille (1977–83). Directeur d’études at the École Pratique des Hautes Études, Section des Sciences religieuses (chair: Religions of Rome), 1983–2001. State doctorate in 1987 (Strasbourg). Professor at the Collège de France, Chair for Religion, institutions et société de la Rome antique, 2001–16. Member of the Académie des inscriptions et belles-lettres in 2017. Corresponding fellow of the British Academy in 2012. Honorary Doctor of the University of Chicago, of the University of Erfurt. Alfred Schmid was tutor in ancient history at Basel and Berne university from 2000 to 2019. He has held a DFG research post at Martin-Lutheruniversity Halle-Wittenberg (from 2018) and is currently investigating the ‘invention’ of horoscope astrology in late Ptolemaic Egypt. He has published monographs on the early principate and astrology, and on early Greek historiography. Other main areas of his interest include: kingship, the history of the nature concept, modernity and ‘antimodernity’ in antiquity. He is co-editor in the ‘Jacob Burckhardt-Edition’

xviii

List of Contributors

at Basel (of the last volume of ‘Griechische Culturgeschichte’ in 2012, and of a volume with Burckhardt’s unpublished lectures on Roman and ancient oriental history, in print). Christopher Smith is Professor of Ancient History at the University of St Andrews. He was also Director of the British School at Rome from 2009 to 2017. He has worked on archaic religion in central Italy, and edited Religion in Archaic and Republican Rome and Italy: Evidence and Experience, with Ed Bispham in 2000.

Translators Christine North read Modern Languages at St Hugh’s College Oxford, before working as a senior lecturer at Middlesex Polytechnic, teaching French language and literature, History of Ideas, and Literature and Philosophy. Her research interests are French Post-Romantic poetry and in particular the works of Stéphane Mallarmé. Her published translations include selected poems by Mallarmé and Laforgue and the first translations into English of the main works of the musicologist Pierre Schaeffer, ‘In Search of a Concrete Music’ and the award-winning ‘Treatise on Musical Objects’. At present she has just completed a new translation of Mallarmé’s major poems and is working on a translation of ‘Journal de mes Sons’ by Schaeffer’s colleague Pierre Henry. For this volume, she translated the paper by Olivier de Cazanove (Chapter 7). Ivan Palencec was the translator of the papers in German by Jörg Rüpke (Chapter 4), Andreas Bendlin (Chapter 5), and Alfred Schmid (Chapter 14). He has been teaching in London, particularly at UCL and Birkbeck College; currently he is working at the British Museum as a research manager. He is not working as a translator at the moment, but hopes to have the opportunity in the future. Jake Wadham taught modern French literature and language at St Edmund Hall, Oxford, for a number of years, specializing on the period of French literature from the nineteenth century to the present; he is leaving Oxford, and Academe, this year. His research interests have concentrated on the relationship between literature and philosophy in the twentieth century. He has also taught French to English translation and, more broadly, the theory and practice of Translation Studies. He translated both the papers by John Scheid (Chapters 2 and 13). The chapter by FILIPPO COARELLI (Chapter 11) was initially translated by the Editor, but then much improved by the author himself.

Introduction John North

This book has taken a long time to see the light of publication. It was on the way, even well on the way, at the time of the early and much to be lamented death of my fellow editor, Simon Price, to whose memory this volume is dedicated. The planning of the book, agreement with the Press, even the commissioning of the translations, were all under way before Simon died in 2011—ten years ago now, as I write. It was conceived as a companion volume to one on the Imperial period which had already appeared in the same series, under the title The Religious History of the Roman Empire: Pagans, Jews and Christians. Both volumes had been planned to have the same structure— equal numbers of chapters, half translated, half in their original English. Simon’s death was the first (and worst) of a series of misfortunes and distractions, which caused prolonged delay. First, I agreed, under some pressure, to act as Director of the London Institute of Classical Studies, initially for one year but eventually for three; this turned out to have been a job well worth doing, but then my health became problematic and was rescued only by a series of explorations and operations. By 2019, we were discussing the possibility of reviving the long-abandoned project. Paradoxically enough, it was the end of 2019 and the beginning of the Covid19 pandemic that saw these efforts coming together at last. It seemed all too likely that we should find we were too late to revive a plan of a decade ago, even if it would complete the twin-volume Religious History of the Roman Empire. However, of our fifteen promised contributors, it turned out that all but one were still available and even willing to be included. The missing (and regretted) chapter was that of Karl-Joachim Hölkeskamp, who had just signed up with another team. As the pandemic emerged in its full horror in the early months of 2020 and life grew narrower and more restricted as the year went by, so work progressed. John North, Introduction In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0001

2

Introduction

The passing of the ten years has perhaps even been helpful in clarifying the difference between the historical phases of ancient history writing in this period. The chapters of the new book cover the years between the late 1980s and the first decade of the new century. That is about twentyfive years making notable progress and introducing new themes and paths of research, but within a framework of broad agreement. Before the end of this period, however, the whole approach to thinking and writing about republican religious history had met with a dramatic challenge. As Jörg Rüpke put it in 2016: ‘I invoke this model only to discard it along with other trends in early twenty-first-century scholarship’. The new way of thinking was to put far greater emphasis on the role of the individual religious agent and far less on the role of religious groups—families, associations, cities. By no means everybody was converted to this new approach and it may be long before there is again any kind of consensus. Meanwhile, particular emphasis has fallen on the idea, originally imported from the United States, of ‘lived religion’, which is to say religious experience forged through everyday practice, rather than based on the ideas and convictions of religious teachers—or bishops. It follows that the work collected in this book marks a definite, but contested, phase in the development of ideas about Roman religion: for some scholars it will still present a basis for future study; for some others it will mark the end of one era and the launching of a very different one. The following survey of the contents of this book is intended to help convey its central themes, but note that this list does not follow the chronological structure underlying chapters of the book: Gods and goddesses— Polytheism and its implications (Bendlin); Myth and ritual (Beard, North); Early festivals and archaeology (Smith); Ritual precision and error (North). War-rituals— The stages of Roman war-ritual: return and triumph (Rüpke); War relations with Italians in particular (Cazanove, Coarelli); Distribution of great religious cult-centres across Italy (Scheid). Priests— The role of priests in relation to politicians and senators (Scheid); Priestly activities around the measuring of land and its relation to augural ritual (Gargola). Communication— Divination via the gods and goddesses (Beard); Communication through the interpretation of the stars (Schmid).

Introduction

3

Innovations— Literature and its role in understanding religion (Feeney); The place of women in Roman religion (Coarelli, Flemming).

These are, of course, no more than a selection of the themes treated in the chapters of this book. There is a good deal of coherence, since the primary concern is with the Romans in peace and war; with their institutions and ideas; and above all with their religious practices. But in many ways, this summary of a long text is liable to be misleading. In the course of the period from Rome’s early monarchic period to the end of the Republican centuries, with which the book is concerned, Roman society was totally transformed. Rome started this period as a minor state of Central Italy and finished as what we would call today ‘a Great Power’—even ‘the Great power’—of the whole Mediterranean area; the Romans started having no claim to any cultural pre-eminence and finished as leaders of art, architecture, and literature, competing only with the Greeks. It is therefore essential to remember that in the course of the chapters this is a society undergoing profound change, from century to century. It will be clear from the above summary of themes that ritual and the analysis of ritual play crucially important roles in the picture of religion and religious life as conveyed by this book. It could be said that this prominence reflects the central position occupied by ritual activities in Roman religion as it was imagined and understood in the period. Ritual was strongly linked with the activity of various groups—clan, family, association, community, village, city, and so on—it was in such groups of varying size that communal rituals were maintained with a god or goddess presiding. It has sometimes been argued (though never universally agreed) that the consequence of this emphasis on the religion of the group was that the Roman gods and goddesses received their recognition through ritual, but that Roman worshippers held no beliefs whatsoever, or at least no beliefs in common. It is important to be clear that both the view that the Romans had rituals, but no beliefs, and that they had beliefs in exactly the same way as do modern Christians are absurd exaggerations of their real position. Any attempt to model ancient pagan religious behaviour on modern Christian religious behaviour leads to confusion, because they are so different from one another. Perhaps the most convincing way of resolving the evident tension between the ritual world of religion and its other world of belief, debate, and argument is to reconsider the relationship between ritual and other media of communication. There can hardly be any doubt that both ritual action and speech are habitual modes of conveying meaning, in religious

4

Introduction

contexts as well as anywhere else. It is perfectly true that ancient pagan religious action seems, at least to modern observers, far more remote from speech or at least from any kind of explanation or analysis than does modern. In fact, religious historians frequently complain about the lack of explanatory material to accompany and clarify religious action. It seems only by problematizing the boundary between religious ritual, on the one hand, and religious argument, on the other, that we have any hope of seeing how the two halves can be fitted together. It is only by fitting together different modes of communication between men, women, gods, and goddesses that we can expect to approach some at least of the problems with which the authors of this book have been concerned. In this second volume of our collection, we have followed the principles applied in our previous volume and, of course, sought to reprint only what was written at the time of first publication, without up-dating. It is more difficult to follow this principle when a text is being translated; it is not always easy to decide what should count as an elegant or more up-to-date translation, what as a serious adjustment of the original meaning. As before, contributors have been invited to add an ‘Afterword’.

I Early Rome

1 Dead Dogs and Rattles Time, Space, and Ritual Sacrifice in Iron Age Latium Christopher Smith

INTRODUCTION In this paper,1 I have tried to move from one Roman festival, the Robigalia, to explore wider issues concerning the nature of ritual both in the Latin world, and in archaeological interpretation. The central theme is to explore the nature of Latin religious experience from the eighth to the sixth century  as it is revealed through dynamically changing social structures, for I believe that to treat ritual, without regard to the effects which it had and which it sought to create, misses crucial opportunities for the interpretation both of ritual and of the world in which that ritual takes place. My starting-point is a discussion of some recent approaches to ritual, in which I identify anthropological and archaeological approaches that form the methodology for my approach. I then proceed to present a body of evidence from a Latin site, Osteria dell’Osa, and from Rome, which gives the material structure of objects and actions to be analysed. My interpretation of this evidence starts from an account of the Robigalia festival, an ancient agricultural festival celebrated annually at Rome probably from the sixth century . I believe this festival gives a clue to how the different aspects of the material evidence can be seen as a part of a coherent evolving practice. In my final section, I attempt to use an experiential understanding of religion to develop an understanding of 1 I am grateful to John Wilkins, Ruth Whitehouse, and Judith Toms for helpful criticism of this paper when delivered in London; the views expressed, and any errors, are my own.

Christopher Smith, Dead Dogs and Rattles: Time, Space, and Ritual Sacrifice in Iron Age Latium In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0002

8

Dead Dogs and Rattles

the changes indicated by the archaeological evidence in ritual symbolism and practice, and to relate these to the nature of the society in which these rituals took place.

THEORETICAL APPROACHES

New Archaeology and the Identification of Cult Places We must begin by meeting the frequent objection that archaeologists are far too keen to assign what they cannot understand or identify to ritual. It is undoubtedly the case that such misinterpretations can occur, but there are ways around this and Colin Renfrew tried to meet the problem head on in his study of the cult site of Phylakopi on the island of Melos (Renfrew 1985). Renfrew proposed a framework of inference based on Binford’s ‘Middle Range Theory’. This is defined by Renfrew and Bahn as ‘a conceptual framework linking raw archaeological data with higher-level generalisations and conclusions about the past which can be derived from this evidence’ (Renfrew and Bahn 1991: 489). This rigorous observation language, ‘a system of scientific inference allowing past cultural systems to be read off the archaeological record’ (Shanks and Tilley 1992: 13), is in fact a combination of empiricism and cross-cultural comparison which can be shown to be deeply flawed.2 When Renfrew tried to draw up his framework, what he actually did was to postulate certain universal truths about religion and ritual. These were seen to have correlates in the material world; hence ritual involves attention focusing, special aspects of the liminal zone, presence of the transcendent and its symbolic focus, and participation and offering. The correlates include special buildings, ritual paraphernalia, repeated and redundant symbols, offerings, and so on. The material remains are then presented in minute detail and checked against the list of correlates which have been postulated. From clear fits like Phylakopi, which no one would question as a cult site since it is strewn with an abundance of votive deposits, and as a building was unlike residential buildings identified on the same site, one can move to other more controversial Aegean sites. Renfrew’s publication is exemplary and valuable, but it is methodologically suspect. Generalizations about religion have their place, but it is not clear that they are universally applicable. Nor is it clear that the observation and identification of objects, which must be the first move in 2 For theoretical objections and analysis, see Gibbon 1989; Hodder 1991:107ff.; Shanks and Tilley 1992: chs 2 and 3. Renfrew has returned to the subject in Renfrew 1994: 47–54 and the gap between cognitive archaeology and post-processual archaeology seems to be closing.

Dead Dogs and Rattles

9

such a procedure, as in any archaeological work, is not itself subjective and value-laden. Processual archaeology has become discredited because it has failed to fulfil its promise of finding the universal framework, and in the meanwhile has produced a large body of work which has proceeded from material evidence to social structure and organization but has left human agency out of the picture. What is central in any interpretation of ritual is the careful use of context within the cultural system under investigation. This is necessarily and admittedly subjective, but we should remember that artefacts in themselves have little or no meaning; they are given meaning by the uses to which they are put, which context can sometimes reveal to us (Hodder 1991). In other words, although one may be misled about a particular site or object, this is inevitable, even with the most scientific approach, and such an approach risks skewing one’s perception of the past towards a systems theory which operates at the ‘high’ level of cultures and economies, and which, although in itself valuable and interesting, excludes much of the past as experienced by us and as experienced in its own time.

Understanding Ritual Anthropologists have long been concerned to examine ritual. It was Robertson Smith who first proposed certain features of ritual which are still sometimes propounded today, that ritual is the real core of religion, and that the purpose of ritual was, as we would say, functional; that is, it was aimed at the restoration of the solidarity of the group (Robertson Smith 1889; Lewis 1994: 569–71). Functionalism is largely rejected by recent writers because it implies a degree of intentionality—the ritual was intended to promote the solidarity of the group—and because it implies a rather straightforward mapping of ritual onto society. If the effect of ritual is to promote the solidarity of the group, this should not necessarily be seen as its cause (Turner and Maryanski 1979). The approach of Fustel de Coulanges, influential in anthropology through the intermediation of Durkheim, is a reverse presentation of the same functionalist approach; here, rituals are not created by social groups, but rather are the creators of such groups (Fustel de Coulanges 1980). Fustel de Coulanges set out to show that ancestor worship was at the heart of Roman society, creating the notion of the ancient city and its patriarchy, but this causation is hard to demonstrate, and Fustel de Coulanges himself actually presents a more nuanced account.3 3 On Fustel de Coulanges, the introduction by Momigliano and Humphreys (Fustel de Coulanges 1980: ix–xxiii) remains fundamental; see also Humphreys (1978) for methodological approaches to the anthropology of ancient history.

10

Dead Dogs and Rattles

Functionalism came to be regarded as problematic because it was so reductionist; rituals are far more complex than a simple social explanation allowed. Evans-Pritchard and others saw rituals as an explanation of the universe, a quasi-theological discourse. These approaches, generally termed intellectualist and symbolist, can be criticized in that the degree of participation which a ritual demands is not required by theology. The power of ritual seems to need another explanation. Symbolic conceptions of ritual tended to run aground when they denied the complexity and ambiguity of the ritual process. As Turner and Lewis have stressed, symbols are not univocal, they are ambiguous (Turner 1967; 1969; Lewis 1980). Turner’s approach emphasized the way in which the complexity of symbolism makes it suitable for social manipulation, because it is associated with moments of heightened emotion and sensuousness. I quote here from Turner’s brilliant account of ritual among the Ndembu of Zambia: ‘I found that I could not analyse ritual symbols without studying them in a time series in relation to other “events”, for symbols are essentially involved in social process. I came to see performances of ritual as distinct phases in the social processes whereby groups became adjusted to internal changes and adapted to their external environment. From this standpoint the ritual symbol becomes a factor in social action, a positive force in an activity field. The symbol becomes associated with human interests, purposes, ends, and means, whether these are explicitly formulated or have to be inferred from the observed behaviour. The structure and properties of a symbol become those of a dynamic entity, at least within its appropriate context of action’ (Turner 1967: 20). Turner’s analysis, here and elsewhere, is subtle and complex, but it shares with Geertz’s view of the symbol the concept that symbols are, to use Asad’s terms, ‘meaning-carrying objects external to social conditions and mental states’. Asad’s critique of Geertz is an attack on what is seen as a generalization of religion which cannot be held to apply to medieval Christianity, but in fact the problem is a much deeper one, as Asad shows when he offers what for me is a more satisfactory view of the symbol as focused on ‘the relations between socially signifying and psychologically organising practices’ (Asad 1983: 241). In a previous article I have explored various more recent approaches to these issues (Smith 1994b). Asad sought to reinterpret the nature of religious symbolism, focusing rather on the authorizing practices and discourses, and the historical conditions, that are necessary for ritual and religion to be effective (Asad 1983). Asad’s Foucaultian project of investigating the ways in which power creates religion may be related to other

Dead Dogs and Rattles

11

important work stressing the ways in which ritual practice and ritual meaning can be manipulated and passed on. Bloch for instance showed how the circumcision ritual of the Merina of Madagascar developed over two centuries, the practice remaining fairly constant but the significance and use of it altering dramatically in reaction to French colonial activity (Bloch 1986). Harrison revealed ways in which ritual can be seen as a form of intellectual property, and can thus be socially controlled (Harrison 1992). The focus on ritual in context may be traced back ultimately to the important contributions of Giddens and Bourdieu, who both stress in different ways the interaction between practice and the social and physical landscape in which it takes place (Giddens 1984; Bourdieu 1977; see also Barrett 1988; 1991; 1994). The tendency of this scholarship is to direct us to approach ritual from the point of view of the material landscape in which it takes place, the knowledgeable activity of the participants, and the dispositions, the habitus, which it creates in them. In my conclusion I shall attempt to apply these approaches to the body of evidence which is presented in detail here.

THE EVIDENCE The evidence which I present in the following section is from Osteria dell’Osa and Rome. The evidence from Osteria dell’Osa was first gathered by Bietti Sestieri and de Santis (Bietti Sestieri, de Santis, and La Regina 1989–90), and I summarize this evidence with documentation of parallels elsewhere in Latium in the sections that follow. The table in Figure 1.1 is borrowed from this article. The evidence for votive deposits from Rome is presented in Bartoloni (1989–90) and the catalogue Il Viver Quotidiano. The interpretations which follow are my own.

Burials at Osteria dell’Osa Osteria dell’Osa is presently a small town about 15 miles outside Rome.4 It is on the edge of the Lago di Castiglione and was one of a number of 4

The site has been published by Bietti Sestieri (1992a) and discussed further in Bietti Sestieri (1992b). Full bibliographies can be found there. All references to specific types of material in this section come from Bietti Sestieri (1992a). Smith (1996) gives an account of developments in Latium as a whole; see also Smith (1994a).

IIA

IIA/B

IIB

IIB

M6/7

M7

M4

F5

F7

F4

F6

F4

F3

F4

F7

F8

357

87

578

113

153

433

394

580

218

581

43

482

CR

IN

IN

IN

IN

IN

IN

IN

IN

IN

IN

CR*

CR*

CR*

CR*

CR*

CR*

X

X

X

X

X

XX

XX

X

X

Stat.

X

X

X

X

X

X

X

X

X

Knife

XX?

X

Broken vases

X

X

Rattles

X

?

X

X

X

X

?

X

Central pos’n

X

X

X

X

Huturn

X

X

X

X

Burner

X

X

X

X

X

X

X

X

X

X

?

X

Arms Razor

Fig. 1.1 Graves with associated ritual or cultic objects (from Bietti Sestieri, de Santis, and La Regina 1989–90).

* indicates that the tomb belongs to one of the central two grave groups at the beginning of the necropolis.

IIB

IIB

IIB

IIA/B

IIA/B

IIA

IIA

IIB

IIA

IIA

M5

IIA

M6/7

M5

127

IIA

IIA

308

M4

142

139

M4

126

Tomb Sex Age Period Ritual Vases

X

?

X

X

X

Wealth

X

X

X

X

X

X

X

X

X

X

X

X

X

X

Fine pottery

X

X

X

X

X

Faunal remains

Dead Dogs and Rattles

13

small prehistoric settlements that coalesced into the urban settlement of Gabii around 600 . We have no settlement evidence from Osteria dell’Osa except for surface scatters of pottery. At Osteria dell’Osa, a necropolis of about six hundred individuals has been discovered and reported in full. The necropolis covers the period from around 900 to around 650 . The earliest burials are some male cremations, around which other cremations and more commonly inhumations gather through the ninth century . It is my belief that this burial site, like all the formal necropoleis of central Italy, did not serve all members of the relevant community, but was exclusively kept for an élite.5 Burials in central Italian necropoleis tended to be accompanied by objects, the assemblage of which is called a corredo. Some of the corredi here have objects or collections of objects which have been taken as indicating some form of ritual or cult connection of the deceased during their lifetime. This is the evidence I wish to discuss here. The evidence is summarized in Figures 1.1–3. No object in itself can be taken as necessarily referring to the special status of the deceased in life; the argument must be that the presence of certain assemblages which within the context of the necropolis as a whole are rare, or which are found in undeniably cultic assemblages elsewhere, should indicate something distinct from the modalities of the death ritual itself. Miniaturized vases are in fact a standard feature of male cremations, which all occur at the earliest foundation of the cemetery. In addition, they have a wide distribution in Latium as a whole in the early Iron Age (Atti 1980: 57, 62).6 There are similar objects found in votive deposits, as we shall see later. In the eighth century (Period III), knives are also common, but the knives of the second period, all of which are smaller than those of the third period, are uncommon here and elsewhere.7 The statuettes are rare throughout Latium in burials and are entirely confined

5 Arguments in Smith (1996: 63–5, 112–13); see Morris (1987) for similar views on Athens. 6 The olletta a rete (Type 6a) is found in seven other male tombs at Osteria dell’Osa apart from the ones listed in Fig. 1.1 and can be parallelled at sites in the Alban Hills, Rome, Pratica di Mare, and Satricum. The smaller vases (Type 37) are only found in tombs listed in Fig. 1.1, and some later damaged tombs. The parallels are with early tombs and more particularly with votive deposits throughout Latium. 7 Osteria dell’Osa Type 58. Parallels from Bianco Peroni (1976) are found in Veii, Palombara Sabina, Perugia, and the Alban Hills in tomb contexts. At Bietti Sestieri (1992a: 398) the distinction between the relatively uncommon miniaturized knives of Period II and the more common full-size knives of Period III is sustained.

14

Dead Dogs and Rattles Class

Age

1–2

0–12

3

12–20

4

20–30

5

20–40

6

30–40

7

40–60

8

60+

Fig. 1.2 Age classes at Osteria dell’Osa.

Latial Period

Date 1000–900

IIA

900–830

IIB

830–770

III

770–730

IVA

730–630

IVB

630–580 580–500

Fig. 1.3 Latin chronology.

to Period I and Period II.8 Moreover, all the parallels which are cited for sites outside Latium are significantly different because the statuettes are attached to other vessels, whereas those of Latium are free-standing. The rattles are unusual as well.9 There is one clear parallel from a grave at Valvisciolo, and reports of rattles in two votive deposits in Tivoli, and in the foundation wall of the Palatine defence which Carandini has discovered and which he has characterised as a ‘deposito di fondazione’. Parallels are cited from Emilia Romagna and a burial at the Etruscan 8 Osteria dell’Osa Type 36a. Only found in burials listed in Fig. 1.1. Parallel forms of the free-standing statuette are found in tombs in the Alban Hills; see Bietti Sestieri (1992a: 316) for parallels from elsewhere in Italy. 9 Osteria dell’Osa Type 29. Only found in burials listed in Fig. 1.1. Bietti Sestieri (1992a: 307) and Bietti Sestieri, de Santis and La Regina (1989–90: 68 n. 11), where the deposit is interpreted as the remains of a Period III female infant’s burial, associated with the other Forum burials of this type; see Gjerstad (1966: 57) and Smith (1996: 80 n. 28) for these burials. The careful reinterment of objects within another context reflects the care with which burials were treated in the Latin world, also indicated by the limited unintentional superimposition of tombs.

Dead Dogs and Rattles

15

necropolis of Veii (Quattro Fontanili).10 This grave is an exceptionally wealthy eighth-century grave of a 3-year-old child, and contains a bronze rattle, unlike the clay rattles of Latium. The broken vases are vases, some of them miniaturized, which were deliberately broken and placed in the fill of the tomb. The fact that they are of the same general types as the vessels found in the tombs would seem to exclude the possibility that they were random. Deliberately broken vases are rare at Osteria dell’Osa (only found in burials listed in Fig. 1.1), but seem to be more common in other necropoleis from Period IVA (Bietti Sestieri, de Santis, and La Regina 1989–90: 74–5). Examination of the specific objects identified by Bietti Sestieri and de Santis in their context at Osteria dell’Osa and elsewhere in Latium reveals two central patterns. First, in Periods I and II, the frequency of occurrence is limited, and all the Latin burials (fourteen noted from outside Osteria dell’Osa at Bietti Sestieri, de Santis, and La Regina 1989–90: 73) are also notable for their unusual wealth. In other words, they are unusual even among the relatively small number of burials which were preserved from these periods. If we combine this observation with the belief enunciated above that all preserved burials from this period were themselves exceptional, we have reasonable grounds for believing that these objects have added symbolism beyond the burial ritual itself. Secondly, the objects do not necessarily preserve this rarity over time. Knives and deliberately broken vessels are found in burial contexts at this site or elsewhere in greater quantities in later periods. There are two ways of interpreting this shift; one is to say that the objects lose or change their specific value over time, or never had this value except at Osteria dell’Osa, in other words that their meaning is related to their place and time. The other is to assert that they never had this value in the first place and that their rarity is entirely dependent on the circumstances of archaeological recovery. The latter assertion is obviously extremely damaging to the attempt to identify them as having special ritual associations, and therefore must be met here. It should be noted that burials from the ninth and eighth century  are more common than burials from later periods, so it is not the case that an increasing occurrence of a certain object is related solely to an increasing number of recovered tombs. At Osteria dell’Osa only some eighty burials come from the period 770 to 720  as against about 450 burials from the period 900 to 770  (Bietti Sestieri 1992b: 103–18). No Latin site which is represented in both periods and has been excavated with care has more burials from Period IV than from Period III.

10

Colonna 1988: 297 n. 18 for references; NSc. 1967: 247–8 figs 98.8 (T.GG 6–7).

16

Dead Dogs and Rattles

Secondly it is clearly the case that not all necropoleis are identical in their practices. Castel di Decima, for instance, is distinct from Osteria dell’Osa in the number of military objects and chariots which are found in burials (Bartoloni, Cataldi Dini, and Zevi 1982). Within the generally similar Latin ritual, there were possibilities for different approaches. It is also clear that the practices of deposition changed markedly over time. Cremation gave way to inhumation in the early ninth century; wealth of burials increased markedly, and the number of visible burials decreased dramatically, in the seventh century. Weapons, represented only in miniature in previous phases, are found in full size in Period III (Bietti Sestieri 1992b: 203). The stress I have placed on the context of objects implies that it is dangerous to interpret the evidence from one burial site on the basis of evidence from another, or from the same at a different period. In other words, the rarity of knives or deliberately broken pottery in ninth- and early eighth-century burials at Osteria dell’Osa should not be ignored because the knives become more common in the later eighth century at Osteria dell’Osa, or because deliberately broken pottery is found commonly in other necropoleis in the seventh century.

Sacrifice There is one instance of an apparent sacrifice at Osteria dell’Osa, which is to be found at the edge of the Period II burials, beside an area which was used exclusively for Period III burials. This is represented by the remains of an adult dog, with no other burial deposits. The nearest material from adjacent burials is late IIB (Bietti Sestieri, de Santis, and La Regina 1989–90: 75–6; Bietti Sestieri 1992a: 34, 488–9). The interment is on the very edge of the group of Period III burials which is separated off from the other burials in the necropoleis, and which has been identified by the excavator as marking a radical shift in the ordering of society towards aristocratic groups close in function to the later gens (Bietti Sestieri 1992b: 199–203).

Votive Deposit Right on the northern edge of the necropolis was found a votive deposit; a group of vessels similar to those which are found in the burials from the beginning of the necropolis, and some of which were miniaturized. The material appeared to have been deposited at one time; there are two

Dead Dogs and Rattles

17

major groups of material, with a scatter in between, perhaps representing one single group disturbed at a later date. All the material was found in one level of about 4–5 cm in depth. There were over sixty complete or fragmentary vessels, including two which must have been made outside Osteria dell’Osa. In form the vessels are identical to those which are used in votive deposits elsewhere in Latium (Bietti Sestieri, de Santis, and La Regina 1989–90: 76; Bietti Sestieri 1992a: 34–5, 873–4).

Votive Deposits at Rome Characteristic of the votive deposits at Rome are miniature vessels, impasto biscuits, and bronze statues. A number of deposits are known. Some cluster on the Esquiline Hill, near an area which was used for burials (Bartoloni 1989–90). An important deposit has also been found at the site of the church of S. Omobono on the Tiber in the Forum Boarium, which has been identified as the later site of the temples of Fortuna and Mater Matuta (Editorial Group 1989; Coarelli 1988). The S. Omobono deposit is important in that it also contained analysable vegetable and animal remains, which has proved that food offerings and sacrifices were also made here contemporary with the votive deposits. The vegetable remains were mostly of legumes and grain. The animal remains were largely ovicaprine with some bovine and porcine animals. A number of these animals were either gravid when sacrificed, or new-born. There were also three puppies and one adult dog. The dates of these votive deposits are largely seventh/sixth century . The votive deposits in Rome are thus contemporary with those found throughout Latium; most sites have produced at least one area where there has been the repeated deposition of offerings similar to those found at Rome, and, as at Osteria dell’Osa, the offerings, especially the miniaturized vessels are identical with those found in contemporary burials. In many instances, these deposits can be seen to be the precursors of later sixth-century temples which are built directly on the deposit. The general development seems to be from an open-air deposit, to a hut, to a monumental temple. Satricum offers a particularly good example (Chiarucci and Gizzi 1985: 39). As I maintain elsewhere, it is no accident that the rise of monumental architecture, temples, public buildings, and private residences coincides with the decline in the evidence for burials from the sixth century on (Smith 1996). There is certainly no reduction in population, nor in wealth to explain this phenomenon, which I would attribute to the redirection of wealth by an élite.

18

Dead Dogs and Rattles ANALYSIS

The Robigalia and Other Dog Sacrifices The Fasti Prenestini, a first-century / calendar, gives the following entry for the Robigalia, which took place on 25 April: The Robigalia festival is held at the 5th milestone on the Via Claudia, lest mildew harm the crops. There is a sacrifice and games with races for young boys and adults.11

We have a more informative account of the festival in Ovid (Fasti 4.905ff.): On that day, as I was returning from Nomentum to Rome, a white-robed crowd blocked the middle of the way. A flamen was on his way to the grove of ancient Robigo, to throw the entrails of a dog and a sheep into the fire. Straightaway I went up to him to inform myself of the rite. Your Flamen, O Quirinus, spoke as follows: ‘You scaly Mildew, spare the sprouting corn, and let the smooth top quiver on the surface of the ground. Let the crops, nursed by the stars of a propitious sky, grow till they are right for the sickle. Yours is no feeble power: the corn on which you have set your mark, the sad husbandman gives up for lost. Winds, showers and glistening frost that nips the sallow corn do not hurt it as much as when the sun warms the wet stalks; then dread goddess is the hour to wreak your wrath. Spare, I pray, and take your scabby hands from off the harvest! Harm not the crops; it is enough that you have the power to harm. Grip not the tender crop but grip the hard iron. Forestall the destroyer. Better that you should gnaw at swords and baneful weapons; you have no need of them, the world is at peace. Now let the rustic gear, the rakes and the hard hoe and the curved share be burnished bright; but let rust defile the arms and when someone tries to draw the sword from the scabbard, let him feel it stick from long disuse. But do not profane the corn, and may the husbandman be able to pay his vows to you in your absence.’ So he spoke. On his right hand hung a napkin with a loose nap, and he had a bowl of wine and a casket of incense. The incense and wine and sheep’s guts and the foul entrails of a filthy dog he put upon the hearth—we saw him do it. Then to me he said, ‘You asked why the strange victim is assigned to these rites’ (I had asked). ‘Learn the cause’, the flamen said. ‘There is a Dog (they call it the Icarian Dog), and when that constellation rises the earth is parched and dry, and the crop ripens too soon. This dog is put on the altar instead of the stellar dog, and the only reason why this happens is his name.’12 11

FERIAE ROBIGO VIA CLAUDIA AD MILLIARIUM V NE ROBIGO FRUMENTIS NOCEAT. SACRIFICIUM ET LUDI CURSORIBUS MAIORIBUS MINORIBUSQUE FIUNT. 12 See Appendix 1, below.

Dead Dogs and Rattles

19

There are two major problems with Ovid’s account here. The first, and most quickly dealt with, is Ovid’s account of the reason behind the festival. Ovid’s claim that the Dog-Star was rising in April is in fact untrue; it was setting. It is clear from Ovid that the sacrifice of a dog is unusual, and the false etymology of a kind so common in the Fasti can be dismissed as a poor explanation, reliant on Greek sources which also mention the destructive power of the Dog-Star Sirius.13 The other problem is Ovid’s geography. If the Fasti Prenestini is correct, the sacrifice took place at the 5th milestone on the Via Claudia. Five miles was thought to be the old extent of Roman territory, and indeed fits well with the situation as we understand it in the late sixth and fifth century  (Cornell 1989: 243ff.). If Ovid was coming back along the Via Nomentana, he ought not to have been anywhere near the cult site (see Fig. 1.4). Mommsen ingeniously suggested that Ovid had come back through his gardens at the junction of the Via Claudia with the Via Flaminia at the Milvian Bridge.14 There is another more plausible explanation, and it is one which has been put forward by Palmer, that the procession was actually stopping at a number of places before it reached the final cult site. I am not sure how much weight one can put on Ovid here; it is just about conceivable that Ovid has elided the time between his meeting the procession and the time of the offering. The topographical problem remains though, and I suggest that Palmer is right to see a procession here, even if not a sequence of offerings (Palmer 1970: 110–13). It is worth pointing out in passing that the prayer which Ovid has versified is very similar to that made by the husbandman in Cato’s De agricultura 141, when leading the suovetaurilia (a swine, a ram, and a bull) around his farm in the purification ritual: Father Mars, I pray and beseech you, that you be gracious and merciful to me, my house and my household; to which intent I have bidden this suovetaurilia to be led around my land, my ground and my farm; so keep away, ward off and remove sickness, visible and invisible, barrenness and destruction, ruin and unseasonable influence; and permit my harvests, my grain, my vineyards and my plantations to flourish and to come to good issue, and preserve in health my shepherds and my flocks, and give good health and strength to me, my house and my household. To this intent, of purifying my land, my ground and of making an expiation, as I have said,

13

See Porte 1985 on etymologies in Ovid’s work; Scullard 1981: 108–10 and Warde Fowler 1899: 88–91 on the Robigalia. For the Dog-Star Sirius, see for instance Homer, Iliad 5.5; 22.25ff.—Homer gets his astronomy right. 14 In Corpus Inscriptionum Latinarum 1.1 (2nd edn) (Berlin 1893): 392.

20

Dead Dogs and Rattles

Fig. 1.4 Map of Latium. deign to accept the offering of these suckling victims; Father Mars, to the same intent deign to accept the offering of these sucklings.15

15 Mars pater, te precor quaesoque uti sies volens propitius mihi domo familiaeque nostrae, quoius re ergo agrum terram fundumque meum suovitaurilia circumagi iussi, uti tu morbos visos invisosque, viduertatem vastitudinemque, calamitates intemperiasque prohibessis defendas averruncesque; utique tu fruges, frumenta, vineta virgultaque grandire beneque evenire siris, pastores pecuaque salva servassis duisque bonam salutem

Dead Dogs and Rattles

21

This early second-century  prayer is probably just as mannered in its way as Ovid’s, but there may be less intrusive material; Ovid has cast his prayer for peace very much in the Augustan mould. Other references to the Robigalia are limited. Pliny (Natural History 18.285) writes: Numa instituted the Robigalia in the 11th year of his reign; these take place on the 25th April because it is at that time that mildew attacks the crops.16

There was in fact another dog sacrifice around this time of the year. Pliny the Elder gives us this information (Natural History 18.14): The following occurs in the Memoranda of the Priesthood: ‘Let a day be fixed for taking augury by the sacrifice of a dog before the corn comes out of the sheath and before it penetrates through into the sheath.’17

This must be a day some time in spring. We learn more about this from Festus (358L): Red dogs, that is, those that are reddish in colour, are sacrificed, Ateius Capito tells us, in the Dog Sacrifice, on behalf of the crops against the savagery of the Dog-Star.18

and Festus (39L): The Porta Catularia at Rome is so called because not far from there red dogs are sacrificed to placate the Dog-Star which is inimical to crops, so that the growing crops may come to maturity.19

Other references to sacrifices of dogs are also attested; a dog is sacrificed to the Luperci (Plut., RQ 111), and to Geneta Mana, and the Lares Praestites (ibid. 51) were depicted with dogs at their feet and wearing dogskins, which Fowler long ago suggested indicated that a dog was sacrificed to them. Among the numerous prohibitions under which the Flamen Dialis laboured was one against touching or owning a dog.

valetudinemque mihi domo familiaeque nostrae; harumce rerum ergo, fundi terrae agrique mei lustrandi lustrique faciendi ergo, sicut dixi, macte hisce suovitaurilibus lactentibus immolandis esto; Mars pater, eiusdem rei ergo macte hisce suovitaurilibus lactentibus esto. 16 Robigalia Numa constituit anno regni sui XI, quae nunc agunter a.d. VII Kal. Mai., quoniam tunc fere segetes robigo occupant. 17 ita enim in commentariis pontificum: Augurio canario agendo dies constituantur priusquam frumenta vaginis exeant nec antequam in vaginas perveniant. 18 rutilae canes, id est non procul a rubro colore, immolantur, ut ait Ateius Capito, canario sacrificio pro frugibus deprecandae saevitiae causa sideris caniculae. 19 Catularia Porta Romae dicta est, quia non longe ab ea ad placandum caniculae sidus frugibus inimicum rufae canes immolabantur, ut fruges flavescentes ad maturitatem perducerentur.

22

Dead Dogs and Rattles

It is possible that a similar festival is recorded in the Iguvine Tablets at IIIa 15ff.20 If the translation is to be trusted, the parallels between this sacrifice and the Robigalia sacrifice are striking, and although the possibility of borrowing is always there, the Umbrian example is an important indication of the detailed knowledge required for cult activity. For what it is worth, Columella (10.342) associated the Robigalia with other practices which are Etruscan, so the borrowing may have been by Rome from Etruria:

20 The Hondia: the day for (the sacrifice of) the dog is appointed for the final period of the Cerealia occurring between moons. If he desires to perform it the adfertor is required, after having observed the birds, to perform it at the full moon. He shall bring the things pertaining to the Hondia; he shall bring the dog, grain, a strusla cake, a ficla cake, mead, wine, ground salt, a maniple, wet and dry vessels, and unguent. He shall place fire on the altar; he shall perform the sacrifice with mead. He shall slay in honour of Hondus Jovius an unblemished dog from the temple-property for the gens Petronia among the Atiedian brothers. The sacrifice shall be (performed) upon the ground. He shall take the underparts of the dog; he shall take the lower limbs. He shall roast the sacrificial pieces on a sufficient number of spits; he shall place a sufficient number of roasting pans under (the spits). He shall watch the ground. He shall present grain offerings with mead, pour a libation, and dance the tripudium, offering wine at each turning. Nine times he shall dance the tripudium. ‘(I invoke) thee with mead, (I invoke) thee with wine’ he shall say. He shall take the spits and the boiled portions. That which he carries last (with it) he shall carry the fat drippings at the same time. He shall pray with a libation upon the ground. He shall cut off two (pieces) of the dog as burnt-offerings, and the third as an erus. At the same spot he shall cut the fat-drippings. He shall add a strusla cake and a ficla cake. He shall make the presentation of the dog. He shall pray with the portion around the (dog’s) foot; he shall pray with the uncut portion; he shall pray with the vempersondro; he shall place the underparts over at the side; he shall pray with vessels of which ever type he desires. He shall pour a libation, dance the tripudium, approach (the altar), and stop. He shall place the underparts at the back. The liver (which is) the erus, he shall carry in his hand. He shall go to the Obelisk. He shall bring mead in two bowls; he shall bring the spits, the smearing sticks not yet used, the vessels wet and dry and the unguent. He shall pour a libation from a bowl to Hondus Jovius for the gens Petronia among the Atiedian Brothers. With the ceremonial spits he shall pray on the other side of the Obelisk. At the same spot he shall pray with the smearing-sticks. With the ceremonial vessels wet and dry he shall pray at the Obelisk. He shall pour a libation and dance the tripudium. He shall anoint the Obelisk, pray with the ceremonial unguent, and wash his hands away from the altar. He shall return to the altar; at the altar he shall pray silently with ceremonial wine. He whose duty it is shall carry and distribute the erus. He shall distribute the wine and mead. He shall grind (a portion) of the strusla cake, the ficla cake and the limbs. He shall extinguish the fire with a bowl of mead. He shall pray with the whole and the ground (grains). He shall stand up and take way the things which remain. The sacrifice shall have been completed. The dog shall be buried at the altar. Lucius Tetteius, son of Titus, approved (the foregoing) in his quaestorship. (Poultney’s 1959 translation.) On the Iguvine Tablets, see now Wilkins in Malone and Stoddart 1994: 152–72. Wilkins sums up Table III as “obscure” at 171. Nothing in my interpretation of the Robigalia rests on the apparently similar festival at Gubbio.

Dead Dogs and Rattles

23

So that country people do not suffer from these monstrous pests, varied experience and toil and custom their teacher has shown the poor farmers new skills of safety, to calm the furious wind and avert the tempest by Tuscan rites. Hence lest foul Robigo parch the green herbs, she is placated with the blood and entrails of a suckling puppy; hence Etruscan Tages is said to have fixed the skinless head of an ass at the edge of the field; hence Tarchon, to avert great Jupiter’s thunderbolt, often hedged his domain with bryony . . . .21

The Robigalia is only one of a series of important and probably ancient agricultural cults around this time of year. In fact the month of April is full of purification rituals. In order, we have the Fordicidia, in which gravid cows were sacrificed on the Capitol and each of the thirty curiae; the Cerealia to Ceres who presides over the generative powers of the earth; the Parilia in which the flocks were purified; the Robigalia; and the Feriae Latinae, the great festival of the Latins on the Alban Hills, which celebrated the common ancestry of the Latins (Scullard 1981: 96–115). What are the essential elements of the Robigalia? The purpose, as with most of these cults, is the avoidance of crop failure or agricultural disaster, which requires the purification of the land and the appeasement of the deity. The processional nature of the Roman festival, taken together with the parallels from other festivals of similar nature would also suggest a concern with boundaries. The intimate connection of boundaries and disease is brought out by Columella, but also by the great Vegoia prophecy, dated by Heurgon to the period of the Social War in the early first century , in which the Etruscan prophetess predicts dire consequences for anyone who tampers with boundaries (tular in Etruscan), consequences which include damage to crops from rain, hail, heat, and rust (robigo) (Heurgon 1959). A specific and unusual animal was chosen for sacrifice. This animal, the dog, is connected elsewhere with chthonic deities and protecting deities. It is notable that an animal which was considered unfit for eating 21

Haec ne ruricolae paterentur monstra, salutis Ipsa novas artes varia experientia rerum Et labor ostendit miseris, ususque magister Tradidit agricolis, ventos sedare furentes, Et tempestatem Tuscis avertere sacris. Hinc mala rubigo virides ne torreant herbas, Sanguine lactentis catuli placatur et extis. Hinc caput Arcadici nudum cute fertur aselli Tyrrhenus fixisse Tages in limite ruris. Utque Iovis magni prohiberet fulmina Tarchon, Saepe suas sedes praecinxit vitibus albis.

24

Dead Dogs and Rattles

is sacrificed to protect food; one might also note that the notion of the scapegoat is widespread in the ancient world.22 The evidence which we now have also allows us to say something about Roman topography. The augurium canarium took place at the Porta Catularia. This was thought to be unknown, but Coarelli has identified it by a process of elimination, since we know from a scholion to Suetonius’ Life of Augustus that it was between the Carmentalis and Fontinalis gates. One option is open, a gate and staircase still visible in 1524 near the church of S. Andreae in Vincis, behind which was a magnificent set of stairs up to the Capitoline Hill. As is quite clear, the gate is very close to the S. Omobono complex, where puppies have been found as required by the festival, and I regard this as important confirmation that Coarelli’s topographical guess is correct.23 We have seen therefore that a detailed consideration of the Robigalia festival permits us to fix it spatially and with reference to other ceremonies in the city of Rome which are connected to it either by the nature of the sacrifice involved, or by the place of the Robigalia in the calendar and the functions it fulfilled.

Knives There is little to say about the knives at Osteria dell’Osa. Knives were of course essential to the act of sacrifice. Whilst weaponry is found in other male cremations of the period, the knife is not a standard item. So its presence in the context of a few special graves which have other unusual features seems to mark a different kind of statement about the deceased, and not simply another piece of armour. What convinces me that the knives refer to the cult responsibilities of the deceased is that in Period II, they are only found in male cremations. The control of religion by heads of family is clearly attested in Roman religion, and this control will become even more important for our argument in due course (North 1989: 604ff.).

22 Bremmer 1983 is mostly concerned with Greek examples, but note the focus on boundaries (p. 314) and the belief that infertile trees promoted infertility, the same sort of ‘sympathetic magic’ suggested for the dog in the Robigalia (pp. 308ff.). 23 Coarelli 1988: 41, 369 with references; the Porta Catularia can probably be seen on a fragment of the Severan Marble Plan (Carettoni, Colini, Cozza, and Gatti 1960: tav. 29). It was about 150 m from the S. Omobono complex.

Dead Dogs and Rattles

25

Rattles To demonstrate the cult associations with the rattles is more difficult. Rattles are found on a group of Geometric vases in Greece, and in all cases seem to refer to a funeral occasion. Kurtz and Boardman write that the gift is appropriate both to a child, and to a ritual (Kurtz and Boardman 1971: 76ff., 352). In the Veii case, this connection is shown by the age of the child; it is not evident in the Osteria dell’Osa cases where the two women are teenage to twenties. It is possible that there is some reference here to their having died in childbirth, but there is another possibility, that the rattles were used in some form of apotropaic ritual. The importance of percussion is attested by the tripudium, the noisy dance which is a common feature of early ritual. Wille claims to see rattles in Etruscan paintings in a ritual context (Wille 1967: 567–72). Finally, Ovid, Fasti 5.441, describing the Lemuria festival to cast out the ghosts from the house, tells us that the head of the household clashes bronze vessels to scare off the dead. It is possible therefore that noise was an important part of early rituals, and that although the rattle is ousted by the wind instruments which are far more common in the historical period, the rattle was used in earlier periods. If this is correct, it is interesting that the rattles are in women’s graves, and this reminds us that women were important in religion, though males held the offices of priests.

TOWARDS AN INTERPRETATION The rhythms of life in Osteria dell’Osa in the ninth century  were governed by the basic facts of existence: the rising and setting of the sun; the agricultural year, with its particular moments of dearth, plenty, and stress; the birth and death of individuals in the community. All these moments were marked, and rituals were remembered and repeated which reaffirmed their centrality to life. The death of a member of the community was a moment for the remembrance of the dead, and the hope for the continuity of the living. I have no doubt that death ritual was an act in which Turner’s moments of liminality and communitas were prominent, but I do want to repeat my belief that burial in the formal necropolis at Osteria dell’Osa was not the natural right of all members of the community, but the privilege of the leading kinship groups or gentes.

26

Dead Dogs and Rattles

At the burial, the mourners came out from the place in which they lived to an area which was given over to the dead, which would never be visited without the recollection of former rituals, former deaths, and which would increasingly come to reflect the growth and development of the communities which used it. They would also have come with their own memory of the deceased, and this would have been reinforced by the markers which were placed in the grave. These included personal objects of value, but also status markers. The very nature of the burial, whether cremation or inhumation, was itself a clear marker of status. In some of the burials, as we have seen, objects were placed in the grave, and ritually broken to make the fill, and this signalled a special rôle for a deceased man, a rôle which seems intimately connected with ritual. By acknowledging in this rite of deposition the special status of male heads of family in the ritual process, the mourners reaffirmed that that status should continue, perhaps being handed on to another member of the family. The ritual knowledge which the deceased had does not die with him, but survives, and is itself a mark of status for the new possessor. Some time at the end of Period II, a special sacrifice took place. This was the deposition of an adult dog. This cannot have been a random act; not only is it unique in an unusually well-known necropolis, it also took place at a significant border. The sacrifice will have taken place at a particular time, and I would suggest that April, or the spring period, would be the most natural given the associations which we have seen elsewhere. The animal chosen was significant; it was different. It was associated, as we have seen, with crop success and with the deities of the earth. Those attending the sacrifice will have moved from their homes, through their land to the liminal space of the necropolis and brought their knowledge of the annual cycle of the year and its attendant problems. Their world was one structured by hierarchical assumptions, which were tested against the experience of sacrifice, almost certainly conducted by a male head of family or head of clan, acting for the community. The ritual moment is a timeless one, but the social structure and the material world which surround the event are also part of the real world: a necropolis confined to an élite, a ritual act which is in the knowledge of the same élite. This cultural knowledge was brought by the participants to the ritual, tested and reaffirmed by the ritual, and then brought back into the world of routine life. Some of these modalities changed over time. In particular, the knife seems to have become a non-cult object in Period III, and was no longer

Dead Dogs and Rattles

27

miniaturized. It is possible that the presence of a ritual object in graves became vestigial rather than indicative in this period, as part of a distinction between ritual as the province of the living as separate from the world of the dead. On the other hand, the knife became more common in female tombs, and this may indicate the beginnings of a shift away from male control of ritual and cult, as may the Osteria dell’Osa rattle. This might be supported by the evidence from votive deposits in Rome and elsewhere. The votive deposit represents a different time-space field of activity, and a different kind of discourse (Smith 1994b). In the seventh and sixth centuries, as burial grounds became less frequented in an orderly fashion, the areas were reclaimed for a more private kind of religion, which may in fact have been less exclusive in terms of sacred knowledge. The finds from votive deposits are qualitatively similar to those from burials, but show a preponderance of female offerings (Bartoloni 1987). There is no evidence that offerings could only be made at a certain time, and it seems to me that the rhythm of offering might have been significantly different in votive deposits, in that it might have followed the rhythm of an individual’s life rather than the more immutable rhythms of the agricultural year. The rising importance of the votive deposit may be seen as the acquisition of a kind of knowledge for the individual from the specialists: the truth against which the agent measeures her/his experience is not the hierarchical truth, but the private truth of experience. That some finds are indubitably high-status objects does not invalidate the point. This may not be the case for the deposit at Osteria dell’Osa, which does not represent the frequent return of individuals to the same place but a single act of deposition. Since it is at the edge of the necropolis, and dateable to the period of the last burials at the site, the excavators have suggested that it might represent the closing of the cemetery. If this is correct, it is interesting that the deposit here is so different from those at Rome; more ordered and organized. It was a communal act, less accessible to the individual. Immediately after the closure of the necropolis, it seems that all the communities around the Lago di Castiglione coalesced on the site of Gabii, which has a seventh-century votive deposit and a sixth-century temple on top. This represents communal and controlled activity, and even the votive deposit reflected the social conditions of control (Bietti Sestieri 1992a: 49–50; Almagro-Gorbea 1981; 1982). The building of temples follows the rise of the votive deposit. Here again, we enter a different world of experience. Temples were very

28

Dead Dogs and Rattles

different in appearance from the preceding huts; although built in tufa, which is quite a soft coloured stone, they were nonetheless clearly artificial rather than natural monuments. They represented the control of energy for building them; for some time they will have been incomplete, allowing for the long experience of seeing something new on the landscape. Temples do not necessarily work in the same way as votive deposits, for although they contained votive deposits, and, as far as we know, could be visited at any time, they were monumentalizations for a certain deity, and had festivals which focused on them once a year. They involved priests as well, and here we see the return of certain special knowledge, marked visibly. The control of votive deposits by temples is spatial in that a temple is axial and has one entrance, and focuses attention onto a cult image of the deity. What we know of the decoration suggests the visual presentation of a hierarchy of deities, from the antefixes, elaborate heads, to the full-size figures of the deities (Rystedt, Wikander, and Wikander 1993). Elsewhere I have characterized temples as the redirection of wealth by an élite from the temporary and transient act of a funeral, to the permanently visible creation of a building (Smith 1996: 185ff.). I wish to deepen this analysis here. The temple takes its place as one of the markers of the developed urban state, as opposed to the small settlements which were served by the necropolis at Osteria dell’Osa. Within the urban state, there are indications of a rigidified or crystallized social hierarchy, patriarchal in nature, and focused on the clans or gentes who were the favoured associates of the kings. It is possible that the religion of early Rome was strongly dominated by these clans, for whom the religious rites were a kind of private knowledge, perhaps distributed by the king, or claimed as an ancestral possession, as Harrison suggested for other forms of ritual knowledge. Within the growing states, the temples were a means for the élite to authorize its position with reference to a higher truth. The temples were probably built by free members of the community—there is little evidence for slavery at this stage. To attend a ritual at one of these temples was to enter a world very different from the acculturated world of the ordinary house (though significantly less different from the acculturated world of the élite house). Whereas previous rituals had taken place in the open, these rituals had an inbuilt opposition of outside, where the people stood, and inside, where only the priest and his acolytes could enter. There is another aspect to the urbanization of ritual activity. The sacrifice of a dog to prevent mildew may have begun as a local celebration, but it seems, as I have said, to have connections with boundaries. The evidence of Ovid, however, shows that this ritual has become more

Dead Dogs and Rattles

29

elaborate, that it keeps pace with the military expansion of Rome from the hills out into the territory which was progressively conquered by Rome. If Ovid met the procession of the Robigalia five miles along the Via Nomentana, he would have been just by the town of Ficulea, which Livy (1.38.4) tells us was conquered by Tarquinius Priscus early in the sixth century. The experience of participating in the festival was therefore not simply that of a small community protecting its own crops and fields, but of an urban state laying claim to land which it had won.

A FT E R W O R D Osteria dell’Osa remains unparalleled in archaic Latium for its size and relative completeness of excavation. Further work has been conducted at Gabii (The Gabii Project) and also at S. Omobono (The Sant’Omobono Project | Faculty, Lab, and Research Template Site (umich.edu)). The evidence base for the archaic period continues to grow, and the most recent substantive summary for Rome is the exhibition catalogue, Damiani (ed.) 2019. On dogs in burials and sacrifice, see now Jean-Claude Lacam (Lacam 2008: 29–80). On rattles, see the helpful survey, by Rystedt 1992: 125–33. Our understanding of the relationship between deposits, temples, and urban foundations has also evolved. For the archaic period, Potts 2015 is the complete survey, while Terrenato 2019 offers the most compelling reconstruction to date. My own article contains some casual gender assumptions around priesthood which I would now substantially qualify; and the discussion of ritual has developed apace. See Swenson (2015: 329–45) and Nizzo (2018) for a substantial collection of essays.

B I B L I OG R A P H Y Almagro-Gorbea, M. (1981). ‘L’Area del tempio di Giunone Gabina nel VI–V sec a.C.’ Archeologia Laziale 4: 297–304. Almagro-Gorbea, M. (1982). El Santuario del Juno en Gabii. Rome. Asad, T. (1983). ‘Anthropological conceptions of religion: reflections on Geertz’. Man 18: 257–9. Barrett, J. C. (1988). ‘Fields of discourse: reconstituting a social archaeology’. Critique of Anthropology 7: 5–16. Barrett, J. C. (1991). ‘Towards an archaeology of ritual’, in P. Garwood, P., D. Jennings, R. Skeates, and J. Toms (eds.), Sacred and Profane: Proceedings of a Conference on Archaeology, Ritual and Religion (Oxford 1989). Oxford University Committee for Archaeology, Monograph 32. Oxford, 1–9.

30

Dead Dogs and Rattles

Barrett, J. C. (1994). Fragments from Antiquity: An Archaeology of Social Life in Britain, 2900–1200 . Oxford. Bartoloni, G. (1987). ‘Esibizione di richezza a Roma nel VI e V secolo: Doni votivi e corrredi funerari’. Scienze dell’Antichità: Storia Archeologia Antropologia 1: 143–59. Bartoloni, G. (1989–90). ‘I depositi votivi di Roma arcaica: Alcune considerazioni’. Scienze dell’Antichità: Storia Archeologia Antropologia 3–4: 747–59. Bartoloni, G., M. Cataldi Dini, and F. Zevi (1982). ‘Aspetti dell’ideologia funeraria nella necropoli di Castel di Decima’, in G. Gnoli and J.-P. Vernant (eds.), Le Mort, les Morts dans les sociétés anciennes. Cambridge, 257–73. Bianco Peroni, V. (1976). I coltelli nell’Italia continentale. Prähistorische BronzeFünde VII/2. Munich. Bietti Sestieri, A. M., C. Ampolo, et al. (1980). La Formazione della città nel Lazio. Dial.Arch. n.s. 2: 125–232. Bietti Sestieri, A. M., A. De Santis, and A. La Regina (1989–90). ‘Elementi di tipo cultuale e doni personali nella necropoli laziale di Osteria dell’Osa’. Scienze dell’Antichità: Storia Archeologia Antropologia 3–4: 65–88. Bietti Sestieri, A. M. (1992a). La necropoli laziale di Osteria dell’Osa. Rome. Bietti Sestieri, A. M. (1992b). The Iron Age Community of Osteria dell’Osa: A Study of Socio-political Development in Central Tyrrhenian Italy. Cambridge. Bloch, M. (1986). From Blessing to Violence: History and Ideology in the Circumcision Ritual of the Merina of Madagascar. Cambridge. Bremmer, J. (1983). ‘Scapegoat rituals in ancient Greece’. HSCPh 87: 299–320. Bourdieu, P. (1977). Outline of a Theory of Practice. Cambridge. Carettoni, G., A. M. Colini, L. Cozza, and G. Gatti (1960). La pianta marmorea di Roma antica. Rome. Chiarucci, P. and T. Gizzi (1985). Area Sacra di Satricum: Tra scavo e restituzione. Rome. Coarelli, F. (1988). Il Foro Boario. Rome. Colonna, G. (1988). ‘La produzione artigianale’, in A. Momigliano and A. Schiavone (eds.), Storia di Roma I: Roma in Italia. Rome, 292–317. Cornell, T. J. (1989). ‘Rome and Latium to 390 B.C.’, in Cambridge Ancient History, vol. VII.2, 2nd edn. Cambridge, 243–308. Damiani, I., ed. (2019). La Roma dei Re. Il racconto dell’archeologia. Rome. Editorial group (1989). Il Viver Quotidiano in Roma Arcaica: Materiali dagli scavi del Tempio Arcaico nell’area sacra di S. Omobono. Rome. Fustel de Coulanges, N. D. (1980). The Ancient City: A Study on the Religion, Laws and Institutions of Greece and Rome. Baltimore. Gibbon, G. (1989). Explanation in Archaeology. Oxford. Giddens, A. (1984). The Constitution of Society: Outline of the Theory of Structuration. Cambridge. Gjerstad, E. (1966). Early Rome IV: Synthesis of Archaeological Evidence. Acta Instituti Romani Regni Sueciae, Series in 4 XVII.4. Lund. Harrison, S. (1992). ‘Ritual as intellectual property’. Man 27: 225–44.

Dead Dogs and Rattles

31

Heurgon, J. (1959). ‘The Date of Vegoia’s prophecy’. JRS 49: 41–5. Hodder, I. (1991). Reading the Past: Current Approaches to Interpretation in Archaeology, 2nd edn. Cambridge. Humphreys, S. C. (1978). Anthropology and the Greeks. London. Kurtz, D. C. and J. Boardman (1971). Greek Burial Customs. London. Lacam, J.-C. (2008). ‘Le sacrifice du chien dans les communautés grecques, étrusques, italiques et romaines: approche comparatiste’, in MEFRA 120.1: 29–80. Lewis, G. (1980). Day of Shining Red: An Essay on Understanding Ritual. Cambridge. Lewis, G. (1994). ‘Magic, religion and the rationality of belief ’, in T. Ingold (ed.), Companion Encyclopedia of Anthropology: Humanity, Culture and Social Life. London, 563–90. Malone, C. and S. Stoddart, eds. (1994). Territory, Time and State: The Archaeological Development of the Gubbio Basin. Cambridge. Morris, I. (1987). Burial and Ancient Society: The Rise of the Greek City-State. Cambridge. Nizzo, V., ed. (2018). Archeologia e antropologia della morte: 2. Corpi, relazioni e azioni: il paesaggio del rito. ESS. Rome. North, J. A. (1989). ‘Religion in Republican Rome’, in Cambridge Ancient History, vol. VII.2. 2nd edn. Cambridge, 573–624. Palmer, R. E. A. (1970). The Archaic Community of the Romans. Cambridge. Porte, D. (1985). L’Etimologie Religieuse dans les Fastes d’Ovide. Paris. Potts, C. (2015). Religious Architecture in Latium and Etruria, c.900–500 . Oxford Monographs on Classical Archaeology. Oxford—New York. Poultney, J. W. (1959). The Bronze Tables of Iguvium. Philological Monographs of the American Philological Association 18. Baltimore. Renfrew, C., ed. (1985). The Archaeology of Cult. The Sanctuary at Phylakopi. London. Renfrew, C. and P. Bahn (1991). Archaeology: Theories, Methods and Practice. London. Renfrew, C. (1994). ‘The archaeology of religion’, in C. Renfrew and E. Zubrow (eds.), The Ancient Mind. Elements of Cognitive Archaeology. Cambridge, 47–54. Robertson Smith, W. (1889). Lectures on the Religion of the Semites. Edinburgh. Rystedt, E. (1992). ‘Notes on the rattle scenes on Attic geometric pottery’. OpAth 19: 125–33. Rystedt, E., C. Wikander, and O. Wikander, eds. (1993). Deliciae Fictiles: Proceedings of the First International Conference on Central Italian Architectural Terracottas at the Swedish Institute in Rome, 10–12 December 1990. Acta Instituti Romani Regni Sueciae, Series in 4 L. Stockholm. Scullard, H. H. (1981). Festivals and Ceremonies of the Roman Republic. London. Shanks, M. and C. Tilley (1992). Re-Constructing Archaeology: Theory and Practice, 2nd edn. London. Smith, C. J. (1994a). ‘A review of studies on Iron Age and archaic Rome and Latium’. JRA 7: 285–302.

32

Dead Dogs and Rattles

Smith, C. J. (1994b). ‘Ritual and archaeology in early Latium’, in Papers of the Fourth Theoretical Archaeology Conference. Oxford, 57–64. Smith, C. J. (1996). Early Rome and Latium c.1000 to 500 : Economy and Society. Oxford. Swenson, E. (2015). ‘The archaeology of ritual’. Annual Review of Anthropology 44.1: 329–45. Terrenato, N. (2019). The Early Roman Expansion into Italy: Elite Negotiation and Family Agendas. Cambridge. Turner, J. H. and A. Maryanski (1979). Functionalism. Menlo Park—London. Turner, V. W. (1967). The Forest of Symbols: Aspects of Ndembu Ritual. Ithaca. Turner, V. W. (1969). The Ritual Process: Structure and Anti-Structure. London. Warde Fowler, W. (1899). Roman Festivals of the Period of the Republic. London. Wille, G. (1967). Musica Romana: Die Bedeutung der Musik im Leben der Römer. Amsterdam.

2 The Priest and the Magistrate Reflections on the Priesthoods and Public Law at the End of the Republic John Scheid

As Pierangelo Catalano has shown,1 the Romans and historians of Rome refer to two main conceptions of the division of power, one determined by Roman tradition, the other by Greek thought. As attested in Cicero, and through into the Empire, canonical opinion tended to divide the ius publicum into two complementary yet hierarchical realms: the ius of the sacra and the sacerdotes, and the ius of the magistrates. It is well known that Cicero, in his Laws, first deals with religion, from the dual perspective of laws pertaining to the sacra and to the sacerdotes, before turning to the magistrates, an order which he defines according to the principle: ‘id est [the magistrates] quod constituta religione rem publicam contineat maxime’ (2. 69).2 This hierarchy is very clearly reproduced in the definition of public law given by Ulpian: Publicum ius in sacris, in sacerdotibus, in magistratibus consistit (D. 1. 1. 1. 2). We can obviously not say with any certainty that Cicero reduced all public law to these two categories, as only three books (out of five?) of De legibus have been preserved. But in any case— and this is sufficient for our purposes here—in his presentation of the supreme laws of the ideal republic Cicero outlines (quod contineat maxime) two types of legislation providing the basis for the two main aspects of power, religious laws and laws concerning magistrates, in accordance with a legitimately hierarchical order that corresponds on

1

Catalano 1974: 667–91.

2

For Cicero’s sources see Catalano 1974: 670.

John Scheid, The Priest and the Magistrate: Reflections on the Priesthoods and Public Law at the End of the Republic In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0003

34

The Priest and the Magistrate

every point to that of Ulpian.3 A further division, which we see in a different context in Quintilian4 and already in Cicero, extends out of this conception. This does not formally contradict the traditional canonical order but expands it on a point of detail which we will come back to. For the moment, let us just note that it is still the leges sacri iuris that take precedence. For Polybius on the other hand, city power is split into three parts, the consuls and other magistrates, the senate, and finally the masses.5 Catalano has shown how Polybius leaves out of this division the priests, who are studied elsewhere,6 and includes amongst the magistrates the plebeian tribunes, while denying them any autonomy.7 Notwithstanding this oversight and distortion, the Polybian model exerted considerable influence, even on a thinker like Cicero, who, while re-establishing a division more in keeping with Roman tradition, reaffirms the model of the mixed constitution.8 For Catalano,9 there are three reasons for the partition made by Polybius: the impossibility of getting a specifically Roman world to fit with a Greek constitutional schema, the fact that in this period the priests and the plebeian tribunes tend to lose their ‘autonomous’ role, and finally the desire to reaffirm (against the wishes of the populares) the integration of the tribunate into the framework of the magistrates, putting aside two aspects of the ‘Roman constitution’ whose powers and modes of action were entirely opposed to the moderate design of the mixed constitution. Suggestive as they may be, these theories inevitably bring with them a number of problems, two of which are fundamental. It has not escaped Catalano’s attention that given the state of our current understanding, it

3 We find the same kind of thinking in Valerius Maximus, 1. 1. 9: omnia namque post religionem ponenda semper nostra ciuitas duxit, etiam in quibus summae maiestalis conspici decus uoluit; see also the preface, 5, and the very order of the early chapters in the work of Valerius Maximus. 4 Quintilian, Institutio oratoria, 2. 4. 33–4: Nam et genera sunt tria [i.e. legum], sacri, publici, priuati iuris. Quae diuisio ad laudem magis spectat, si qui seam per gradus augeat, quod lex, quod publica, quod ad religionem deum comparata sit; Cicero, De domo sua, 12. 32: (on his cause and the inquiry of the pontiffs) quae cum sit in ius religionis et in ius rei publicae distributa, religionis partem, qua multo est uerbosior, praemittam, de iure rei publicae dicam. Clearly, these two passages are only of relative value for our purposes here, as the context and intention are different. Lastly, let us highlight a passage by Ausonius, which, as Catalano has pointed out, concerns the same representations: Ausonius 336, 61ff. (p. 203 Peiper), Ius triplex, tabulae quod ter sanxere quaternae: / Sacrum, priuatum et populi commune quod usquam est. 5 6 Polybius 6. 11ff. Polybius 6. 1. 9 = 21. 10 (13). 11. 7 8 9 Catalano 1974: 668–9. Catalano 1974: 670. Catalano 1974: 669.

The Priest and the Magistrate

35

is difficult to know whether other divisions, other theories, also existed.10 But a little history is better than none. Let us accept, then, the fact that we know of at least two conceptions of the division of power—and what’s more, famous ones—and establish the principle that other representations, just as rational and as warranted, would doubtless also have existed. The conclusions drawn will thus be relative and will only reflect one aspect of Roman thought, but then when can we ever hope to encompass the full gamut of opinion? With this doubt settled, another difficulty now emerges. Should the theories we know about be thought of as secondary elaborations, representations (traditional or otherwise) that have no relation to deep-lying structures, assembled out of a selective wish to justify a certain vision of the political and social order, as is the case for example in Polybius? Or rather, in spite of this superficial bias, do these schemas reveal the institutional and conceptual Roman system as such, the historical product of a timeless ideology? Let us straightaway admit that this age-old problem will not be resolved here, and that we can only take a few small steps towards the light. Just in passing will we consider the Polybean theory. Its historical foundations have often been examined, and may in any case, according to Catalano’s analysis, result in a less close-up experience of Roman institutions or concepts, or rather in a political and philosophical vision different from that of Cicero, for example, even if it does not fundamentally contradict it. We will retain here the terms of Ciceronian theory, and in the light of these concepts examine an order of reality less determined by subjective re-elaborations. While this institutional reality may itself also to a large extent be founded on literary sources, it would nonetheless avoid the criticism we have already addressed to the political theories. And if this comparative approach to two complex realities, not pertaining to the same order, were to show up familiar kinds of convergence or divergence, we would be in a position to appreciate the degree of fiction contained in the political theories considered and to define their relation with the structure as well as with the process of historical development. We will allow, then, that all aspects of supreme power can be split into two complementary areas, the sacrum and the publicum. By examining the prosopographical series of priests in the Republican period we will thus seek to establish what place these priests held in the respublica, and what was their relation to the magistrates, who held the other part of supreme power.

10

Catalano 1974: 676–7. Cf. also Kaser 1949: 74ff.

36

The Priest and the Magistrate

With the dissertation of Carl Bardt11 on the priests of the four major colleges of the Republic, studies concerning the Roman sacerdotal institution left the realm of antiquarian speculation and fully entered that of historical inquiry. Instead of using such and such a document in order to justify a more or less naïve reconstruction of origins, Bardt made it his goal to study the priesthoods of the historical period in their own right and in their entirety. Putting together all the data, he reconstituted the fasti of the various colleges and asked questions of a historical nature, in relation to historical reality, rather than to the representations given by ancient or modern writers. His conclusions were raised to a more general level by Theodor Mommsen in Roman Public Law, notably in the pages devoted to the pontifex maximus,12 where for the first time we encounter an in-depth analysis of the relations existing between priesthood and magistracy, while—in a good example of the combinatory method—the sacerdotal institution is used capture the spirit of Roman public law. Ever since, scholars have unceasingly taken the path mapped out for them, studying the priestly world from two complementary perspectives. One approach followed Bardt’s endeavour to establish the complete and exact fasti; this reached a provisional conclusion with the most recent study to have appeared, G. J. Szemler’s Priests of the Roman Republic.13 The other approach, begun by Mommsen, consists in using all the information to emerge from prosopographical inquiry and the study of sources in order to get to the religious core and to fathom out the relations between the various different priesthoods; or rather—and this often amounts to the same thing—so as to reconstruct the institution in its oldest form. The work of Georges Dumézil14 should also be cited here on the organizing ideology of behind the most ancient representations of Roman religion, which provide an insight into the historical structure of the major priesthoods, and likewise—as we often forget—the priesthoods’ profound character. In these studies, the relations obtaining between the priesthood and the magistracy, as well as the specific place occupied by the priest in the Roman Republic and society, have been subject to much fine-tuning. The judicial positions held by priests have been brought out, the grip exerted by the senatorial elite on the major priesthoods highlighted, and the work of certain great plebeian priesthoods uncovered.15 In short, there is

11

12 Bardt 1871. Mommsen 1887: 2. 1. 16–70. Szemler 1972, with bibliography. To which may be added: Szemler 1976: 53–8; 1974: 72–86; 1978: cols. 331–96; Parrish 1977: 623–33. 14 Dumézil 19742; 1970 and passim (for the flamens); (opposition between flamens and pontiffs). 15 See in particular Szemler 1972; Richard 1968: 786–801. 13

The Priest and the Magistrate

37

little that we do not know about the normal cursus and activity of a priest. Just one small detail has not received the attention which to my mind it deserves, one which may turn out to be important insofar as it relates directly to the deep-lying structures of public religion and its place in the ‘constitution’ of the Roman city. This concerns the social rank of priests, and in particular how it develops from Sulla on, finding its conclusion in the reign of Augustus. For Mommsen, for example, every citizen had under the Republic the right to apply to become a judge or a priest, other than in the case of some patrician priesthoods, which were few in number and without great political importance. Even if the facts say otherwise, no form of censitary qualification was required by law in order to seek these honours. Under Augustus, on the other hand, recruitment to both priestly and judicial authorities was governed by these kinds of criteria, and priesthoods, for instance, were shared out amongst the members of the supreme orders.16 This theory, which has been refuted by Claude Nicolet17 where judicial recruitment is concerned, has nonetheless never been questioned, as far as I am aware, in respect of access to the priesthood.18 To be sure, G. J. Szemler19 notes how it was well known that not all the priests of the Republican period were senators, but like his predecessors he does not take the observation further and devotes himself instead to the priests who effectively sat in the senate. This problem is a fundamental one in terms of what we are looking at here, for the examination of the social status of priests in the light of recent work on Roman orders20 makes it possible to deepen our understanding of the relations between priesthood and magistracy, ius sacrum and ius publicum, in short to respond the questions raised at the beginning of this study.

TO WHAT ORDER DID THE ROMAN PRIESTS BELONG? Before answering this question, the first thing to say is that the priests do not form an ordo comparable to that of the knights or senators. The sources, which are fairly abundant, have not left the slightest hint of any censitory requirement for the election of a priest, and this fact alone proves that there never was any ‘priestly order’. The ‘ordo sacerdotum’ 16 18 19 20

17 Mommsen 1887: 3. 1. 566–7. Nicolet 1976c: 20–38; 1977: 365–6. Wissowa (19122: 479ff.) takes up Mommsenian thinking. Szemler 1972: 76. Szemler 1972: 76; Chastagnol 1971: 282–310; 1973: 583–607; Nicolet 1976: 20–38.

38

The Priest and the Magistrate

evoked by Festus21 should not deceive us here. It refers not to a specific ‘order’, reserved for public priests, but to the hierarchy of priests that make up the pontifical college: in the ‘order’ were the rex, the flamines Dialis, Martialis, and Quirinalis, and the pontifex maximus. So to what order do the priests belong? The answers put forward up until now, which all derive from Mommsenian theory, are neither entirely correct nor entirely false, and therefore do not provide a full picture of the situation. In any event, we are unable to give a single, unifying answer that would be valid for all Roman history, in that Roman thinking fluctuates between two conceptions of the priest, which, as we will see later on, come to light in various different ways in one period or another. But let us have a look at the data first. From Augustus on, the facts are clear. The public priesthoods are split into two groups. One is senatorial and includes the priests of the four major colleges, the salii, the Arval Brethren, the fetiales, the sodales Titii, the curio maximus, and later on the sodalities of the diui.22 The other is equestrian, with the pontiffs and the minor flamens, the luperci, the curiae, the Laurentes Lavinates, and the members of the other ancient Latin priesthoods.23 This very clear division revealed by prosopography24 includes some exceptions (spread over three centuries) which

21 Festus, 198–200 L: Ordo sacerdotum aestimatur deorum maximus quisque. Maximus uidetur rex, dein Dialis, post hunc Martialis, quarto loco Quirinalis, quinto pontifex maximus. Itaque in soliis rex supra omnis accumbat licet; Dialis supra Martialem et Quirinalem; Martialis supra proximum; omnes item supra pontificem. Rex, quia potentissimus; Dialis, quia universi mundi sacerdos, qui appellatur Dium, Martialis, quod Mars conditoris urbis parens; Quirinalis, socio imperii Romani Curibus ascito Quirino; pontifex maximus, quod iudex atque arbiter habetur rerum diuinarum humanarumque. See Wissowa 19122: 504 and Dumézil 1974₂; 1970. We should not confuse this ordo, the hierarchy of rex, flamines, and pontifex maximus with the ordo haruspicium LX, which certainly is a social order. See Wissowa (19122: 548), whose views on the nature of this order have been corrected by Rawson (1977: 140). 22 Mommsen 1887: 3. 1: 567ff.; Howe 1903: 8; Wissowa 19122: 74, 492. 23 For the Latin priesthoods see the excellent study by Wissowa 1915: 1–33. 24 There is no text that gives us any information on this reform. It may be possible to see a direct allusion to this construction in the words of Claudius, who presented the children of a knight of Vienna as follows: ILS 212. II. 1. 12–14. L. Vestinum . . . cuius libe/ri fruantur quaeso primo sacerdotiorum gradu, post modo cum / annis promoturi dignitatis suae incrementa; ‘may his children, I entreat you, enjoy the 1st rank of priesthoods, so that with the years they may further the growth of their dignity’. Commentators have understood this primus gradus in different ways. For E.-G. Hardy (1913: 81), it means the most lustrous priesthood; whereas Fabia (1927: 103) opts for the lowest priesthood, i.e. the first, without giving a clearer indication as to why. Fabia’s view will be the one taken here. In the end, though, whichever interpretation we decide to adopt, for the present purposes the result is the same: this ‘first rank of priesthoods’ proves the existence of ranks in an area where precisely there were none before Augustus.

The Priest and the Magistrate

39

Mommsen was easily able to account for25 by invoking the status of priests at the point that they were co-opted. This may involve a senator co-opted before the quaestorship, that is, when he still belonged to the equestrian order; a member of the senatorial order who had never held or been able to hold a magistracy; or else a knight incorporated into the senate having obtained an equestrian priesthood. We do not possess any information that would allow us to decide whether the change in the personal status of the priest carried any kind of consequence at a sacerdotal level. If the facts, then, are pretty clear during the Empire, the data for the Republican period is less obvious. G. J. Szemler has rightly highlighted the difference in the status of priests, and in particular their very variable age at the point of their co-option. But his commentary is focused more on the political position of priests, and their political power, seeing the priesthoods obtained early on in a life as ‘the means utilized by the nobiles as a training ground for potential advancement’.26 His approach is entirely justified and there is nothing else to add to the results of this inquiry. Priesthoods are in general presented as the inseparable complement, or rather as the opening, to a brilliant career. One aspect of these co-optations has nonetheless been left out by Szemler: the rank held by priests at the point when they are co-opted, and for some, their whole life. Before developing the implications of this situation as it is set out, we first need to draw up an inventory of cases attested by the sources. Three scenarios may be outlined: either no magistracy is attested for the priest in question, or he became a magistrate after his co-optation, or else he was co-opted after having been a magistrate. This last category is not of any interest to us in the present context, as the priests who were co-opted after holding one of the curule magistracies, or later on one of those which qualified them to be placed on the senatorial album, belonged to the senatorial order even before their co-optation. Thus, insofar as their case does not present a problem and as a rule does not reveal any discernible difference from the senator-priests of the Imperial period, we will only draw up a table of the first two groups mentioned. We will come back later to some of those priests belonging to the third category, whose example will support the conclusions reached from the second.

25 26

For examples: Mommsen 1887: 3. 1. 568, nn. 3, 7; Howe 1903: 8. Szemler 1972: 192, 56.

40

The Priest and the Magistrate

Given the uncertainty of our information, we will only examine those cases we can be sufficiently sure about, which does not hold us back in our inquiry, as a few examples may suffice to buttress our interpretation of the facts. We will leave aside the Early Republic, where the state of our sources does not really allow us to form reliable conclusions.27 Let us first draw up the inventory for all Roman priests that we know of between 300 and the death of Caesar who are not attested as magistrates.28

A. Pontifical College Reges sacrorum 1. L. Postumius L. F. Albinus, ca. 275 (p. 68, n 9); 2. M. Marcius, ?–210 (p. 174, n 1); 3. Cn. Cornelius Dolabella, 208–180 (p. 175, n 2); 4. L. Cloelius Siculus, 180–? (p. 175, n 3); 5. [Sul]picius Ser. f., perhaps ca. 63 (p. 175, n 5). Flamines 6. P. Cornelius P. f. Sulla, fl. Dialis, ca. 250 (p. 68, n 13); 7. C. Claudius, fl. Dialis in 211 (p. 74, n 36); 8. Cn. Cornelius, Fl. Dialis, 174–? (p. 168, n 6); 9. Ti. Venturius Philo, fl. Martialis, 204–? (p. 167, n 3); 10. [Q?] Sulpicius, fl., ?–223 (p. 70, n 21). Pontiffs 11. Q. Aelius Paetus, ?–216 (p. 71, n 27); 12. P. Scantinius, ?–216 (p. 71, n 29); 13. C. Sulpicius Galba, 202–199 (p. 110, n 11); 14. P. Mucius Scaeuola, 73/69–ca. 60 (p. 132, n 51); 15. L. Pinarius Natta, ca. 58–56 (p. 135, n 64). Minor Pontiffs 16. L. Cantilius, scriba pontificis in 216 (p. 71, n 26); 17. P. Albinouanus, before 68–after 57 (p. 133, n 53); 18. Q. Cornelius, in 57 (p. 133, n 54).

27

Szemler 1972: 47, 62. References to the corpus of Szemler, on the basis of which this list is established, are given in parentheses. 28

The Priest and the Magistrate

41

B. Augurs 19. 20. 21. 22. 23. 24. 25.

C. Genucius (Augurinus), 300–? (p. 66, n 6); M. Minucius Faesus, 300–? (p. 66, n 7); T. Publilius, 300–? (p. 66, n 8); Q. Fabius Maximus, 203–196 (p. 140, n 8); P. Cornelius Scipio (Africani f.), 180–? (p. 142, n 13); Ti. Veturius Gracchus Sempronianus, 174–? (p. 143, n 15); P. Licinius Crassus, ca. 55–53 (p. 152, n 39).

C. X(V)uiri s.f. 26. 27. 28. 29.

M’. Aemilius M’. f. (Numida), 236 (?)–211 (p. 69, n 17); M. Liuius M. f. Salinator, in 236 (?) (p. 69, n 18); C. Sempronius Longus, 174–? (p. 161, n 13); L. Valerius, in 76 (p. 165, n 25).

D. Epulones 30. C. Licinius Lucullus, 196–? (p. 172, n 1); 31. Q. Fuluius M. f., 180–? (p. 173, n 4).

E. Curio maximus 32. M. Aemilius Papus, ?210 (p. 175, n 1).

F. Luperci 33. 34. 35. 36. 37. 38. 39. 40.

L. Herennius Balbus, ca. 56 (p. 178, n 2); Q. Tullius Cicero, ca. 46–43 (p. 178, n 3); Geganius Clesipus, mag. lupercor. (p. 178, n 4); Statius, 46–? (Cic., ad. Att., 12, 5, 1); Philotimus, 46–? (Cic., l.l.); Q. Considius Q. L. Eros, Caesarian period (?) (CIL, VI, 1933); C. Iulius Caesaris l. Saluius, Caesarian period (?) (CIL, XI, 7804); C. Curtius Post(umi). l. Helenus (CIL, VI, 32437).29

29 Szemler does not include these five luperci. For Statius and Philotimus, we know that they became luperci at the same time as Q. Tullius Cicero. For Clesipus, C. Iulius, and

42

The Priest and the Magistrate

For none of these priests is it attested that they held a magistracy or belonged to the senate. There is a particular problem posed by a certain number of priests, for while they may be known to have managed lesser magistracies, it is not clear, and it is unlikely, even, that they belonged to the senate.30 Despite this, we are not including them in the first group of priests, and with the benefit of the doubt put them in the second group. This concerns the augur and (quin)decimvir (C.) Coelius Caldus (p. 150, n 32; p. 164. n 22) who was a moneyer; the decemviri C. Papirius L. f. Maso (p. 72, n 31, Xuir ?–213, IIIuir agr. d. ass. in 218), M. Aurelius Cotta (p. 160, n ; 204–200, aed. pl. 216), M. Opimius (p. 164, n 20; ca. 125–120, moneyer). Here now is the list of priests who were co-opted well before their first (known) magistracy.

A. Pontifical college Flamines 42. C. Valerius Flaccus, fl. Dialis 209–before 174, aed. cur. 199, pr. 183 (p. 166, n 1); 43. L. Postumius Sp. f. L. n. Albinus, fl. Martialis 168–154, aed. cur. 161; pr. 157 (p. 169, n 7); Pontiffs 44. Cn. Cornelius Scipio Hispallus, 199–176, pr. 179 (p. 112, n 13); 45. M. Aemilius Lepidus, ca. 60–12, cos. 46 (p. 134, n 63).

B. Augurs 46. Q. Fabius Maximus Verrucosus, 265–203, quaest. before 237, aed. cur. 237 (?) (p. 68, n 10);

Q. Considius, we do not know if they belonged to the Republican period or to the beginning of the Empire. Mommsen (CIL, I2, 1004), Wissowa (19122: 491, n. 6), and Klose (1910: 57) put Clesipus in the Caesarian period; Mommsen (1887: 3. 1. 566, n. 3), Klose and Wissowa also add in here C. Iulius and Q. Considius, on the basis that these priests were originally freedmen, proving in their view that we are in the period of the ‘Verwirrung vor der augusteischen Reform’. This is the dating we are adopting here. As for C. Curtius Post(umi) l. Helenus, Treggiari (1969: 196, n. 4) suggests that he should be considered as a freedman of Curtius Postumus, also of the Caesarian period. As is often admitted, it is difficult to decide whether all these luperci were Julian luperci or not. For our purposes, this problem does not matter so much; at most we may note that even for a new priesthood the traditional rules were respected. 30

For recruitment to the senate see Nicolet (1977: 363–9).

The Priest and the Magistrate

43

47. L. Quinctius Flamininus, 213–170, aed. cur. 201, pr. urb. 199 (p. 143, n 16); 48. Ti. Sempronius Longus, 210–174, tr. pl. 200, aed. cur. 198, pr. 196 (p. 137, n 2); 49. Ti. Sempronius Gracchus, 204 (?)–174 (?), tr. pl. 187 or 184 (?), aed. cur. 182, pr. 180 (p. 140, n 6); 50. C. Claudius Pulcher, 195–167, pr. peregr. 180 (p. 141, n 9); 51. T. Quinctius Flamininus, 167–?, pr. in 153 (p. 144, n 17); 52. M. Valerius Messalla (Rufus), augur for 55 yrs, cos. 53 (p. 154, n 45).

C. X(V)uiri s.f. 53. Ti. Sempronius Ti. f. Longus, 210–174, pr. 196 (p. 141, n 9); 54. L. Manlius Torquatus, ca. 65–46, IIIuir monet. ca. 65, pr. 49 (p. 165, n 26); 55. M. Porcius Cato, 64–46, co-opted before the quaestership (p. 165, n 27).

D. Salii 56. Ap. Claudius Pulcher, before 167, pr. 146 (p. 177, n 2). If we do not include two pontiffs for which there is insufficient evidence,31 or the seven known cumuls, it is possible to say that in total we know of 222 priests between the years 300 and 44 .32 Fortyone of these priests are not known to have held a magistracy,33 fifteen became priests well before being placed on the senatorial album, and lastly there are four priests of whom we cannot say for sure whether they belong to the first or the second group.34 The other priests, that is, Szemler 1972: 116, n 25; 120, n 27. This assessment is based on the work of Szemler which in all brings out 217 priests (36 before the year 211, and 188 between 211 and 44 ). To this number may be added the five luperci whom Szemler has not included. Of course, these figures only offer an indication; they are not absolute. The number of cases attested could be reduced or increased, depending on how one approaches the sources. 33 To these we can add A. Castricius Myriotalenti f. (CIL 14. 2105), who was a knight (Nicolet 1974: 2. 882). He belongs to the early years of Augustus’s reign. Very far behind on the prosopographical research from the beginning of the century (and somewhat casual with the Latin grammar) is the study of Simon (1981: 12–14). 34 See Nicolet 1984: 7. 31 32

44

The Priest and the Magistrate

roughly three-quarters of the cases we know about, already belonged to the senate at the point when they were co-opted. What might have been the social rank of the priests from the two categories used to draw up the list? The priests who did not hold any magistracy have to be considered knights, with the exception, as we will see later, of the flamen of Jupiter. Indeed, the vast majority of the priests were nobles, or in any case came from a senatorial family, and as the sons of senators belonged to the equestrian order.35 Likewise, the thirteen individuals who became priests very early on in life, before they could demonstrate the qualities needed to be placed on the senatorial album, were also part of the equestrian order at the point when they were coopted, insofar as their later career shows that they possessed one of the conditions for access to the senate: the equestrian census. I am not, though, claiming that a quarter of all the priests we know of between 300 and 44 were knights, at least not when they were co-opted. Any such assertion would be unwise, as from the state of our sources we cannot say with any certainty that such and such a priest really exercised a magistracy. We are sure that the priests from the first group did not hold the consulate, but some doubt remains over the curule aedileship, and from the second century on, the praetorship, the quaestorship, the plebeian tribuneship and aedileship,36 and other possibly other lesser magistracies. In any case, this is not the point. The aim here is to show that in the Republic, priests were not necessarily senators or knights according to the priesthood in question, and even that not all the priests belonged to the primi ordines. It will therefore be sufficient to find a few definite examples to obtain a clear answer to the questions raised. Such examples exist. The most obvious is provided by the Luperci Clesipus Geganius, C. Iulius Saluius, Statius Philotimus, Q. Considius Eros, and C. Curtius Post. l. Helenus, who are all freedmen. Of course, Cicero is rather sarcastic about the honour of belonging to a sodality 35

For all this, see the articles cited in n. 10. While it is unlikely that there were many unknown curule aediles among the priests attested from the Second Punic War on, in that a large number of those who held this function were known, it remains the case that the Censors may have taken account of certain other inferior functions in order to inscribe such and such a priest. This is the case, for example, with C. Licinius Lucullus (our n 30), who was a plebeian tribune in 196 and IIuir aed. dedic. in 191. For Willems (1885: 345, n 193), he belonged to the senate; however, on p. 306 the author comments—and he is quite right—that not all the tribunes were inscribed, and that in this period we can only be sure that they were senators if they had fulfilled a senatorial mission, such as that of the legate or deputy. It is therefore not at all clear that Lucullus was a senator, even if this possibility exists. A similar case is raised in respect of the plebeian aedile M. Aurelius Cotta (Szemler 1972: 160 n 9); see Nicolet (1984: 7). 36

The Priest and the Magistrate

45

where the son of a praetorian, his nephew Q. Tullius Cicero, rubbed shoulders with freedmen and bragged about it,37 and Wissowa attributes this crowding in, to the period of the pre-Augustan ‘Verwirrung’. Nevertheless, no Roman protested against this intrusion on the part of freedmen. It may have appeared telling, even humiliating, in the eyes of senators avid for exclusive distinctions and disdainful of functions which were not, but in fact this was never stated explicitly. Indeed, Cicero would certainly have contested this transgression of customs if it were a new thing, an anomaly. We may compare the recruitment of the Luperci to that of the colleges of Mercuriales and Capitolini. In these quasipriestly colleges, knights like M. Furius Flaccus, A. Castricius Myriotalenti f. rubbed shoulders with freedmen (Clesipus Geganius) or figures who in any case were neither senators nor knights (L. Spetumius).38 Furthermore, Cicero recounts without being the slightest bit critical or sarcastic, how a knight was excluded from these colleges where freedmen were, like the Luperci, able to take up the presidency.39 The absence of criticism on Cicero’s part, in a context in which there was nothing compelling him to hold his tongue, shows, I would argue, that this situation was neither scandalous nor new. In other words, the social openness of the lupercan sodality does not go back to the period of the ‘Verwirrung’; it always existed. If we are only aware of it in the middle of the first century , this is due to the lack of sources and not because of an anarchic process of development. In short, the case of the recruitment of Luperci in the Ciceronean period shows no kind of censitary criterion needed to be met in order to assume this function. It might be objected, however, that the example of the Luperci cannot be used to describe the situation of the major priesthoods. This is true, although under the Empire the lupercan priesthood became one of the priestly dignities reserved for knights; in other words, a very clear evolution can be observed, and the priesthood itself is neither marginal nor derisory. But let us accept that this example will not on its own Cicero, ad. Atticum 12. 5. 1: ‘Quintus pater quartum’ uel potius millesimum nihil sapit qui laetetur luperco filio et Statio, ut ‘cernat duplici decore cumulatum domem’; addo etiam Philotimum tertium. o stultitiam, nisi mea maior esset, singularem! (Text by Shackleton Bailey; July/August 46.) See the commentary of Shackleton Bailey, Letters to Atticus, 5 (1966: 303). 38 See the list given by Nicolet 1974: 2. 891–2; for the Mercuriales and Capitolini see Livy 5. 50. 4; Treggiari 1969: 196. 39 Cicero, Ad Quintum fratrem 2. 5. 2: Non praetermittam ne illud quidem: M. Furium Flaccum, equitem Romanum, hominem nequam, Capitolini et Mercuriales de collegio eiecerunt praesentem ad pedes unius cuiusque iacentem. (I am not even omitting this: the Capitolini and Mercuriales have expelled M. Furius Flaccus, an eques Romanus, of wretched character, from their college, when he was present and lying at the feet of each one). 37

46

The Priest and the Magistrate

suffice to elucidate the customs of priestly recruitment in the Republic, and let us instead try to look at the data we can be confident about in respect of the major priesthoods, so as to establish a rule which may or may not fit with the example of the Luperci. We have outlined a number of priests for whom there is no known magistracy, or whose co-optation took place well before they held qualifying magistracies, and we have concluded from their scenario that they were had been, or were still, knights. At the same time, we have to work out whether this observation has a real bearing on things here, and whether in fact we are only aware of these knight-priests because, as a result of illness40 or premature death, they were unable to hold magistracies or enter the senate. The fact that the co-optations often occurred quite early on may explain the situation. Four texts explicitly highlight the youth of priests such as Cn. Domitius Ahenobarbus (Szemler, p. 115, n 23), Q. Caecilius Metellus (p. 129, n 42), Ti. Sempronius Gracchus (p. 140, n 6), or Q. Fabius Maximus (p. 140, n 8).41 If one of these priests died shortly after being co-opted, then this leaves us with a kind of false problem: our priest was a knight because he did not have time to complete his cursus. The example of L. Pinarius Natta illustrates this situation well. In 63  he was in all likelihood a knight,42 and he died before 56, without having been able to seek a magistracy. Might this prematurely deceased adulescens demonstrate our hypothesis? If, indeed, we take the view of D. E. Hahm and G. J. Szemler,43 we then have to say that for the majority of priests, the priesthood in fact represented the first step towards a political career rather than being a result of that career. These youths were therefore co-opted shortly after reaching their majority, spurred on by their fathers and parents, and in this way set a date for the swift progression towards honours. All in all, the priesthood, according to this conception of the recruitment process, represents the first ‘honour’ to open the door to political influence, in other words towards the senatorial order, and the prematurely deceased youths—let us recall

40 P. Cornelius Scipio, for example, (our n 23) was of a weak constitution; it may be that his health prevented him from entering the senate (Cicero, Cato maior 2. 35; De officiis 1. 33. 121; Brutus 19. 77). Willems (1885: 350, n 232) nonetheless puts him amongst the quaestorians and doubtless sees him as a senator. I take a distinctly more sceptical view: there is nothing to prove that Scipio was a quaestor and a senator. 41 Cn. Domitius: Livy 42. 28. 10–13; Q. Caecilius Metellus: de viris illustribus 63. 3; Ti. Sempronius Gracchus: Livy 29. 38. 7; Q. Fabius Maximus: Livy 33. 42. 6 (Q. Fabius Maximus augur mortuus est admodum adulescens, priusquam ullum magistratum caperet); see also Szemler 1972: 191. 42 Nicolet 1974: 979, n 270; Szemler 1972: 135, n 64. 43 Hahm 1963: 76–7, 82–3; Szemler 1972: 56, 192.

The Priest and the Magistrate

47

that the Villia Annalis law set the minimum age for applying for the aedileship at 36 years—were unable to reach the end of the cursus they had begun. Let us first take here a passage from Livy44 which, on the occasion of one of four co-optations of adulescentes, notes that this practice was not current at the time. Such a remark makes it possible to reduce the possible interval between the date of co-optation and the date when first honours could be applied for. But this detail hardly moves us on. It seems more important to consider in itself the situation of those who were co-opted, rather than connecting up the co-optation with a possible senatorial future. Indeed, one important fact must first be underlined: at no point do we learn that applying to the priesthood was subject to conditions comparable to those required for equestrian or senatorial dignities; there is no question either of censitary criteria, an age limit, or any other condition such as carrying out military service.45 On the contrary, it was possible to be co-opted while still a knight, or a senator, and for some priesthoods while still a freedman; recruitment never required a minimum age. Lastly, military conditions probably played no role, as with the benefit of immunity, that is, above all uacatio militiae, the priests were in some sense placed outside of the ‘normal’ system. This question of immunity calls for a number of observations. We should start with chapter 66 of the lex coloniae Genetiuae, according to which iisque pontificibus/auguribusque, qui in quoque eorum collegio/ erunt, liberisque eorum militia munerisque public(i) uacatio sacro sancta esto uti pon/tifici Romano est erit (a)e(r)aque militaria ea omni/a merita sunto.46 We see that in a colony, the pontiffs and augurs (as well as their 44

Livy 29. 38. 6–7: . . . in M. Pomponi Mathonis auguris . . . locum (creatus) dandis sacerdotiis erat. Hahm’s assessment is too optimistic; in this connection. see the critical remarks of Szemler (1972: 193). Prosopography is unable to provide clear information on this point. Let us first note that the situation varied considerably from case to case. Indeed, the prohibition against any two members of the same family being present at the same time in the same college (see the very fine analysis of this rule in Bardt 1871: 34–6) could delay the co-optation of the son, if for example the family was keen for him to follow his father into the same college; the case of younger sons is also relevant here. Lastly, an ambitious youth could well bide his time and gain a pontifical seat rather than straightaway applying for a less lustrous priesthood like an epulo or decemvir. There is no need to point out that these subjective motivations have never been recorded by our sources. 45 One rule did exist, but it too demonstrates the particularity of priestly recruitment; the requirement to select the major flamens and the salii from the patricians. Other than the splitting of the major colleges into patricians and plebeians, this is the final vestige of the disputes from the beginning of the Republic. The fact that the major priesthoods were not open to the plebeians until 300 also shows that the evolution of political institutions and of sacred institutions did not follow an exactly parallel trajectory. 46 See Girard and Senn (eds.) 1977: 2. 16 ff.

48

The Priest and the Magistrate

children) enjoy a kind of immunity from military service and from public responsibilities which is similar to that enjoyed by the Roman pontiff. This privilege raises certain questions. First, we may wonder whether the pontiffs really held the privilege. For the Republican period, we do not find any evidence of a period of immunity from public responsibilities other than in three passages in Dionysius of Halicarnassus. By contrast, the sources tell us both about the military immunity of pontiffs and, from the reverse angle, about the lifting of immunity from tax or back-payments of tax in the case of pontiffs. For Mommsen, these texts, in particular a passage by Livy (33. 42. 3), do not establish the existence of pontifical immunity, but rather express the will of the priests to obtain it.47 To my mind, the view of Nicolet48 that this passage in Livy concerns only the claiming fiscal back-payments comes closest to the truth, especially when we consider the other sources: the pontiffs enjoyed immunity from military responsibilities except in the case of the tumultus Gallicus,49 and according to Plutarch, the law establishing this privilege included this exception (Plutarch, Camillus, 41; Marcellus, 3). But insofar as the pontiffs were not, as a result of their uacatio militiae, assidui, they did not, according to the logic of the fiscal system, have to pay the tribute either.50 This is why the conflicts of the Punic Wars (Livy 33. 42. 3), or the requisitions of 43 (Appian, Civil Wars 4. 34), provide a clear illustration of uacatio militiae from the perspective of the exceptional restrictions to which it could be subject. These restrictions placed on pontifical immunity are not without interest. Military immunity, along with its fiscal implications, was granted to priests by the city, but it could be lifted at any given moment. While the priest’s immunity in some sense distances him from the centuriate system and the responsibilities which it places on citizens, he is nonetheless not situated outside the realm of the city. He remains subordinate to the magistrates, and 47 Mommsen 1877: 100–1; 1887: 3. 1. 226. For Ziegler (1914: col. 1135), this relates to back-payments of the tribute. It is possible to explain this dispute in the following way: immunity no doubt only concerned normal responsibilities, those required in normal times. In this case, though, the issue was an extraordinary tribute which did not fall under the rule of ordinary immunity. For their part, the pontiffs tried unsuccessfully to extend their immunity. There are three passages in Dionysius of Halicarnassus (2. 21. 2; 4. 62. 5; 5. 1. 4) that mention immunity from military and civil responsibilities. 48 Nicolet 1976b: 218; 1976a: 77; see p. 69 for extraordinary measures. 49 Plutarch, Camillus 41. 7; Marcellus 3. 4; Appian, Civil Wars 4. 34; see also Dionysius of Halicarnassus 2. 21. 2; 4. 62. 5; 5. 1. 4. This exception of the tumultus is in keeping with the ter uacatio mentioned by the lex. col. Genet., which in fact refers to relative rather than absolute immunity (see Kübler 1933: cols. 649–50). 50 Nicolet 1977: 246; 1976a: 29, 33). The position of the priests is similar to that of the proletarii et capite censi.

The Priest and the Magistrate

49

despite his ordinary immunity, he may be called up or required to pay the tributum. In other words, sacred law is not radically cut off from public law; it is the latter which assigns the ius sacrum its place in the city, a very specific place quite separate from public law but not absolutely independent. The second point to be examined concerns the extension of this privilege. The texts mention the Pontiffs, the Greek of Plutarch no doubt referring to the same pontiffs. Dionysius of Halicarnassus evokes the case of priests chosen by the Romulean curiae (2. 21. 2), that of the duumuiri s.f. (4. 62. 5) and the rex sacrorum (5. 1. 4). The lex coloniae Genetiuae nonetheless guarantees this privilege for the pontiffs and the augurs of the colony, and Cicero (Academica priora 2, 121; Brutus 117) speaks of all priests indiscriminately. Furthermore, we know from Pliny (Natural History 7. 19) that the Hirpi of Mount Socrates possessed perpetuo senatus consulto militia omnium aliorum munerum uacationem. Lastly, the sacerdotes Lanuuinorum were also immunes.51 Faced with all this evidence showing the extension of the privilege of immunity, it is difficult to deny to the other Roman colleges, and in particular the three other major colleges, this privilege initially attested for the pontiffs alone.52

PRIESTLY RECRUITMENT To return to the main argument, it is now possible to say that the recruitment of priests took place in a very different context to the recruitment of magistrates. We must therefore avoid making a fixed rule out of prosopographical data showing that a priesthood granted to 51

CIL IX, 4399; 4206–8; CIL X, 3704. It is reasonable to think that the extension of the privilege goes back, at the latest, to the Augustan reform, where various benefits were granted to priests: subsidies and stipendium for individuals, lands and resources for colleges (see Suetonius, Augustus 31 and the analysis of Mommsen 1887: 2. 1. 62). One final point deserves mention. If on taking up their functions the municipal priests of the Imperial period offered games, this is probably more to do with an extension of the traditional cena aditialis of Roman priests than an imitation of the gratifications offered by magistrates on assuming office, even if the form which these festivities took may have been borrowed from the munera of the magistrates. Likewise, priestly immunity—which goes beyond the magistrates’ munera—cannot be justified by the munera, which in any case they offered to the people, and it does not constitute a kind of compensation, but is part of a more general system that separates the priest from the magistrate, prescribing for him certain specific forms of conduct such as the cena aditialis, for example, and its later extensions. 52

50

The Priest and the Magistrate

youths very often represented the first step in the conquest of honours. And we must not confuse the Republican situation with the Augustan data, where the sons of senators, practically the only individuals who could aspire to a senatorial priesthood, already officially belonged to the senatorial order as soon as they had reached their majority. From this period on, a senatorial priesthood obtained before the quaestorship must, just like the right to wear the laticlave, be considered the first sign of the future career that the senator’s young son would have.53 If the ambiguous status of these youths, who in a formal sense were knights, made possible the bestowing of an equestrian knighthood, an honour comparable to the sevirate of the Roman knights, the exceptional situation would rather be that of a young man who wears the laticlave and has been co-opted into a major college, but who prefers to remain a knight all his life.54 In any case the priests’ activity did not absolutely depend on their personal status. The liturgical acts performed by a freed or equestrian priest had exactly the same value as those of a priest belonging to the senatorial order. It would not occur to anyone to question the acts of freed Luperci at the time of Cicero, even if a man jealously guarding his rank might laugh at these rather peculiar colleges. The example of L. Pinarius Natta illustrates the point here perfectly. This young man had, in 57, dedicated the chapel of Liberty on the site occupied by Cicero’s house. On his return, Cicero set about getting his property back by having the dedication abolished, and in the speech pronounced before the pontifical college examined the problem in relation to Pinarius’ status. He attacked this dedication celebrated by an adulescens as opposed to a pontiff who was honoribus populi Romani ornatus.55 This argument, though, can be deceptive. The validity of the act executed by the young Pinarius, who had not yet sought honours and who was essentially still a young knight, is in no way contested on the grounds that it came from someone who was not a senator, from a pontiff who had not held a magistracy. Cicero simply wants to show that this act had 53

Suetonius, Divus Augustus 38. 2; Dio Cassius 59. 9. 5; see Chastagnol 1973: 589–91. For this kind of behaviour, see Chastagnol 1973: 590. By way of example, we may cite the following cases: CIL VI. 1397; CIL VI. 31716; see the commentary of Mommsen (1887: 3. 1. 568). 55 Cicero, de domo sua 45. 118: Hanc tu igitur dedicationem appellas, ad quam non collegium, non honoribus populis Romani ornatum pontificem, non denique nisi adulescentem quemquam, cum haberes in collegio familiarissimos, adhibere potuisti? (Do you therefore call this a ‘dedicatio’ to which you were unable to invite the college members, or to invite a pontifex who had received the honours of the Roman people, or, to sum up, invite anybody except some teenager, when your closest friends were college members). 54

The Priest and the Magistrate

51

been performed on the sly by a young man—related, what’s more, to the wife of Claudius—to the exclusion of the other members of the pontifical college. Let us not forget that it is precisely before all these other pontiffs that Cicero is pleading, and that it is not in his interest to implicate them in the outrageous dedication. He moreover explains the situation by saying: ‘Very well, you did not call in the college; but what? Who, in the end, then, from this college was present? Because for the authority that you needed, if they all have it, age and rank enhance dignity, and, for the knowledge that you also needed, if they all possess it, age always gives more experience . . . . But if it is knowledge we are looking for, who had less experience that the one who had entered the college but a few days previously?’56 This is an interesting account, because even making allowances for the polemic and the captatio beneuolentiae, it reveals a certain tension, or rather a new mindset, in which youth and inexperience are opposed to age and knowledge, but above all youth and non-senatorial status are opposed to the authority of a priest who is also a magistrate. Nevertheless, Cicero recognizes the auctoritas of the young Pinarius, and in any case this is not what makes the dedication invalid. If there is an irregularity, it is because the dedication was carried out without being expressly ordered by the people.57 In short, the invalidity of the act derives not from the rank of the celebrant priest, but from a violation of customs that Cicero puts down to the inexperience of a young pontiff who had only recently been co-opted. Other examples may be used to show that the priests do not have to be senators for their acts to be valid; in other words, the personal status of the priest is of no importance, and is not subject to any kind of censitary requirement. As we know, the reges sacrorum could not hold their religious function in combination with any kind of public honour.58 A ‘king’ who had been chosen young, that is, before he could enter the senate, will therefore never be part of the senatorial order, for we do not possess any 56 Cicero, de domo sua, 45. 117: Esto, collegium non adhibuisti; quid? De collegio quis tandem adfuit? Opus erat enim auctoritate, quae est in omnibus, sed tamen auget et aetas et honos dignitatem; opus erat etiam scientia, quam si omnes consecuti sunt, tamen certe peritiores uetustas factit . . . 118 . . . Sin autem scientia est quaesita, quis erat minus peritus quam is qui paucis illis diebus in collegium uenerat? 57 Cicero, de domo sua 53, 136. We understand that Cicero, keen to show that a serious irregularity going against to all the customs had been committed, attempts to erect a barrier between the young, inexperienced pontiff and the dignity of the pontifical college, in particular that of the pontiffs who are honoribus populi Romani ornati. 58 Dionysius of Halicarnassus 4. 74. 4; Plutarch, Roman Questions 63.

52

The Priest and the Magistrate

information about an automatic inscription of priests on the album of senators: as the son of a senator, the rex sacrorum will remain a knight his whole life through.59 One final argument of substance is provided by the case of C. Valerius Flaccus (our n 41). In 209, having led an adulescentia neglegens luxuriosaque (Livy 27. 8. 4–11), Flaccus was against his will seized upon by the pontifex maximus P. Licinius Crassus and inaugurated as a flamen Dialis. Contrary to all expectation, the new flamen distinguished himself by the extraordinary care with which he carried out his functions; specifically quoted is the conflict between him and the moneylender P. Licinius on the subject of an ancient accepted right belonging to the flamen Dialis. This is the privilege in question: [C. Valerius] reclaimed a right which had been suspended for many years due to the unworthiness of his predecessors, namely the right to enter the senate. 8. He came to the curia and, having been ejected by L. Licinius, he appealed to the plebeian tribunes. The flamen repeatedly reclaimed an ancient right of his priesthood: this right, he said, was given along with the toga praetexta and the sella curulis. It was the praetor’s wish, on the other hand, for the right to be founded not on precedents that time had forgotten, drawn from the annals, but always from the most recent usage; never in the memory of their fathers or grandfathers, had any flamen of Jupiter made use of the right in question. 10. The tribunes, having rightly deemed that the lapsing of this advantage, due to the negligence of certain flamens, was harmful to the flamens themselves, but not to their priesthood, had the flamen admitted to the senate, without the opposition of the praetor himself and with the general agreement of the senators.60

First of all, the above account shows that if this right was formerly granted to the flamens of Jupiter, it is because a number of them, at 59 The complete lack of information means that we cannot further elaborate. However, it is possible that certain reges did not even belong to the equestrian order, if for example that had been chosen before they had been inscribed in this order. See also Szemler 1972: 99. 60 Liv., 27. 8. 7–10: Rem intermissam per multos annos ob indignitatem flaminum priorum priorum repetiuit ut in senatum introiret 8. Ingressum eum curiam cum P. Licinius praetor inde eduxisset, tribunos plebis appellauit. Flamen uetustum ius sacerdotii repetebat: datum id cum toga praetexta et sulla curuli ei flamonio esse. 9. Praetor non exoletis uetustate annalium exemplis stare ius, sed recentissimae cuiusque consuetudinis usu uolebat: nec partum nec auorum memoria Dialem quemquam id uis usurpasse. 10. Tribuni rem inertia flaminum oblitteratum ipsis non sacerdotio damno fuisse cum aequum censuissent, ne ipso quidem contra tendente praetore, magno adsensu patrum plebisque flaminem in senatum introduxerunt, omnibus ita existimantibus magis sanctitate uitae quam sacerdotii iure eam rem flaminem obtinuisse. See Mommsen 1887: 3. 2. 860, n. 1, VIII n. 1; for Richard (1968: 789) the priest in question may be identified with the great pontiff P. Licinius Crassus Diues.

The Priest and the Magistrate

53

least, did not belong to the senate.61 For the plebeian tribunes, the fact that some flamens did not claim this privilege was not detrimental to their function, only to the individuals most directly concerned. As a result, several flamens of Jupiter prior to the year 209 did not belong to the senate, and their lower rank, in all likelihood the equestrian rank, in no way affected the authority of their acts. Quite the opposite, in fact, as one final example proves. In 200, C. Valerius Flaccus applied for the curule aedileship. His candidacy aroused bitter dispute, no doubt with the same fastidious pontifex maximus, P. Licinius Crassus,62 and once more bears witness to the fact that in the eyes of many Romans there was no disadvantage in the flamen of Jupiter remaining outside the political cursus. As for the other flamens and priests, no hint is given as to the extension of the exceptional privilege granted to the flamen of Jupiter. It is therefore very unlikely that the priests would have been automatically inscribed on the senatorial album. Likewise, what was true for the most august priest of the Republic was true a fortiori for the other priesthoods: the fact of being mere knights did not devalue the acts they performed. Lastly—and this piece of evidence is even clearer—the unimportance of the priest’s status within the social system of the Republic is strikingly demonstrated by the existence of female priesthoods of the highest rank, that of the vestals, the regina sacrorum and the flaminica. Putting all this data together, we may establish the principle that unlike the priests of the Augustan period, the Republican priests belonged indiscriminately to the senatorial or equestrian order, and even to other plebeian groups, without there being any difference in 61

If this privilege had lapsed, it is doubtless not so much on account of the obscurity of certain flamens, as Mommsen imagines (1887: 3, 2. 860 n. 1), but rather because certain flamens had wished to take part in public affairs and had applied for magistracies; see Richard 1968: 794. Thus, in the case of consular flamens, for example, the privilege had become pointless. Mommsen is right to draw attention to the same right—with the necessary adjustments for context and period—recognized for the provincial flamen at Narbonne, which is derived from the right which the flamens of Jupiter enjoy (1887: 3, 2, VIII n. 1). 62 Livy 31. 50. 7–9; Richard 1968: 789. We will not examine here the difficulties which the taboos of the flamen presented for the administration of a magistracy. We are only concerned with the general consequences of this debate. Let us note in passing that the opponents of C. Valerius Flaccus also referred to the specific character of the flamenal function which was to be outside of time, to be a ‘living statue’ of the god, the human culmination of Jupiter’s connection with the city. These aspects clearly placed him outside political life, unlike the pontiffs for example, who were committed, engaged in the events of the time. See Dumézil 1966: 550–66; 1974: 2 [trans]: 580–93. However, as Bleicken (1957b: 446) for example explains, there is no question in the Roman tradition of priests being forbidden to administer magistracies. The isolated case of the rex rather proves the opposite.

54

The Priest and the Magistrate

their respective remits. A kind of counter-evidence to this custom may be brought in here by studying those priesthoods which in the Republican period were reserved for knights alone: the minor pontificates and flaminates, and the sodality of the luperci. Whether or not M. Popillius Laenas was the aedile of 364 and the consul of 359, 356, 350,63 it is in any case clear, according to the testimony of Cicero,64 that he was flamen Carmentalis and consul. The minor pontiffs of the Ciceronian period, P. Albinouanus and Q. Cornelius are not attested as magistrates or senators, and it is difficult to give a verdict as to their rank. P. Volumnius, on the other hand, has been identified by T. R. S. Broughton as P. Volumnius, the friend and judge of Cluentius in 66. Moreover, as Nicolet highlights, he is named ‘in a list starting with knights, after Cn. Tidicius, who is a senator: he therefore has every chance of also being a senator’.65 For his part, the minor pontiff Q. Terentius Culleo was a plebeian tribune in 58, meaning that he was a senator. As for the luperci, we may cite the examples of M. Caelius Rufus and Mark Anthony,66 which show that the senators could perfectly well take part in the famous February festival. These examples cited thus also demonstrate the absence of any desire to reserve certain priesthoods for the equestrian order or other types of social group. Our inquiry has thus established that the recruitment of priests took place outside the recruitment system for magistrates or orders, such that we have to qualify somewhat the objections voiced earlier concerning the value of the list of priests for whom no magistracy is attested. It may well be that a number of them never belonged to the senate. However, it is also undeniable that the majority of priests were senators, with knights only representing a limited proportion of all priests. The information we have about the other social groups is too flimsy to warrant an estimate. Among the priests of the last three centuries of the Republic, this predominance of members of the primi ordines nonetheless shows that the Romans approached religious dignities in the same way as political functions: in a way, the censitary reactions also played out at a level where they did not in principle need to intervene. In any case, as we will see later on, attitudes seem to have evolved towards a convergence in the 63

Broughton 1950–60, 1: 121 n. 1; Szemler 1972: 60. Cicero, Brutus 14. 56: ‘qui cum consul esset eodemque tempore sacrificium publicum cum laena faceret, quod erat flamen Carmentalis’. (‘. . . while he was consul and at the same time performing a public ritual, wearing his priestly cloak as the flamen Carmentalis . . .’). 65 Nicolet 1974: 2: 1082–3, n 1. 66 Szemler 1972: 178, n 401; Mark Antony: Cicero, Philippics 2. 84ff.; Plutarch, Antonius 12. 3 (note the passage where Plutarch mentions the participation Greek, clearly an allusion to recruitment in the Republican period); Dio Cassius 45. 30. 2; 46. 5. 2. 64

The Priest and the Magistrate

55

two systems of recruitment. For now, let us look at the contrasting opinions held by the opponents of C. Valerius Flaccus and of Cicero in order to bring out a progressive conjunction of the two types of public dignities, priesthoods, and magistracies. At the dawn of the second century, some Romans wanted to maintain a separation, at least for the flaminate of Jupiter, between priesthoods and political dignities. For his part, Cicero assessed the experience and authority of a pontiff in terms of the political honours that he had held.67 Compared with factual reality, however, these opinions turn out to be no more than that. Indeed, not until the period of the Augustan reform was the recruitment of priests ever subject to the same rules as the election of magistrates, although this does not mean that opinions held no importance. In short, the place of priests in Republican society did not change during the Republic. It is only with the fall of the entire Republican system that the position of the priest began to evolve, such that we may assume that it was closely bound up with the whole institutional edifice that characterized the free Republic.

FROM REPUBLIC TO PRINCIPATE Having made these observations, we can now go a step further and ask what were the reasons behind the development that took place between the last century of the Republic and the reforms of Augustus, and what is the significance of this development? To answer these questions, we should first set out the situation as it was in the Republic. Mommsen, whose Roman Public Law remains on a number of points the best manual of religious institutions, drew a very clear line between priests and magistrates, corresponding to the one which, in the realm of the ius publicum, distinguishes the two complementary sectors of the ius sacrum and the ius publicum.68

67 It is interesting to note that Dionysius of Halicarnassus presents the election of priests in the Archaic period from a perspective that dates back from the time of the Gracchi and from the first century : this is why he evokes the conditions of age, birth, and fortune. For the ideological background of his Antiquities, see Pohlenz 1924: 157–89; Gabba 1960: 176–225; 1961: 98–121, 42; 1964: 29–41. 68 Mommsen 1887: 2. 1. 52 and notes; Wissowa 19122: 380–1; Bleicken 1957b: 446. Mommsen places particular emphasis on the fact that sacred law is only one category of public law, citing Ulpian, Digest. 1. 1. 1. 2; see also Catalano 1974: 667–91.

56

The Priest and the Magistrate

According to this conception, the normal and traditional worship of the city gods is the duty of priests and is free from the control and interference of magistrates. It belongs to the irreducible realm of the priest’s independence. Of course, the magistrates may and indeed must pray, sacrifice, vow, and make dedications in the name of the Republic, but they can do no more. In other words, fully to bring out Mommsen’s way of thinking, it is not by these acts that the priest is defined, insofar as any citizen, and thus any magistrate as well, may perform them. Being a priest and acting as a priest means something else, something more. This priestly quality is clear in the case of the flamens, the rex, the vestals, and the members of sodalities, in that some forms of action and behaviour belong to them exclusively. We also see it in certain functions of the pontiffs, augurs, and quindecemvirs (the celebration of regular festivals, captio, inaugurations, the consultation of the libri sibyllini), to which we must also add competence for the sacred tradition, and the assistance and recourse that these priests offer to the magistrates for sacred law, which is their respective responsibility. By contrast, in the same way that the sacred always takes precedence over the public, while remaining subject to the authority of the magistrates,69 priests do not hold any political power. As Jochen Bleicken demonstrated not so long ago,70 no priest, not even the pontifex maximus, possesses the imperium, the auspices, or the potestas. If they do intervene publicly, it is to safeguard the rites entrusted to them by virtue of the rights and duties that they receive at the point when they are co-opted. Without going into the ceremonies of regular worship, it is possible to takes as an example the prerogative of the augurs to end or cancel the comitia. Reading again the famous passage from De legibus in which Cicero extols this privilege,71 we can see how he relates it not to the free conduct of augurs or to their political powers, but to their ius, the ius augurum, whose supreme auctoritas he underlines. The highly political intervention of the augurs is justified not by an imperium or by auspices superior or at least equal to those of the supreme magistrates, but only by the ius augurum, by the sacred customs that give them the permanent authority to make these restricting obnunciations. The augurs, in fact, take on responsibility for the sacred aspect of holding comitia, the part that falls to the priest and which complements the political and religious duties of the magistrate. This is why we should not 69

Mommsen 1887: 3. 2. 941; Wissowa 19122: 380–1; Catalano 1960: 164; 1962: 133–4, 137–8; Sabbatucci 1961: 154ff.; Nicolet 1977: 421; Cardauns 1976: 139; Gioffrodi 1958: 22ff. 70 Bleicken 1957a: 345–66; Latte 1960: 400ff. 71 Cicero, De legibus 2. 12. 31.

The Priest and the Magistrate

57

be fooled by the elliptical formulas of the ancient historians. When a priest participates as a priest in some or other public act, he does so accompanied by a magistrate—whether or not the magistrate is named— and only takes care of the properly religious part that this act may involve. The convening of comitia does not fall within his remit, and when a priest appears to be convening assemblies, it is always because he is also a magistrate, or else the meetings in question are very unusual. Thus in all probability—although this is something that is often questioned—72 the pontifical college in its entirety, and through its spokesperson, the pontifex maximus, convenes and presides over the comitia calata of the curia in respect of the testament, the inauguration of certain priests, the detestatio sacrorum, and the adrogation. But there we are dealing with very unusual assemblies where the curia only provide assistance and do not take any decision. In a more recent period (at the earliest after the end of the third century), the comitia curiata vote for the adoption of a man sui iuris, but this is a new procedure, an exception owing to the importance that adoption has increasingly taken on. It cannot, therefore, be used to describe the normal powers of pontiffs. The power to convene the curiate comitia so that they could vote through this lex curiata was in a sense added to the normal powers of pontiffs, because this was in the nobility’s interests, but it does not in any way derive from the ordinary powers of this college. Conversely, the priests, whose decrees, interventions, or advice had to be requested and respected by the magistrates, cannot exercise these ‘powers’ without some issue first being referred to them, or as we said earlier, without having received this right when they were co-opted. Without being expressly consulted by the magistrate and the senate, without the publication of a magistrates’ decree, priestly advice does not exist. Once expressed with the authority of the magistrate, it becomes mandatory for all.

72

The opinion of Mommsen (1887: 2. 1. 35 n. 3; 38–9), according to which the pontifex maximus shared royal powers with the consuls, before the power to convene legislative comitia passed almost entirely over to the consuls and the centuriate comitia, has been questioned by Bleicken (1957a), followed by Kunkel (1962: 22 sq.) and Szemler (1978: cols. 344–6), who take the view that it is not possible to establish from the cases invoked by Mommsen that the pontifex maximus convened and presided over the curiate comitia. For them, the pontifex maximus only attends these comitia (they also, therefore, reject Brecht’s ‘spiritual imperium’). However, Magdelain (1978: 82–5) has refuted this critical standpoint, demonstrating that the comitia in question were indeed convened and presided over by the pontifical college. His is the position I am adopting here, drawing on his interpretation of the comitia concerned. Essentially, these comitia are to the other comitia what the priestly comitia are to other electoral comitia; see Bleicken 1957a: 357.

58

The Priest and the Magistrate

A very clear separation exists, then, between priest and magistrate, and according to Mommsen, this separation finds expression in a specific mode of recruitment for priests (co-optation, captio, in any case never a ‘real’ election, even after the lex Domitia), in the lifelong duration of the function, and of course in rules, insignia, and particular privileges. And Mommsen subtly brings out that over the Republican cursus, priesthoods and magistracies always exist side by side of one another.73 The two functions are separate, even if they are often taken up by the same man. This observation is reinforced if we consider that in the Republic, at least, priests belonging to the senate were never viewed there as priests. Their religious function did not set aside for them a particular rank in the senatorial hierarchy; on the contrary, it is only as consulars, praetorians, or quaestorians that they figure, in their right and proper place, amongst the witnesses of the S.C. on the Bacchanalia (186 ), of the Tiburtibus (circa 159 ), and of the consular provinces (51 ). Likewise, as Szemler has shown,74 it is only as magistrates or ex-magistrates that priests can speak in the senate. In short, there was always a very clear separation, even within the same man, between, his religious and his public functions. While kept strictly apart in the Republican period, the priesthood and the magistracy are nonetheless intimately bound up with one another, and the entire institutional edifice of the Republic is founded on this necessary juxtaposition of the two interconnected powers. In a way, the two functions are quite similar, and Mommsen has compared the main priesthood, the ‘great pontificate’, to the magistracy, in the sense that the latter secures the guardianship of the community of citizens, unable to express itself directly, and the former the guardianship of the gods, who also, in spite of their incontestable authority, ‘agree’ to be in some sense subject to the authority of a particular magistrate.75 This comparison is entirely pertinent and felicitous. The Republic is effectively an 73

74 Mommsen 1887: 2. 1. 18; 1887: 1. 544. Szemler 1972: 53–8. Mommsen 1887: 2. 1. 22–3: ‘Just as the magistrate may, in a sense, be thought of as the guardian of the people, who are unable directly to express their will, so the great pontiff is the guardian of the gods of the people, that is, the one who carries out for them, and in their name, the necessary manifestations of their will; and he is the rightful representative of all the deities recognized by the Roman people, while the other priesthoods who are charged to act within the cult are assigned to a single deity. In this respect, the pontificate stands in at least as vigorous an opposition to the priesthoods allocated to a cult as to the magistracy. Its religious function is, in a very real sense, to represent the element of the magistracy in the realm of religion’. As we have said, Mommsen went too far in comparing the great pontificate to a magistracy. It is the relations of the pontifex maximus with the gods that are similar in kind to those of the magistrates and citizens, not its public powers (see n. 72). But it is still the case that the interest of the gods is represented by the colleges of priests, and that the pontifex maximus represents all the gods. 75

The Priest and the Magistrate

59

association of three partners: the gods, the people, and the magistrates. But although Mommsen connected up the representatives—the ‘guardians’ of these first two partners, the magistrates and the pontifex maximus—it would be more accurate to say that gods and citizens alike are subject to the power of the magistrates, to a form of public power which, to be absolute, must include both sacred elements and strictly public elements.76 It is the strictly religious elements—essentially, divine participation in a public act—that the priests uphold. Insofar as it is the involvement and understanding of priests, rather than the direct intervention of Jupiter, the direct contact of the magistrate with Jupiter, that are required, we may speak in terms of ‘guardianship’. In the same way that the magistrates look after the interests of the people, and that they speak and act in the name of the people, so the priests look after sacred law and speak for the gods. Priests and magistrates are therefore the trustees of certain permanent mandates, of a particular tradition, and they contribute towards the practice of performing complete public acts, with the one exception that in whatever exceeds the framework of his normal assignations, the priest does not have a voice unless the matter at issue has been referred to him by the magistrate and the senate. This is the reason for the view taken here that the gods themselves are also subject to the authority of the magistrates. Before looking at how the historical forms taken by this opposition evolved, it may already be possible to answer the question formulated above: is the conception of Cicero and Ulpian confirmed by historical inquiry? Yes, with one qualification. Touching on an area not already shaped by the intervention of an ancient thinker, prosopographical analysis has established two facts. First, it has shown that in spite of the historical variations that we will examine later on, priests cannot be assimilated to the holders of temporal power, the supreme magistrates. 76 It is not possible within the scope of the present study to set out the condition of the gods in the Roman city. I hope to return to this elsewhere. Let us simply note that, like the citizens, the gods are first of all the owners of a plot of land, a ‘house’, and inalienable cultural objects. In fact, they only become Roman gods after a public act, a naturalization: what counts as a god is not the deity who reveals itself to the Romans, or one whom the Romans worship, but one who has officially received ‘Roman citizenship’ from the senate. As we have said, after being enthroned in this way, the deity furthermore receives an inalienable property. Besides, the gods cannot intervene directly in affairs. They may, of course, show their agreement or disagreement, but this opinion is received and interpreted by the magistrates, with the gods— meaning above all Jupiter—‘voting’, as it were, like the people with a yes or a no during the ‘comitia’ convened by the magistrate cum imperio. Like the people, the gods may make a violent show of their displeasure, but once more the omina are received and interpreted by the magistrates, after consulting the relevant priests, and without this intervention the signs do not exist. Without the magistrate and the senate, the Sybil remains silent.

60

The Priest and the Magistrate

Unlike these, the priests are in fact free from any censitary criterion, are not subject to any of the rules that apply to the recruitment of magistrates, and are not automatically bestowed a place in the pyramid of orders. As such—and despite the points of convergence, which we will examine—the priests are not placed in the group of citizens that are geared towards temporal action, the assidui, the knights, the magistrates, or senators. This does not mean so much that the priests are ‘renouncers’, forming a caste removed from historical reality and acting outside of the city. Priests, especially the members of the major colleges, are part of the Roman elite; they come from the same social class and the same orders as the knights or senators, and form a very tight circle, which for the Romans implies a high function. But while they may, for the most part, be recruited from amongst the descendants of the primi ordines, their place and their role remains unaffected by the functions belonging to these. Whether or not he is a senator, whether or not he has been decorated with higher magistracies, the priest possesses an authority and a legitimacy of a different order, analogous to that of magistrates. What has been suggested by prosopographical analysis, the brief description of the priest’s work has been able to make plain. The realm belonging to the priests is the sacrum, conceived as a necessary part of the activity of the Republic, and not as a timeless activity. Their specific activity complements that of the magistrates. But if the priests occupy a rank analogous to that of the magistrates, they are not on the same level. Superior in the absolute realm to the magistrates with imperium, they nonetheless remain subject to their temporal power, and in public life do not have any other initiative than that which these magistrates ask of them, or is conferred on them when they are co-opted. At a temporal level, the magistrate with imperium is therefore invested with a fuller power than that of the priest; he is active and entirely geared towards action, even at a religious level, and thanks to the auspicial investiture, which only he is able to request, he possesses a capacity for extraordinary action, uniting the foundations of the sacred and the profane with the complete deed. But while he exercises a superior temporal power, even at a religious level, insofar as he gives voice to the gods or celebrates one part of the cult in the name of Republic, the magistrate remains, as such, excluded from the sacred realm; he does not represent the interests of the gods, is not master of the ius sacrum whose rules are necessary for him to act, and is of course still subject to the will of the gods. Prosopographical analysis and the outline sketched of the respective powers belonging to the higher priests and magistrates demonstrate a clear convergence between institutional reality and the theory formulated by Cicero and Ulpian. In the Republic, supreme power effectively

The Priest and the Magistrate

61

appears to break down into two areas which are quite separate but intimately bound up with one another: sacred power and the power of magistrates. The theory of the supreme laws of the city, or the distinction between the two hierarchical categories of public law, are most probably not fictions, selectively drawing on certain elements of power, but express a global and exhaustive vision of supreme power, one which is doubtless founded on traditional mindsets, but also on the timeless structure of the foundations of power and the way in which it is put into effect at an everyday level. On one point, though, the theory examined departs from what prosopographical analysis and the comparative assessment of powers have revealed. The hierarchical interconnectedness of the two powers has shown itself to be more complex in reality, for if the priest and the sacred come first or are superior in the realm of the absolute, they remain subordinate to the temporal power of the magistrate. This difference is in part down to the fact that Cicero and Ulpian are describing laws, or public law, from a static rather than a dynamic perspective. Restricting themselves to abstract data, they are not concerned with what the actual facts, such as prosopographical data or the exercise of powers, may point up. As much as an oversight, this is therefore a question of a different approach to these facts. However, it is clear that the Ciceronian lesson also has a certain bias. Influenced by the excesses of his time, Cicero places strong emphasis on the pre-eminence of the priests, and in particular that of the augurs, who could be a powerful force for moderation (e.g. p. 2. 12. 31). As a result, the aspect of this that interests us here, and which could be considered to justify abuses of power, is passed over. Having ascertained that the theories of Cicero and Ulpian involve a fundamental representation of supreme power, we may now turn our attention to the way this ideology played out, that is, towards the historical evolution of a timeless template. As something which is required, which is bound up with this ‘secular’ power, but subordinate to it, the priesthood, as we have said, retains its independence, as well as a ‘spiritual’ authority. It concerns the heavenly sphere, the absolute, with the magistracy taking care of the terrestrial sphere. And of course, it is for this reason that the priestly institution, like its sacred counterpart, obeys specific rules, is not subject to the conditions required for the magistracy, and in short develops in parallel to it. When, beginning with the fall of the monarchy, an entirely new system of ‘secularized’ power gradually came into being, the priests remained, as such, untouched by this development, and up until the end of the Republic, the formal separation would, as we have witnessed, remain so keen that we might wonder whether this state of affairs was

62

The Priest and the Magistrate

not connected to the deeply set character of the free Republic, or at least to the idea that the Romans had of it over the last two or three centuries. According to this conception, which pervades every institution and the entire tradition, powers must be spread out as far as possible. A number of different means are used to this effect: election with universal suffrage and an increase in magistrates, yearly tenure, limited, hierarchical powers, and restrictions on repeat-appointments. From a religious standpoint, this division of powers is established the day that priests alone, rather than a magistrate-priest, have responsibility for the interests of the gods, in other words when power becomes secularized (insofar as this was possible in the ancient world). As Bleicken has emphasized,77 this evolution did not happen overnight. It developed and progressed over centuries, moving towards an ever greater separation of powers: a rise in the number of priests, the opening of priesthoods to plebeians, and the creation of new colleges. Nevertheless, the fundamental principle of the division of religious and political powers stems directly from the advent of the Republic, which according to André Magdelain was not ‘an embryonic but a radical beginning’.78 It seems to me that the specificity of priests, or the necessary complementarity of priests and magistrates, expresses a single wish: to restrict as far as possible the subjective, nondemocratic element in the exercise of power, to avoid as far as possible the concentration of powers. In this system, no magistrate could now make claim to a total freedom of action, to total power. Even dictatorship, at least in the historical period and up until Sulla, does not allow all powers to be controlled.79 Henceforth, we may say, complete action comes about through the collaboration of two interconnected but separate bodies, and is no longer only the result of two antithetical but complementary principles of action combined in the hands of a single agent. The place assigned to priests by institutional development made them key elements in the balance of power, and their ‘indifference’ in respect of the criteria for recruiting lay agents stemmed from this autonomy of the sacred as much as from the mindset of the free Republic. As all scholars here have stressed, the role of the pontifex maximus, and above all that of the augurs, constantly expanded over the course of history.80 It was not by chance, then, that the regulation of

77

78 Bleicken 1957a: 447. Magdelain 1979: 698–713. Nicolet 1977: 412–14. 80 For example, Fears (1975: 592–602) has shown that the augural priesthood gained in importance in the final year of the Republic, because in some sense it demonstrated the excellence of the auspices taken by a magistrate, which when exegesis became difficult, was better able to interpret them using the most authorized tradition. The religious writings of 79

The Priest and the Magistrate

63

priesthoods formed the stakes of a bitter political dispute from the end of the second century, and even before then, when it came to deciding on whether to open up the recruitment of priests or to keep in check the election of the pontifex maximus. As the expression of a political will tied into the very principles of the free Republic, all these conflicts at the same time pave the way for a later development which would increasingly bring about a different version of the timeless design of supreme power. This evolution took place in two phases, and, through a desire for popular oversight of priests and the reactions elicited by this desire, culminated in an important alteration in the place of the priesthood in the constitutional structure. Thus, roughly between 200 and 100, a number of measures and religious conflicts can be explained by a will on the part of the plebeians, and later on the ‘populares’, to give ever clearer expression to the relative autonomy of the priesthoods, but also to exert ever greater control over religious power, meaning the pontifex maximus and the other priests.81 We have already seen how Polybius, who was doubtless no stranger to debates taking place in circles which were starting to promote a new conception of supreme power, eliminates the priesthoods from his system. In the same way, this circle, whose aspirations Polybius may be considered to be voicing, is opposed to any measures that tend to accentuate the autonomy of sacred power. Fluctuating between rigid Republican traditionalism and an openness to new ideas, these two perspectives, which are in principle opposed, end up coming together, or in any case, preparing the ground for an important shift in the conception of religious power, closely connected to the advent of a new political regime. It is in relation to the system of recruitment for priests that the first battles were waged and the first breaches very gradually opened up in traditional representations. Up until the end of the second century, priests would be renewed by co-optation, that is, by a system entirely free from popular control, meaning that ‘religious power’, one of the constitutive elements of power as a whole, lay outside the control of the people. However, from at least the middle of the third century,82 Cicero contain many allusions to this exceptional place acknowledged for the augurs, and Cicero’s vanity is not enough to explain the importance that he gives them. This role was bestowed on them both by institutions and by historical development. 81

See Rawson 1974: 193–212; Cornell 1981: 27–37. Richard (1968: 786–801) has shown that around 200, the plebeians adopted a somewhat conservative religious policy, directed of course at their enemies who were interpreting religious law too loosely, in other words, from what we have just said, in too subjective, too ‘personal’ a fashion. 82 For this see Mercklin 1848: 115ff. Pais (1915b: 339–46) dates this reform from the years 255–252; whereas Taylor (1942: 421) declines to give an exact date, placing the reform within the period of Livy’s second lost decade (292–219 ).

64

The Priest and the Magistrate

the pontifex maximus was elected by a special assembly comprising seventeen tribes drawn by lot from the thirty-five at the point of election.83 The people thus acquired a right of inspection over the appointment of the most important priest, but essentially this supervision was only relative, as the pontifex maximus had to be chosen from among the priests, themselves co-opted according to ancient tradition. After an initial attempt on the part of C. Licinius Crassus, tr. pl. in 145, the lex Domitia endeavoured in 104/103 to democratize the process by which priests were nominated by conferring on the special assembly responsible for appointing the pontifex maximus the right to elect all members of the four major colleges.84 The candidates, though, were exclusively appointed by the priestly colleges themselves. The progress signalled by the lex Domitia is important. Its intentions were commendable: to put an end to certain types of feudal system which were deleterious to religious life and therefore politics. But in the process, the balance between priesthood and magistracy was destroyed. Indeed, as Mommsen has already highlighted, the colleges which were recruited by the new procedure, that is to say the pontiffs, the augurs, the decemvirs, and in all probability the septemvirs,85 thus received for the first time since the foundation of the Republic the rank of quasi-magistrate.86 Not only were the quasi-comitia convened, but the four priesthoods concerned gained a lustrous position by virtue of this comitial procedure (amplissima collegia).87 To be sure, the separation of the sacrum and the

83

Mercklin 1848: 138; Bleicken 1957b: 357. For the lex Domitia see Mercklin 1848: 135–43; Taylor 1942: 421–4; Parrish 1977: 623–33. 85 Mommsen (1887: 2, 1, 28 n. 1) is no doubt correct in thinking that the eminent position held by the college of the epulones was acquired at that point by its addition to the three great state priesthoods. To which the following argument may be appended: the four priesthoods affected by the lex Domitia were practically the only ones which concerned public religion as such, and which perfectly expressed the Republican aspect of religion, as opposed to the sodalities, which demonstrated a different, as it were pre-poliad, kind of interconnectedness. In a way, the septemvirs shared in the pontifical remit, in particular on one very important political point: games. 86 Mommsen (1887: 2. 1. 28 n. 3): ‘. . . there is a strong probability that their separation from the other priesthoods is founded precisely on the lex Domitia, which by establishing a popular quasi-election for these four priesthoods gave them the rank of magistracy. The particular political influence of these colleges must surely have been the determining motive’. 87 Mommsen 1887: 2. 1. 19: ‘. . . It is even likely that the priesthoods were only brought into line with honours by extending to them the popular quasi-election, and only at the beginning, for the duration of this extension; the four priesthoods that go through to this election do not do so as the most elevated any more than they are the most elevated as a consequence of going through it’. 84

The Priest and the Magistrate

65

publicum was maintained. Not only was the election held by the minor pars populi, but what is more, the applications were subject to the authority of the colleges themselves. Lastly, the co-optation itself was carried out, after the comitial election, by the priests.88 From a formal perspective, appearances were kept up at least, but I would argue that fundamentally the lex Domitia is the expression of a profound shift in the political mindset, a shift which first comes out into the open in the time of Sulla, and then in that of Augustus. While going back to the tradition prior to the lex Domitia, Sulla in fact continued to view the priests of the four colleges as quasi-magistrates. The information we have about his activity at religious level is pretty slender,89 but it is possible to say that while he may appear to have put back into force a clearer separation between the recruitment of priests and that of magistrates, he also raised the number of priests to the same proportions as magistrates and senators.90 In other words, these 88 According to Pais (1915b), the assembly of the thirty-three tribes, formed so as to elect the pontifex maximus was meant to show that the people only held one half of the power in the appointment, the other half reverting to the pontifical college which carried out the nominatio of candidates, and, after the election, the co-optation of the candidate chosen. Bleicken (1957a: 357) is of the view that this minor pars populi, this people which is not a people, expressed the fundamental principle that the sacrum could not be subject to the laws of the publicum. This is the interpretation that I am adopting here. Insofar as the election of priests is concerned, ‘co-optation’ now involved three, and for the augurs even four stages: the nominatio, the petitio/creatio by the tribes, and the cooptatio by the college concerned, possibly followed by the inauguratio. Based on an analysis of the election of the augurs and the pontifex maximus, the only elections for which there is good evidence, Mercklin (1848: 138–40) was able to suggest that the nomination could only be performed by two priests for a single candidate, and could not involve more than three candidates for each post vacant. It is between these candidates that the tribes had to choose. The petitio— which could also be carried out in absentia—was then followed by the actual cooptatio by the members of the sacerdotal college concerned. See also Catalano 1960: 230ff. 89 In fact, we only know of two texts that are anywhere near specific: Livy periochae 89; (Sulla) pontificum augurumque collegium ampliauit, ut essent quindecim; Servius, ad Aeneid, 6, 73 (restored): primo duos librorum fuisse custodes, inde decem usque ad tempora Sullana, inde quindecim uirorum uocabulum; Ps. Victor, de uiris illustribus 75: numerum sacerdotum auxit. It is from texts concerning the reforms of Caesar that we are able to deduce that the number of decemvirs went up to 15, and triumvirs to 7: Dio Cassius, 42. 51. 4; 43. 51. 9; cf. Wissowa (1912₂: 485 n. 5). 90 If we say that the number of normal magistrates was increased from 23 (10 quaestors, 4 aediles, 7 praetors, 2 consuls), to 34 (20 quaestors, 4 aediles, 8 praetors, 2 consuls), and therefore went up by a third, it is interesting to observe that the number of priests, which goes from 31 (9 pontiffs excluding flamens, 4 augurs, the decemvirs and triumvirs) to 52, also increases by roughly a third. In a similar vein, we know that the number of senators doubled. Mommsen (1887: 29) thought that Sulla had gone right to the extreme and abolished the election of the pontifex maximus. We have to accept, though, that there is no available evidence for this. I am inclined more towards the view expressed by Taylor (1942: 421–4), namely that Sulla did not dare interfere with this procedure for election.

66

The Priest and the Magistrate

measures suggest that the reform of sacerdotal institutions in fact follows the same line of thinking as his reform of magistracies. The quarrel over the augurate of Sulla, his conception of the auspices, and his policy in general, do not in any way go against this interpretation of priestly reform; quite the opposite, in fact. In 63 the lex Labiena gives tribes the right to elect the priests of the four major colleges, testifying to a desire to go back to the privileges established at the end of the previous century following the Sullan interlude, which, as stated earlier, is not fundamentally contradicted by that desire. But it is the position of Cicero that most clearly expresses the two elements involved in the debate. While in De Legibus, observing the collapse of the political system, he strongly affirms the jealously guarded autonomy of the priests, and the augurs in particular, he also formulates very markedly the new conception of the priest and even the balance of the two powers. Indeed, as we have seen, in De Domo Sua, Cicero brings the religious authority of a priest into alignment with his political rank,91 and downgrades the priest’s actual authority as such. Henceforth, the complete priest is to be an eminent consular, bringing together within his own person two kinds of dignity, two kinds of authority, of which the authority bestowed by comitial election and by the exercise of the imperium has precedence. Likewise, in the same speech Cicero distinguishes between the ius religionis and the ius rei publicae (in the narrow sense), as if wishing to convey a new balance between the two complementary remits of supreme power,92 which he celebrates in its traditional form in the Laws. We are not in a position to specify what Caesar’s role was here. We know that he slightly increased the number of priests, and we can relate this increase to the rise in the number of magistrates. However, it is with Augustus that the new conception of power most clearly emerges. Recent studies have brought to light Augustus’ involvement in recruitment to the senate, that is to say, the creation over the first decades of his reign, especially between 18 and 13 , of a senatorial order distinct from the equestrian order.93 It is in the same spirit that the princeps reorganized, from 29  on, recruitment to the priesthoods, which were shared out amongst the senators, the knights, and for the local cults, a little later on, even amongst the freedmen. In other words, Augustus clearly made 91 Dionysius of Halicarnassus presents the election of the early priests from the same perspective; see n. 67. 92 Catalano (1974: 671); see. n. 4 in relation to the text in question, Cicero, De Domo Sua 12. 32. 93 See n. 20.

The Priest and the Magistrate

67

priestly recruitment subject to censitary criteria, viewing the priesthoods along similar lines to the magistracies. Long left out of the censitary system, the priesthoods are henceforth part of the edifice of orders and public responsibilities. This represents the logical conclusion to an evolution in thinking that began in the second, and even third century .94 Did priestly power suffer as a result? At first glance, no, because in a society where censitary criteria were clearly very widespread, the matching up of priesthood with magistracy recruitment95 could only bolster and enhance the prominence of the priesthoods, which enjoyed an authority analogous to that offered by comitial election. But is this analysis correct? We all know what became of the magistrates from Caesar and Augustus on. The fact that they hold the first social rank cannot conceal their demotion at a political level. From being responsible and independent magistrates, within the limits set by laws an customs, they in effect became councillors, executors, a pool of high-level auxiliaries. As soon as the Republic had a man in permanent possession of a superior imperium and the exclusive right to the supreme auspices, the imperium and the auspices were by force of circumstance diminished. When the first ‘magistrate’ of the Republic was also a pontifex maximus, and alone enjoyed the supreme auspices, when his agnomen and his victories testified to the inviolability of those auspices; when, furthermore, he belonged to all the important priesthoods, and could thus both refer decisions to the colleges and debate within them, while, thanks to his auctoritas, imposing his view, the other priests were demoted to the rank of councillors in sacred law or liturgical assistants. In short, Mommsen is right to say that the radical separation of religious and of civil power ends the moment that the supreme magistrate regains the full breadth of royal power.96 Of course, this transformation of institutions did not happen overnight, either for the magistrates or for the priests, if only because Augustus only became pontifex maximus in 12 . Nonetheless, the intention and driving principle was apparent from 36 on. Indeed, from 94 The passage cited earlier from the speech of Claudius (see n. 24) fully confirms what we are saying here. It is interesting to note in this context that during the Empire the epigraphical cursūs use two procedures to celebrate the priesthoods of the honoured figure: one traditional, placing side by side at the head of the cursus itself the magistracy (the consulate) and the priesthood, the other in keeping with the new mindset, inserting the religious function within the cursus in its chronological place. 95 Sacerdotal recruitment was also formally matched to that of magistracies: the appointment was made by the senate, and only the renuntiatio was made before the comitia, in the present case the 17 tribunes. See Mommsen 1887: 29–30. 96 Mommsen 1887: 11–13; Pais 1915a: 1. 273–335.

68

The Priest and the Magistrate

36 , the young Caesar made sure to have his supporters co-opted into the pontifical college, over which he thus had complete control following the victory at Actium.97 By contrast, the political importance of the augurs was considerably diminished by the new conception of the auspices. Once Octavian, himself an augur, became Augustus, the presence within him of this almost miraculous ‘fullness of power’, equivalent to the power invested by the permanent, favourable auspices, combined with the exclusive possession of the auspices,98 allowed him to dispense with what was previously the essential collaboration of the augural college, or in any case its autonomy was severely curtailed. In short, in one way or another, all the great Republican priesthoods were subordinated to the prince, who became the sole source of religious and political law and power. The priestly colleges once more became what they were—according to our sources—in the Royal period: auxiliaries to a magistrate-priest. With this loss of sacerdotal independence was also lost the principle of the distribution of power between two interconnected but formally separate poles, the sacrum and the publicum, and within the sphere of the sacred between the different priestly colleges. It is in this context that we may understand the formulation of Quintilian in Institutio oratoria 2. 4. 34. Quintilian makes a distinction between three kinds of law, sacred, public, and private laws, and sets himself in opposition to the definitions of Cicero and Ulpian, in the sense that these laws are now more juxtaposed than inextricably intertwined. They are henceforth two fields for the application of imperial power, just as with private law, even if the ancient hierarchy is maintained. The priest’s place in society and politics thus evolved in ways closely determined by political evolution. As soon as the wish for a stronger power even began to make itself felt in Rome, we see attempts to suppress religious power, and as a result, a concentration of the opponents of these initiatives around these same institutions, which they wanted to defend precisely because they presented an obstacle to the establishment of a strong personal power. Whether it is the attempt at democratization, followed by the conservative reaction of Sulla, or the quarrel between triumvirs, priesthoods were always at the heart of the conflict. The motivations may seem different, opposed even, but the outcome was always the same: whether it is ‘populares’ trying to reduce the power of a restricted elite, or an imperator anxious to establish his personal authority, each time the priestly powers were affected, and each time the sacrum lost

97 98

See Scheid ANRW 2, 16: 273–335. See Gagé 1930, 1931; Dumézil 1969: 273–335.

The Priest and the Magistrate

69

some of its deep-seated autonomy. Priests thus became more and more like magistrates, and the distinction between priest and magistrate became less and less complete, both at a formal level and in factual reality. All this goes to show just how far this principle of the ‘collegiality’ of Roman power, the inevitable interconnectedness of two laws, two independent powers, the sacred and the public, were bound up with the Republican constitution, and were the necessary issue at stake in the conquest of power. Deeply rooted in the Roman mindset, this founding structure of the Republican constitution then becomes the principle of Imperial power, changing in form but not in substance. What changes is the place of priests. At the height of their glory in the free Republic, where they were one of its main agents, they begin to fade into background at the very point that the sacred becomes the prime area of jurisdiction for imperial power.99

A FT E R W O R D I remain essentially in agreement with my 1984 study. The approach was above all to insist on the connection between priests and magistrates insofar as they expressed a public power; this approach was by way of reaction against a superficial assimilation of Roman priests to those of the Christian Church. This return to Mommsen and to the appropriate categories of Roman public law is accepted today by the majority of colleagues and has been developed by certain studies. I would, however, add to this account a warning that if we are researching the operatives of Roman public religious life, we should not forget the magistrates themselves. For the priests were far from being the only celebrants of the religious life of the Roman state: it was consuls, praetors, aediles, or censors—amongst others—who offered the great public sacrifices, just as they did the vows for the year’s beginning, the triumphs or the Roman Games. It follows that the Roman priests, even if they didn’t follow a specific training, were counsellors, experts and preservers of tradition rather than celebrants of cult, even though it did happen that they too might celebrate this or that public ceremony. This same difference emerges equally if one considers the system of the auspices, since it was undeniably the magistrates before all who took the auspices, but the augurs also did so in the case of their own operations, for the installation of temples or for the inauguration of certain priests. But this represents an infrequent though important part of their duties; far more frequent was their assistance to the magistrates when taking the auspices, or the advice they offered to the senate on these questions. And if it was 99 These reflections have benefitted from the advice and comments of Ségolène Demougin, Jean-Louis Ferrary, André Magdelain, Claude Nicolet, Philippe Moreau, and Y. Thomas, to whom I offer my most sincere thanks.

70

The Priest and the Magistrate

the (quin)decemviri who consulted the Sibylline Books and produced the required Sibylline oracle, it was the consuls who asked the senate for the consultation to take place. The prophecy announced by the priests to the senate was then debated by the senators before its recommendations would be announced by the consuls, which did not prevent the (quin)decemviri from supervising the application of the oracle and, ultimately, celebrating the prescribed rites. It is not right therefore to forget the magistrates’ participation in the practice of public religion. The volumes published under the direction of Rüpke (2005a; 2005b) have brought together a new priestly prosopography, which greatly exceeds the chronological space covered by this article, which is also limited in its scope to the priests of the respublica populi Romani. Useful information is to be found on the sacerdotal institutions, given that one disregards the theories derived from protestant theology, which in its search for a universal definition of religions and of the priestly role, tends to neglect the ritual and juridical world specific to Roman priests. Also to be found in these volumes is the subsequent bibliography on the priestly careers and their powers. In 2002 F. Van Haeperen published an important study of the pontifical college, one part of which concerns our period (2002). Apart from Rüpke’s monumental three-volume work and Van Haeperen’s synthesis, numerous other books and articles have appeared, of which the following are mentioned here, being concerned with the same period, chronologically and thematically. A general survey on the priests of the period is in Beard and North (1990: 17–71); Beard, North, and Price (1998). Otherwise, a number of studies have been devoted to priestly authority by Santangelo (2013a) and Berthelet (2011/1: 119–289 and 2020: 121–43); and to the auspices by Hurlet (2001: 155–80); Santangelo (2013b); Berthelet (2015); and Van Haeperen (2012: 72–112 and 2015: 225–30).

B I B L I OG R A P H Y Bardt, C. (1871). Die Priester der vier grossen Collegien aus römischrepuklikanischer Zeit. Berlin, 1871. Beard, M., and J. North (1990). Pagan Priests. Religion and Power in the Ancient World. London, 17–71. Beard, M., J. North, and S. Price (1998). Religions of Rome. Vol. 1: A History. Cambridge. Berthelet, Y. (2011). ‘Légitimer les experts religieux, sous la République romaine’. Hypothèses 2011/1 (14): 119–28. Berthelet, Y. (2015). Gouverner avec les dieux. Autorité, auspices et pouvoir, sous la République romaine et sous Auguste. Paris. Berthelet, Y. (2020). ‘De la différence entre l’auctoritas des prêtres et celle des magistrats sous la République romaine’, in J.-M. David and F. Hurlet (eds.), L’auctoritas à Rome. Une notion constitutive de la culture politique. Ausonius Scripta Antiqua 136. Bordeaux: 121–43.

The Priest and the Magistrate

71

Bleicken, J. (1957a). ‘ “Oberpontifex und Pontifikalcollegium”: eine Studie zur römiscjhe Sakralverfassung’. Hermes 85: 345–66. Bleicken, J. (1957b). ‘Kollisionen zwischen sacrum und publikum’. Hermes 85: 446–80. Broughton, T. R. S. (1950–60). The Magistrates of the Roman Republic. New York. Cardauns, W. (1976). M. Terentius Varro Antiquitates Rerum Divinarum, 2. Wiesbaden. Catalano, P. (1960). Contributi allo studio del diritto augurale. Turin. Catalano, P. (1962). ‘Per lo studio dello ‘ius divinum’. SMSR 32: 133–4; 137–8. Catalano, P. (1974). ‘La divisione del potere in Roma. (A proposito di Polibio e Catone)’. Studi Grosso 6: 667–91. Chastagnol, A. (1971). ‘Les modes d’accès au Sénat romain au début de l’Empire: remarques à propos de la Table claudienne de Lyon’. BSAF 1971: 282–310. Chastagnol, A. (1973). ‘La naissance de l’ordo senatorius’. MEFRA 85: 583–607. Cornell, T. J. (1981). ‘Some observations on the crimen incesti’, in Le délit religieux. Rome. Dumézil, G. (1969). Idées romaines. Paris. Dumézil, G. (1970). (= English trans. of) Religion romaine archaïque. (1966 version). Dumézil, G. (19742). Religion romaine archaïque. Paris. Fabia, P. (1927). La table Claudienne de Lyon. Lyon. Fears, J. R. (1975). ‘The coinage of Q. Cornificius and late Republican augural symbolis on late republican denarii’, Historia 24: 592–602. Girard, P.-F. and Senn, F. (eds.) (1977). Les lois des Romains. Vol. 2. (7th edn). Camarino. Hahm, D. E. (1963). ‘Roman nobility and the three major priesthoods, 218–167 B.C.’, TAPA 94: 76–7. Hardy, E.-G. (1913). The speech of Claudius on the adlection of Gallic senators’. JPh 32: 79–95. Howe, G. (1903). Fasti sacerdotum P.R. publicorum aetatis imperatoriae. Halle. Hurlet, F. (2001). ‘Les auspices d’Octavien/Auguste’. CCG 12: 155–80. Kaser, M. (1949). Das altrömische Jus. Göttingen. Klose, A. (1910). Römische Priesterfasten. Diss. Breslau. Kübler, B. (1933). RE s.v. munus, cols. 649–50. Kunkel, W. (1962). Untersuchungen zur Entwicklung des römischen Kriminalverfahrens in vorsullanischer Zeit. Munich. Gabba, E. (1960). ‘Studi su Dionigi d’Alicarnasso I’, Athenaeum 38: 176–225. Gabba, E. (1961). ‘Studi su Dionigi d’Alicarnasso, II’, Athenaeum 39: 98–121. Gabba, E. (1964). ‘Studi su Dionigi d’Alicarnasso, III’, Athenaeum 42: 29–41. Gagé, J. (1930). La Victoria Augusti et les auspices de Tibère, RA 32, 1–35. Gagé, J. (1931). Les sacerdoces d’Auguste et ses réformes religieuses. MEFR 48, 75–108. (German trans. In Binder, G. (c.1988). Saeculum Augustum II. Darmstadt, 52–87). Gioffrodi, C. (1958). ‘Sulle attribuzioni sacrali dei magistrati Romani’. IVRA 9: 22ff. Girard, P.-F. and Senn, F (1977). Les lois des Romains, par un groupe de romanistes des ‘Textes de droit romain’. Vol. 2 (7th edn). Camarino.

72

The Priest and the Magistrate

Latte, K. (1960). Römischer Religionsgeschichte. Munich. Magdelain, A. (1978). La loi à Rome. Histoire d’un concept. Paris. Magdelain, A. (1979). ‘Li suffrage universel à Rome au Veme siècle avat J. Ch.’. CRAI 47, 698–713. Mercklin, L. (1848). ‘Die Cooptation der Römer’ Mitau-Leipzig, 115ff. Mommsen, Th. (1877). Eph Epigr 3, 100–01. Mommsen, Th. (1887–8). Römische Staatsrecht2 2.1. Leipzig. (= Droit public romain 3. Paris, 1893). Nicolet, Cl. (1974). ‘L’ordre équestre, 2, Paris. Nicolet, Cl. (1976a). Tributum, recherches sur la fiscalité directe sous la République romaine. Antiquitas 24. Bonn. Nicolet, Cl. (1976b). Le métier de citoyen. Paris. Nicolet, Cl. (1976c). ‘Le cens sénatoriale sous la République et sous Auguste’. JRS 66: 20–38. Nicolet, Cl. (1977). Rome et la conquête du monde méditerranéen. Vol. 1. Paris. Pais, E. (1915a). ‘Le relazioni fra I sacerdozi e le magistrature civili nella Repubblica Romana’, in his Ricerche sulla storia e sul diritto pubblicco di Roma. Vol. I, Rome. 273–35. Pais E. (1915b). L’elezione del pontifice Massimo per mezzo delle XVII tribù, in his Ricerche sulla storia e sul diritto pubblicco di Roma. Vol. I, 339–46. Parrish, E.-J. (1977). ‘M. Crassus Pontifex: by whose patronage?’ Latomus 36: 623–33.Pohlenz, Max (1924). ‘Eine politische Tendenzschrift aus Caesars Zeit’, Hermes 59: 157–80. Rawson, E. (1974). ‘Religion and Politics in the late second century B.C,’ Phoenix 28: 193–212. Rawson, E. (1977). ‘Caesar, Etruria and the disciplina etrusca’. JRS 67: 140. Richard, J.-Cl. (1968). ‘Sur quelques grands pontifes plébéiens’. Latomus 27: 786–801. Rüpke, J. (2005a). Die Mitglieder det Priesterschaften und das sakrale Funktionspersonal römischer, griechischer, orientalischer und jüdisch-christlicher Kulte in der Stadt Rom von 300 v. Chr. bis 499 n. Chr. Teil 3. Querllenkunde und Organisationsgeschichte, Bibliographie, Register. PAwB Altertumswissenschaftliche Beiträge, 12. 3. Stuttgart. Rüpke, J. (2005b). Die Mitglieder der Priesterschaften und das sakrale Funktionspersonal römischer, griechischer, orientalischer und jüdisch-christlicher Kulte in der Stadt Rom von 300 v. Chr. bis 499 n. Chr. Teil 2: Jahres- und Kollegienliste. PaWB, 12. 1. Stuttgart. Sabbatucci, D. (1961). ‘Diritto augurale e religione Romana’, SMSR 32: 154ff. Santangelo, F. (2013a). Divination, Prediction and the End of the Roman Republic. Cambridge. Santangelo, F. (2013b). ‘Priestly auctoritas in the Roman Republic’. CCG 17: 7–22. Simon, S. (1981). ‘A. Castricius Myrio. An overlooked Luperci (sic)’, CB 58: 12–14.

The Priest and the Magistrate

73

Szemler, G. J. (1972). The Priests of the Roman Republic. A Study of Interactions between Priesthoods and Magistracies. Coll. Latomus 127. Brussels. Szemler, G. J. (1974). ‘The Dual Priests of the Republic’. Rh.Mus. 117: 72–86. Szemler, G. J. (1976). ‘Sacerdotes publici and the ius sententiam dicendi’. Hermes 104: 53–8. Szemler, G. J. (1978). Art. pontifex in RE supp. 15, cols. 331–96. Taylor, Lily Ross (1942). ‘The election of the pontifex maximus in the Late Republic’, CPh 1942: 421–4. Treggiari, S. (1969). Roman Freedmen during the Late Republic. Oxford. Van Haeperen, F. (2002). Le collège pontifical (3ème s. a. C.-4ème s. p. C.). Institut historique belge de Rome. Études de philologie, d’archéologie et d’histoire anciennes. XXXIX. Brussels—Rome. Van Haeperen, F. (2012). ‘Auspices d’investituren loi curiate et légitimité des magistrats romains’. CCG 23: 72–112. Van Haeperen, F. (2015). ‘De la nécessité d’une loi curiate pour les magistrats sans imperium’. CCG 26: 225–30. Willems, P. G. H. (1885). Le Sénat de la République romaine: sa composition et ses attributions, Vol. 1, (1878). Louvain. Wissowa, G. (19122). Religion und Kultus der Römer. Munich. Wissowa, G. (1915). ‘Die römischen Staatspriestertümer altlatinischer Gemeindekulte’. Hermes 50: 1–33. Ziegler, K. (1914). Art. Immunis in RE, col. 1135.

II Republican Practices and Ideas

3 Acca Larentia Gains a Son Myths and Priesthood at Rome Mary Beard

Acca Larentia had twelve sons but lost one of them by death.* In his place Romulus gave himself to Acca as a son and called himself and her other sons ‘Arval Brethren’. Since that time there has always been a college of Arval Brethren, twelve in number, and the insignia of the priesthood are a garland of wheat ears and white fillets.1

With this story of substitution, Masurius Sabinus, a jurist of the Augustan age, ‘explained’ the origin of the priesthood of the Arval Brethren. Fanciful, no doubt. But were legends of this type simply quaint tales of the remote, prehistoric past? Were they not also important in constructing an image of priestly authority at Rome in the more recent, historical period? I argue in this paper that these ‘myths of priesthood’ raise important questions about the status and definition of priests at Rome. Far from being just intellectualizing speculation or quaint legends, the myths formed a medium of debate within the Roman religious system on the nature of priesthood as a whole and on the relationship of priests to the divine. Strikingly, several of these myths offer an image of priesthood— as a ‘natural’, ‘other-worldly’ phenomenon—quite at variance with the image of routinized, politicized priestly activity gained from a functional analysis of ‘what priests did’. I suggest that this disjunction is * A version of this paper was first given at the Triennial Conference of the Hellenic and Roman Societies in Cambridge in 1985. It owes a great deal to conversations with Richard Gordon, Keith Hopkins, and John North. I hope that Joyce Reynolds enjoys it, for it was she who first suggested to me that we should take Roman priests seriously! 1 Aulus Gellius, Attic Nights 7. 7. 8 (quoting Masurius Sabinus). Mary Beard, Acca Larentia Gains a Son: Myths and Priesthood at Rome In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0004

78

Acca Larentia Gains a Son

itself important as evidence of a structural tension within Roman religion—between, on the one hand, the ideological pressure to absorb religion within the public, political sphere, and, on the other, the contrary claims that religion and its officials properly belonged to the ‘other world’, the world of gods, not the world of men. In making this suggestion I take issue with what has become the current orthodoxy: namely, that Roman religion is best understood as an almost wholly political phenomenon.2 The paper is divided into six sections. The first discusses the general character of the Roman mythology of priesthood. The second offers a brief summary of the activity and functions of priests in Roman society, against which the ‘mythological images’ of priesthood will be set. Sections three, four, and five consider individual myths of priestly origins and early history. These are arranged in ascending order of complexity. They demonstrate the different types of debate on priestly function that were conducted within Roman story-telling—from the simplest questioning of the nature of priestly contact with the divine (in the myth of king Numa and the nymph Egeria), to the much more detailed and problematic discussion of priestly power in the story of the early augur Attus Naevius. The final section sets this mythological debate in a wider context, by asking how the myths of priesthood functioned, to legitimate, subvert, or refine the public religious ideology of Rome.

THE MYTHOLOGY OF ROMAN PRIESTHOOD Historians of early Rome relate colourful stories of the foundation and early history of the city’s priesthoods. They tell of elaborate religious planning by Romulus, Numa, and the other legendary kings, of miraculous interventions by the gods themselves, and of stunning and unexpected revelations of priestly skill. Characteristic is a story told by Dionysius of Halicarnassus and others of the origins of the Salian priesthood: a shield fell from heaven into the very palace of Numa; judging it to be a gift from the gods and fearing that its great value made it liable to be stolen, the king ordered a number of exact replicas to be made, to confuse a potential thief; and he created the priesthood of the salii, part of whose job was to keep all these shields and to parade them through the city in a complex dance.3 But equally marvellous tales were 2

See, for example, Wardman 1982: 1–21; Brunt 1978: 159–91. Dionysius of Halicarnassus, Roman Antiquities 2. 71. See also Ovid, Fasti 3. 361–92; Plutarch, Numa 13. 3

Acca Larentia Gains a Son

79

told of other priests and priesthoods: Vestals who saved themselves from charges of unchastity by performing miracles aided by Vesta herself;4 the quindecimviri founded by king Tarquin the Proud to have charge of the books of Sibylline oracles, which had been sold to the king by a strange woman who ‘disappeared from men’s sight’, as soon as the sale had been completed;5 the twelve Arval Brethren, who (as we have seen) were established as a priesthood under Romulus from the eleven sons of his nurse Acca Larentia—with the addition of Romulus himself, to take the place of another brother who had died. These myths of priesthood were part of a widely shared popular tradition at Rome during the late Republic and early Empire; they were not merely arcane, half-forgotten tales resurrected by historians with zealous antiquarian interests. The religious role of Numa, for example, is the subject of casual, even joking, allusions in a wide range of authors— showing clearly how this part of the priestly mythology, at least, could be regarded as ‘common knowledge’ amongst a Roman audience.6 And in the visual imagery of the city Numa and other heroes of the early priestly history of Rome are prominent. Both Pliny and Dionysius tell of a statue of Attus Naevius—the famous augur of the reign of king Tarquinius Priscus—whose miraculous exploits will be discussed below. It was a bronze figure, somewhat under lifesize, which stood just in front of the senate house, in one of the most prominent positions in the Roman Forum.7 Likewise coin types alluded to the themes of priestly myths. Particularly striking is an issue of L. Pomponius Molo of 97 , who claimed descent directly from king Numa and depicted the king on his coins as archetypal priest, offering sacrifice and holding the lituus, the augurs’ curved staff.8 The Roman concept of priesthood was formed not simply by the observable, real-life activity of the various priestly groups; it was determined also by the myths of priests and priesthoods from the legendary past. These myths were part of the common stock of Roman culture, repeatedly recalled and reworked in story-telling, literary representations, and the visual images of the city. Modern scholars of Roman religion have tended to pass over these fictional accounts of priestly history—precisely because they are fictional and offer no sure evidence on the historical origins of the different

4

For example, Valerius Maximus 1. 1. 7; 8. 1. 5; Dionysius of Halicarnassus 2. 69. Dionysius of Halicarnassus 4. 62; Lactantius, Institutes 1. 6. 10–11; Servius on Aeneid 6. 72. For example, Horace, Epistles 2. 1. 86; Juvenal 3. 137–9; 6. 342–4; 8. 155–7; Martial 11. 5. 1–4; 15. 10. 7 Dionysius of Halicarnassus 3.7 1. 5; Pliny, Natural History 34. 11. 21. 8 Crawford 1974: no. 334. 5 6

80

Acca Larentia Gains a Son

priesthoods or the character of early priestly office holding. This is to underestimate their historical value. Of course, I do not claim that the picturesque tale of the foundation of the Salian priesthood was literally true, nor that king Romulus himself established the Arval Brethren within the family of his old nurse Acca Larentia. But to ignore these stories on the narrow criterion of ‘truth’, is to miss their much more important role within the history of Roman religion, namely as a means of ‘imaging’ and understanding Roman priesthood, and as a medium of debate on priestly functions within Roman society. My aim in this paper is to explore the cultural role of these myths of priesthood;9 and, in particular, to discuss the contrast between two images of Roman priests—one, a mythological image which stresses their other-worldly connections; the other, a routinized, political image gained from an analysis of priestly organization and activity during the Republic.

PRIESTS AT ROME: ACTIVITY, FUNCTION, AND HISTORY Analysis of the activity of priests in Rome during the Republic offers three important conclusions:10 first, that Roman priestly organisation was complicated, involving many different types of official, with different religious and civic obligations; second, that the major priests (at least those on whom we are best informed) were closely integrated into the political sphere of the state; third, that priests did not act as the principal mediators between gods and men or parade a special, direct link with the divine. Each of these conclusions will be discussed briefly in turn, as each will reappear in a somewhat different form in my analysis of the mythological stories of priestly origins and early history. The organization of priests in Republican Rome was complex. There were many different groups of priests, with different tasks to fulfil and different qualifications for entry. Some (such as the Vestal Virgins11) 9 Of course, myth has many other functions—and these particular myths could be analysed in many other ways. By treating them in a very general sense as part of ‘what the Romans knew about themselves’, I am not wishing to suggest that (for example) a strictly Levi-Straussian, or modified Levi-Straussian, form of analysis would be inappropriate. That is simply not what I have chosen to do here. 10 For a detailed discussion of the nature of priestly activity in Republican Rome (justifying more fully the points made here), see Beard 1990a: 19–48. 11 For a brief discussion of the obligations of the Vestals, see Wissowa 19122: 504–10. A clear account in English is given by S. B. Pomeroy 1975: 210–14; with my own more detailed analysis, Beard 1980: 19–48.

Acca Larentia Gains a Son

81

were involved in onerous and time-consuming religious duties; others (the salii, for example, or luperci)12 undertook only one or two ritual obligations a year. Most served for life, once they had entered their office; but some (notably, again, the Vestal Virgins) had the option of resigning their priesthood after a fixed term. Most religious offices were entirely compatible with a full social and political life,13 but a few imposed such restrictions on their holders that a normal career was impossible. The most extreme case here is the notorious flamen Dialis (priest of Jupiter), whose day-to-day conduct was bounded by a rigid series of taboos, prohibiting activities as diverse as the eating of beans, the taking of an oath, and even spending more than three nights away from his bed.14 The catalogue of such differences between priesthoods could be extended at great length. It is almost as if (paradoxically) it was the main distinguishing feature of Roman priesthood that it defied generalization as a single category. The political aspect of Roman priesthood emerges clearly if we restrict our attention to the four major colleges of priests—the pontifices, augures, quindecimviri sacris faciundis, and septemviri epulonum. For these priestly groups we are relatively well informed, not only on their relations with other public bodies but also on the identity of a large number of the priests concerned. Analysis of their activity and of their lists of membership shows how closely these priestly offices were integrated with political life: priests were drawn almost exclusively from the political elite of the city, who held their priesthoods alongside the major magistracies of state and who (by the late Republic) competed for priestly office in a process of popular election in much the same way as they competed for magisterial office.15 This overlap between religious and political office holding may partly be understood as just one aspect of a more general, and for the Romans entirely ‘natural’ or ‘unargued’, integration between the spheres of religion and politics. But it was also explicitly recognized by contemporary Romans as an important, quintessentially Roman, practice. So, for example, Cicero, in the opening of a speech addressed to the college of pontifices, puts forward the identity of 12 See Wissowa 19122: 555–61. For an account in English of these two priesthoods, see Scullard 1981: 76–8 (Luperci), 85–6 (Salii). 13 Dionysius of Halicarnassus 2. 67. 2; Plutarch, Numa 10. 1–2. For a full discussion of Vestal ‘resignations’, see Beard 1980: 14 n. 21. 14 For the various taboos and regulations surrounding the flamen Dialis, see Aulus Gellius, Noctes Atticae 10. 15. 15 A form of popular election was introduced for the four major colleges in 104 ; see Cicero, De lege agraria 2. 7. 18; Suetonius, Nero 2. 1. The complicated process is clearly described by Linderski 1972: 91–2.

82

Acca Larentia Gains a Son

religious and political office holders as a paraded ideal of Roman civic life and religion: Among the many divinely inspired expedients of government established by our ancestors, there is none more striking than that whereby they expressed their intention that the worship of the gods and the vital interests of the state should be entrusted to the direction of the same individuals, to the end that citizens of the highest distinction and the brightest fame might achieve the welfare of religion by a wise administration of the state and of the state by a wise administration of religion.

Priests and politicians did not just happen to coincide; for at least some Romans, as well as for modern scholars, that coincidence was a distinguishing feature of Roman religion.16 The third major aspect of Republican priestly activity concerns the relationship of the priestly colleges to the gods. Unlike priests in many other religious systems, members of the Roman priestly colleges did not act as the principal mediators between men and gods; nor were they in a formal sense authoritative in matters of religion. They acted instead as specialist religious practitioners in the performance of ritual and as advisory religious experts, proffering advice to that institution of state most directly concerned with mediating between the spheres of gods and men, namely the senate.17 This summary is, of course, an oversimplification; and ‘mediation’ is itself a vague term of analysis.18 Many priests (as well as thousands of others) could be said to have acted as ‘mediators’, when they performed the ritual of sacrifice and so opened up a direct channel of communication between the sphere of men and that of gods. And the augurs, in particular, ‘mediated’ in the sense that they had the power to create templa on earth (not just ‘temples’ in our sense, but any piece of ground that was deemed to have direct contact with the heavenly sphere).19 Nevertheless various well-documented accounts of the problems affecting religious procedures in the late Republic demonstrate clearly the major role of the senate in controlling relations between men and gods, and the subsidiary (while expert) role of the priests. The standard procedure for the handling of prodigies provides the simplest illustration of these relative roles. At the beginning of each year, the senate considered the prodigies which had been observed and 16

17 Cicero, de domo 1. 1. See Beard 1990a: 30–4. For a discussion of the problems of the term ‘mediation’, see Beard 1990b: 8–9. 19 For the process of establishing a templum, see Varro, De lingua Latina 7. 8–10. Note that the term templum in this sense includes such buildings as the senate house and the rostra: see Cicero in Vatinium 19. 24. A full account of the problems is given by Linderski 1986: 2146–312. 18

Acca Larentia Gains a Son

83

reported during the previous year and voted on which of those were to be recognized as genuine signs sent by the gods. It then ordered immediate action to be taken in expiation, or it referred particular prodigies to one of the priestly colleges for consideration. In this latter case, the priests then reported back to the senate which (no doubt on the basis of the priestly report) took the final decision on the most appropriate course of action.20 So, in other words, it was the senate that took the prime mediating role of determining the will of the gods, while the priests offered expert advice, almost as if they formed standing sub-committees of the senate.

SOME PROBLEMS OF METHOD The formal structure of the Republican priesthoods continued under the principate. Even greater complexity was introduced by the variety of new priestly groups concerned with the imperial cult.21 Priestly office still remained deeply embedded in political life. But political life had changed with the advent of imperial government, and the pattern of religious office-holding necessarily adapted to the new regime. The most striking change was that, from the dictatorship of Caesar onwards, each emperor took a place in every one of the major colleges of priests. This ran quite counter to Republican traditions (which normally ensured that no man held office in more than one of the major priesthoods) and was compatible with a more general monopoly of religious, as well as political, roles by the emperor. So, for example, each emperor took the office of pontifex maximus as an integral part of his imperial powers; and the official iconography of Roman art came to depict the emperor, and only the emperor, in the role of public sacrificer.22 In short, the most powerful political voice in the state became also the voice of religious authority, beyond the traditional mediating power of the senate and the traditional (and continuing) religious expertise of the individual priests and colleges.

20 For a clear account of the procedure for handling prodigies, see Livy 22. 1. 8–20; 40. 19. 1–5, with Bloch 1963: 112–29. 21 Among priestly groups at Rome founded in connection with the imperial cult were: the sodales Augustales, sodales Flaviales, sodales Antoniniani. The sodales Titii were also (probably) refounded in close connection with the imperial cult. For the largely epigraphic sources on these groups, see Wissowa 19122: 564–6; Hoffman Lewis 1955: 111–59. 22 The growing dominance of the emperor both as a member of all the main colleges of priests and in public visual images, is discussed by Gordon 1989: 179–98, 201–31.

84

Acca Larentia Gains a Son

This historical development presents one immediate problem of analysis. I have chosen to concentrate on Republican priestly office holding. Yet most of the extended treatments of Roman priestly mythology (on which I shall rely) are to be found in Greek and Roman writers of the Augustan period or later. This raises a possible objection. Is not the version of the Roman mythology of priesthood in writers of the first century  bound to have been influenced by (and, in its turn, to have influenced) the changes in priestly structure brought about by imperial government? Would it not have been better to set Republican religious activity directly against Republican texts—or, failing that, to have contrasted the early imperial texts with priestly office holding in the early Principate rather than the Republic? If we had full documentation of priestly myths and priestly activity for both periods, it would obviously be the safest course to restrict our focus to either the Republic or the Empire. It is well known that authors such as Livy and Dionysius adopted a specifically Augustan perspective on a variety of themes taken from the history of early Rome; and it would be, at the very least, plausible to imagine that their stories of the earliest priests were similarly cast under the influence of the new imperial ideology.23 But we do not have full documentation. On the one hand, although there are allusions and brief references to the various myths of priesthood in surviving Republican authors, no extended narratives of that mythology survive from the earlier period. On the other, the day-today operations of priests in the early Principate are much less well known than those in the Republic; and the precise stages of the development of priestly structures over the first century  and of the growing monopoly of priestly power by the emperor remain quite uncertain, even though the general direction of that development is clear enough.24 We thus have no option but to take the material where it survives in quantity and so to consider the traditional Republican priesthoods against a priestly mythology drawn from early imperial authors. This procedure is justified by such traces as do survive of the priestly myths of the Republic. No doubt there were changes of emphasis and substance in the stories of early priesthood as the Republic gave way to the Principate, and as the imperial system gradually brought about changes in priestly organization and power; no doubt contemporary understanding of the myths shifted, along with the political position of 23

On the Augustan perspective of contemporary history-writing, see, for example, Syme 1959: 27–87 = Syme 1959: 400–54; Gabba 1984: 61–88. 24 For different approaches to priesthood under the principate, see Gordon 1989: 179–98, 201–31; Hoffman Lewis 1955: 111–59; Scheid 1978: 610–54.

Acca Larentia Gains a Son

85

priesthood in Roman society. But as far as we can tell from the scanty remains of Republican evidence, these changes were not particularly marked. I have already mentioned, for example, the coin of L. Pomponius Molo (97 ), which depicts king Numa in his role of founder-priest, a role treated at length by Dionysius, Livy, and Plutarch. Other, literary, references—a fragment of Ennius and a brief aside in Cicero’s De legibus— allude to a related aspect of Numa’s legendary career, which will be discussed fully in the next section: namely, the divine inspiration he supposedly drew from the nymph Egeria.25 And others, for example, refer to Acca Larentia, sometimes thought of as Romulus’ nurse, the mother of the Arval Brethren.26 If major changes occurred in the mythology, they occurred slowly. There is no evidence to suggest that the principal themes of the priestly myths that will form the focus of this paper underwent serious modification between the first century  and the first century , or that it is grossly misleading to juxtapose (as I shall do) the surviving Augustan mythological accounts and Republican priestly activity.

NUMA MEETS EGERIA: THE PROBLEM RAISED King Numa was, according to legend, the founding father of institutional priesthood at Rome. Certainly, some priesthoods were thought to have been founded after his reign; others27 before his time, in the reign of Romulus;28 still others, like (according to some authorities) the Vestals, were given a history, which seemed, paradoxically, to place their origin before the foundation of the city itself.29 But in most Roman accounts, Numa was the key figure in the establishment of the formal structure of priestly office holding. So, according to Dionysius, he divided religious 25

Ennius, Annales 2. 1 (fgt. 113 Skutsch); Cicero, De legibus 1. 1. 4. Cicero, Ad Brutum 23. 8; Cato, fgt. 16 (Peter) (in Macrobius, Saturnalia 1. 10. 16); Valerius Antias fgt.1 (Peter) (in Aulus Gellius, Noctes Atticae 7. 7. 5–7); Varro, De lingua Latina 6. 23. 27 Apart from the VIIviri Epulonum, established in 196  (Livy 33. 42. 1), note the XVviri sacris faciundis (originally IIviri) in charge of the Sibylline Books, reputedly founded in the reign of Tarquin the Proud. See above, n. 5. 28 For the stories of the religious foundations of Romulus, see for example Dionysius of Halicarnassus 2. 21–2; Plutarch, Romulus 22. 1–2. 29 So, for example, Rhea Silvia, the mother of Romulus and Remus was often portrayed as a Vestal (Livy 1. 3. 10–4.2; Plutarch, Romulus 3. 2–4). She was, strictly speaking, a Vestal of Alba Longa—and so in a sense both Roman and not-Roman at the same time. Compare Plutarch, Romulus 22. 1–2, which recounts Romulus’ establishment of the Lupercalia (and so also the Luperci) on the pre-Roman site of Rome; see Livy 1. 5. 1–2. 26

86

Acca Larentia Gains a Son

observances into eight parts and assigned priests to each of the divisions: the curiones (to perform sacrifices in the thirty regions, curiae, of Rome); the flamines; the commanders of the celeres (bodyguards of the king, with a military religious role); the augurs; the Vestals (granted for the first time an established temple in the city); the salii; the Fetials (to take charge of the religious aspects of Rome’s relations with the outside world); and the pontifices (with a wide competence over all kinds of religious observance).30 The inspiration for Numa’s programme of priestly foundations and other reforms was said by some ancient authors to have come from the divine nymph Egeria, whom he used to meet by night and with whom he even enjoyed a ‘celestial marriage’. Dionysius again explains: Men relate many marvellous tales about him, attributing his human wisdom to suggestions of the gods. For they tell the story that a certain nymph Egeria used to visit him and instruct him on each occasion in the art of kingship; although others say that it was not a nymph, but one of the Muses.31

Dionysius then describes a miraculous banquet provided by Numa for an enormous company, which so astonished the Romans of his day that they were firmly convinced that their king had some direct contact with the gods; and he continues: But those who remove all fabulous elements (muthōdē) from their history say that the report concerning Egeria was invented by Numa so that, when once the people were possessed with a fear of the gods, they might readily pay regard to him and willingly receive the laws he should enact, as coming from the gods.32

The striking feature of this account is its incorporation of two apparent variants on Numa’s relations with the gods: the first, that he really did meet Egeria and could prove his divine contacts by the performance of a miracle; the second, that Numa invented his meetings with the nymph, in order to impress the people and make them compliant to his programme of reform. These variants are not confined to Dionysius’ account. Plutarch likewise admits both the possibility that Numa really did have meetings with Egeria: Is it not likely that the gods are in earnest when they hold converse with men such as these, in order to instruct and advise them in the highest and best way . . . ?33

30 32

Dionysius of Halicarnassus 2. 64–73. Dionysius of Halicarnassus, 2. 61.1.

31 33

Dionysius of Halicarnassus 2. 60. 4–5. Plutarch, Numa 4. 7.

Acca Larentia Gains a Son

87

and the possibility that the contact was a calculated invention: Indeed there is no absurdity in the other account which is given of Lycurgus and Numa and their like, namely that, since they were managing headstrong and captious multitudes and introducing great innovations in modes of government, they pretended to get a sanction from the gods, which sanction was the salvation of the very ones against whom it was contrived.34

How are these variants to be understood? One possible interpretation of these two different views of Numa is that one is a ‘rationalization’ of the other: to the sophisticated, sceptical Graeco-Roman writers of the late Republic and early Principate, the traditional, perhaps popular, story of Numa’s meetings with a nymph could only be retained if reinterpreted as a clever fabrication by an astute monarch, skilled at winning the loyalty of the people. This interpretation is, no doubt, in part correct. Cicero, certainly, adopting a sceptical stance at the beginning of De legibus, treated belief in Numa’s conversations with Egeria as a clear sign of gullible superstition;35 and, more generally, there is plenty of evidence from the late Republic onwards that many traditional Roman beliefs were being reinterpreted by certain sections of the Roman elite in the light of Hellenizing, philosophical approaches to the divine.36 But such ‘rationalism’ may not fully explain the juxtaposition of these variant views on Numa’s divine friend. A very different interpretation of Numa’s relations with Egeria is suggested if the two versions of the myth are taken closely together. Instead of separating the two stories and, in effect, making one a sophisticated or sceptical comment on the other, we can equally well—perhaps better—read them as an integral whole. The two variants taken together, in their sharp conflict, express uncertainty on Numa’s links with the gods, and (by extension) uncertainty about the direct contact with the divine enjoyed by Roman priests in general. The myth of Numa, as presented by Dionysius and Plutarch, ‘has it both ways’, by claiming simultaneously that Numa did have direct links with the gods in the form of Egeria, and that he did not. Neither claim stands on its own, without parading also its own contradiction. On this interpretation, this brief episode in the myth of Numa heralds the theme of much of the priestly mythology discussed in the rest of this paper. An analysis of priestly activity at Rome locates Roman priests firmly in ‘this world’; but the mythology raises the question of their special links with the ‘other world’.

34 36

Plutarch, Numa 4. 8. See Beard 1986: 33–46.

35

Cicero, De legibus 1. 1. 4.

88

Acca Larentia Gains a Son THE SALII AND THE ARVAL BRETHREN: THE AMBIGUITY OF ORIGIN

The suggestion of direct priestly links with the divine is raised more positively in the myths of the Salian priesthood and of the Arval Brethren; although, even here, the suggestion is tempered by various types of uncertainty surrounding the origins of the priesthoods. I have already mentioned the foundation of the Salii by Numa after a shield had miraculously fallen from heaven; and the origins of the Arvals under Romulus, formed from the eleven sons of Acca Larentia with the addition of Romulus himself.37 How do these stories relate to the problem of priestly contacts with the gods? The myth of the origin of the Salii makes an obvious claim for the priests’ special relationship with the other world; while at the same time that claim is offset by a counterbalancing emphasis on the human origin of the priesthood. This ambiguity (while different in detail and weighted more heavily towards the divine status of the priests) is broadly compatible with the ambiguity I have indicated in the story of Numa and Egeria. On the one hand, the myth asserts a ‘natural’ divine connection for the Salian priests, in that their duty is to look after a shield which fell directly from heaven, sent by the hand of a god. On the other, the myth itself undercuts such a clear link with the divine, by stressing that the priests also have in their charge numerous replicas of the divine object itself, made entirely by the hand of man. These claims and counterclaims can best be appreciated if we imagine the scene of the twice yearly ritual dances of the Salii, as they processed through the city carrying, amongst other things, these famous shields.38 The Roman observer who ‘knew’ the origin of the priesthood and of the shields would know that one of these weapons had come to earth as a gift from heaven; but he would also know that that divine object was indistinguishable, and purposely so, from the other shields carried by the priests. He would recognise a direct link between the gods and the priests who looked after their gift; but, just as he would be unable to distinguish the gods’ handiwork from the product of human craft, so he would be unable to define that link precisely. The priests both were and were not seen as directly and naturally in contact with the other world. The myth of the origin of the Arval Brethren is complicated by the ancient debate on the identity of Acca Larentia. The story of the

37 38

For references to the ancient accounts, see above n. 3 (Salii), with n. 1 (Arvals). For details of the rituals of the Salii, see Scullard 1981: 85–6.

Acca Larentia Gains a Son

89

priesthood which I have already quoted is drawn from Aulus Gellius, who preserves the view of Masurius Sabinus that Acca Larentia was the nurse of the first king of Rome. But other Roman authors offered other views. Some saw Acca Larentia as a prostitute in the primitive city of Rome, who left the vast fortune she had made from her trade to the Roman people (or, alternatively, to Romulus himself) and was honoured for her generosity by a public sacrifice and a regular festival in the religious calendar (the Larentalia). Others claimed that she was given as a prize to Hercules by a temple-caretaker (aedituus) who had challenged the god to a game of dice and lost; after sleeping with Hercules, she married a wealthy, childless man and eventually left the property she had inherited from him (as in the previous version) to the Roman people. Still other versions saw a connection between the nurse of Romulus and the renowned courtesan—stressing that the Latin word ‘lupa’ could mean both ‘wolf ’ (the animal traditionally regarded as nurturer of the infant Romulus and Remus) and, colloquially, ‘prostitute’.39 The version of Masurius Sabinus is the only one surviving to link Acca Larentia explicitly with the Arval Brethren; and it has much in common with the myth of the foundation of the Salii. The majority of the legendary forebears of the Arval priests are presented (like all but one of the shields of the Salii) as entirely human—the sons of a human nurse; but one (like the shield that fell from heaven) was Romulus, a man of divine birth, later to be reincorporated among the deities as the god Quirinus.40 The original priests are thus seen to have a clear link with the gods and to be part of the human world. But the other stories of Acca Larentia, even if not themselves explicitly connected to the foundation of the Arvals, seem to set that underlying ambiguity in a yet more extreme form. On the one hand, the mother of the Arvals is presented not merely as human, but as a particularly base form of humanity—a prostitute. On the other, she is presented as divine herself: not merely the sexual partner of a god, but finally the recipient of divine honours in one of the major public festivals of the Roman calendar. The nexus of myths around the

39 See especially Aulus Gellius, Noctes Atticae 7. 7. 5–7 (the wealthy prostitute); Plutarch, Romulus 4–5; Plutarch, Roman Questions 35 (suggesting two different characters of the same name: one the nurse of Romulus; one a prostitute, given as a prize to Hercules); Livy 1. 4. 7 (suggesting a link between the nurse of Romulus and lupa (she-wolf) in the sense of ‘prostitute’). For further references and discussion, see Sabatucci 1958: 41–76; Scheid 1975: 352–64. 40 For Romulus’ metamorphosis into a god, see Livy 1. 16, with the further discussion of Ogilvie 1970: ad. loc., at pp. 84–7. John Scheid has (rightly) stressed to me the significance of the ambiguity of Romulus’ status, as neither fully human nor fully divine.

90

Acca Larentia Gains a Son

figure of Acca Larentia offered one of the most elevated, divine images of priestly origins, as well as one of the most human and most lowly.

ATTUS NAEVIUS: THE ARCHETYPAL AUGUR The myth of Attus Naevius is more complex than those I have considered so far.41 The hero, Attus Naevius (or Navius) was an early augur, who was believed to have lived in the reign of the fifth king of Rome, Tarquinius Priscus. He was commonly regarded by the Romans as the archetypal augur, a role model for the ‘real life’ augurs as they operated in the centre of Roman political life during the historical period, ever present in the form of his bronze statue outside the senate house. Cicero even suggests that Attus Naevius was seen by many as the Roman equivalent of the Etruscan Tages, the legendary founder of the discipline of haruspicy, the Etruscan science of divination. But Cicero himself (in the sceptical voice of the second book of De divinatione) also points out that it was illogical to regard Naevius as a founding father in the strict sense, in that the skills of augury were known at Rome long before the reign of Tarquin, as early as the establishment of the city by Romulus.42 Several versions of the extraordinary career of Attus Naevius are preserved in writers of the late Republic and early Empire.43 I offer here a summary of that career based on the clear and extended account of Dionysius of Halicarnassus: Attus Naevius was a boy of poor family, employed by his father to mind the pigs. One day he fell asleep at his job and woke up to discover that he had lost some of the animals. In desperation, fearing that his father would beat him, he vowed the biggest bunch of grapes on the farm to the ‘heroes’ (that is, the Lares) if they would help him find the pigs. This they did, but the boy realized that he now had another problem: how was he to locate the biggest bunch of grapes? He prayed to the gods for omens to help him and they responded by inspiring him to divide up the farm’s vineyard according to the rules of augury until the place where the biggest bunch grew was finally marked out. These he offered to the Lares as he had promised. But his father, who had found out what had happened, realized that his son had a

41

Important aspects of the myth are discussed by Piccaluga 1969: 151–208. Cicero, De divinatione 2. 38. 80. 43 In addition to the version that follows, see the extended accounts of Cicero, De divinatione 1. 17. 31–3; Livy 1. 36. 3–8. For further references, see Piccaluga 1969: 151–208. 42

Acca Larentia Gains a Son

91

natural talent for divination and eventually sent him to Etruria for proper training in the augural art. Attus Naevius became a celebrated augur (though not, Dionysius stresses, a regular member of the augural college); and he went on to enjoy a stunning success against king Tarquin. Naevius was opposing some political reforms of Tarquin, who therefore summoned him to the Forum in order to discredit the augur’s opposition by revealing him to be a charlatan. Tarquin asked him to determine by augury whether or not the plan that he, the king, had in his mind at the moment was possible to carry out. Naevius took the auspices and declared that the undertaking was indeed possible. Tarquin was jubilant, believing that he had wrong-footed the augur; the plan, he said, had been to cut a whetstone in half with a razor. Naevius was nothing daunted; and, since Tarquin had brought to the Forum a whetstone and a razor, he instructed the king to attempt to cut the stone. Of course, the whetstone fell in two: the razor went straight through and on, straight through the hand that held it! Tarquin was amazed. He gave up his political reforms and honoured Naevius with the famous bronze statue in the Forum. In the account of Dionysius, the end of Naevius follows shortly after this extraordinary success. It was a (predictably) mysterious end. The famous augur simply disappeared; no body was ever found.44

The theme of priestly contact with the divine should by now be easily recognizable. In this story we find a clear emphasis on the special relations of the priest with the gods: it was from direct divine inspiration that Attus Naevius drew his skill in augury, which enabled him to discover the site of the biggest bunch of grapes in the vineyard. But again we find the display of those direct contacts in some way tempered—in this case by the reaction of Naevius’ father, who recognized both the boy’s ‘natural’ talent for divination and his need for formal instruction. Augury is thus seen as dependent on the priest’s direct links with the gods; at the same time it is allowed to be in part the product of human training and skill. New issues are also raised by the detailed narrative of Naevius’ career; and these set up further important contrasts between the ‘real-life’ activities of (in this case) augurs and the mythological image of the augur presented by the legend. Two particular themes stand out: first, the miraculous and, on occasion, mysterious behaviour of Attus Naevius; second, the stance of political opposition taken by Naevius against the power of king Tarquin.

44

Based on Dionysius of Halicarnassus 3. 70–72.3.

92

Acca Larentia Gains a Son

The public activity of the augurs in the Roman Republic and their use of augury is relatively well known. Their main expertise lay in the area of divine signs and the ways in which the approval of the gods could be recognized for whatever project or legislation the state officials were undertaking; and when they marked out the heavens for the observation of birds, it was normally to provide a framework within which the gods might make their will known to Roman magistrates.45 This contrasts strikingly with Naevius’ use of augury: when he marked the sky it was in order to use the practice of augury to perform a miracle—that is, the discovery of the biggest bunch of grapes in the vineyard. His behaviour in general, at every major event in his career, confirms this miraculous tendency: in his conflict with Tarquin he brought about the ‘impossible’ slicing of the whetstone; on another occasion, as related by Pliny, he miraculously transferred the Ficus Ruminalis (the fig tree under which Romulus and Remus had been discovered) from the Lupercal to the Comitium in the Forum;46 and finally, his ‘death’ was no death at all, but a mysterious and miraculous disappearance. It is as if the legendary, archetypal augural priest was cast not simply in a role more spectacular than his real-life historical successors, but in a role that was quite different in kind. The augurs of the Republic were closely integrated with the political process. That is not to say, of course, that they did not sometimes (as individuals or a group) use their special religious status to offer objections to particular political decisions—by announcing, for example, the appearance of unfavourable omens in the midst of an assembly;47 but, by and large, the distinctive feature of the priestly organization at Rome was precisely the overlap between religious and political practitioners and the fact that there was within traditional cult no locus of priestly or religious power separate from the political sphere. The story of Attus Naevius offers a violent contrast. For the main episode of the clash between Naevius and Tarquin not only sets the priest in opposition to the political order, it effectively defines two different forms of authority— the ‘religious’ and the ‘political’—and celebrates the victory of religion over politics. The image of the priest is thus disengaged from the routinized, day-to-day, political activity of the real-life augurs, and given special, even supernatural, status.

45

For the role of the augurs, see Beard 1990a: 34–40. Pliny, Natural History 15. 20. 77. 47 Note for example Cicero’s stress on this aspect of the augurate (De legibus 2. 12. 31); he cites also some examples of augural intervention. 46

Acca Larentia Gains a Son

93

THE MYTHOLOGY OF PRIESTHOOD: FUNCTIONS AND IMPLICATIONS Throughout this paper, I have emphasized the disjunction between the mythological image of Roman priests and the image gained from an analysis of the public role of those priests during the late Republic. How is that disjunction to be understood? I want in this final section to consider three answers to that question: first (a), the disjunction is largely illusory, the result of a misleadingly narrow definition of priestly activity; second (b), the disjunction is a consequence of the partisan quality of the priestly mythology—in its aim either to subvert or to legitimate the monopoly of priestly power by the political elite; third (c), the disjunction is a feature of important debates at the centre of Roman religion, part of a characteristic tension between the political focus of that religion and its simultaneous claims to links with the other world. Of these three options, the third (c) seems to me to offer the most important and interesting insights into the apparently conflicting relations of the mythological and ‘real life’ images of priestly status at Rome; but all three, to a greater or lesser extent, contribute to a fuller understanding of the mythology of priesthood as a whole.

THE SCEPTICAL APPROACH: THE DISJUNCTION ERODED One response to the problem of the apparent incompatibility of the mythological and ‘real-life’ images of priesthood is simply to question the highly political, even routinized, image of ‘real life’ Republican priestly office holding that I have outlined. It could be argued, for example, that by basing my analysis of priestly activity on the major priestly colleges, I have paid too little attention to those priests whose public religious role and obligations did set them apart from the routine of ‘normal’ life among the political elite; the classic case here would be the flamen Dialis, who was marked out as a special, even other-worldly, figure by the extraordinary series of taboos surrounding his conduct. Or again it might be objected that, within surviving literary sources, there are traces of a public, political debate on priesthood that raises concerns very similar to those I have isolated in the mythological representations: the ‘naturalness’ of priestly skill and the possibility of direct contact between priests and the gods. Cicero’s speech De domo provides a clear example here. The main concern of this speech, delivered to the

94

Acca Larentia Gains a Son

pontifical college on the orator’s return from exile, was the religious status of the shrine of Libertas erected by Clodius on the site of Cicero’s house; Cicero was anxious to deny the validity of the consecration of the shrine, in order to rebuild his house—and so throws into question the competence of the young pontifex who had performed the consecration only a few days after his entry into the priesthood. Even in the partisan oratory of Cicero (who obviously wished to deny that the young man’s action had any religious significance), there is a clear overlap with the concerns of the first part of the story of Attus Naevius: what is the status of the novice priest? how far does priestly skill result from training? how far is it natural or innate?48 The sharpness of the disjunction between the two types of image may also be blunted by appeal to those myths of priesthood that appears to be more closely compatible with priestly activity during the Republic. These stories are not numerous; but a clear example is found in the legend of Cornelius the priest. According to Plutarch, a soothsayer (mantis) in the reign of Servius Tullius foretold that whoever should sacrifice a certain heifer at the temple of Diana on the Aventine, his city would eventually rule over all Italy. Cornelius, having heard of the prophecy, cunningly instructed the owner of the heifer—a Sabine named Antro, who was anxious to ensure the future success of his own native land—that he must bathe in the Tiber before offering the sacrifice, since this was the local custom. Antro followed these instructions; but, before he could return from his bathe, king Servius had overtaken him and had himself offered the animal in the sacrifice, thus winning the benefits of the prophecy for his city of Rome. In this story, quite unlike the others I have considered, the priest is cast as the cunning political manoeuvrer, hand in glove with the authority of the state.49 There is some force in these points. For the sake of clarity my arguments have tended to oversimplify the contrast between the different images of priesthood; and I have not explicitly recognized what must necessarily be the case, that the mythological presentation of priesthood must have interacted with, influenced, and have been influenced by the ‘real life’ activity of Roman priests. It is obviously inconceivable that a society should operate with two entirely separate and entirely different images of the same institution, one restricted to myth, the other to 48 See, for example, Cicero, De domo sua 45. 117–18, underlining the paradox of a pontifex without ‘knowledge’—though stopping short of claiming that the young man’s actions had no validity. For the difficulties associated with the idea of a trained priesthood in Rome, see Beard 1980: 12–27. 49 Plutarch, Roman Questions 4.

Acca Larentia Gains a Son

95

‘everyday life’. But even if these observations suggest the possibility, at a further stage of analysis, of a more refined account of the Roman conception of priesthood, the broad terms of the disparity I have identified remain good. Most myths of priesthood do, in fact, present an image of Roman priests at variance with the major characteristics of those officials, as seen in their public activity. That divergence demands to be explained.

THE PARTISAN MYTHOLOGY The second solution to the problem of the divergence in the images of Roman priesthood is precisely the reverse of the first: namely, to regard the claims of the priestly mythology as so consistently opposed to the operation of priests in Roman public life that they can only be read as a directed, partisan comment on ‘real life’ priestly organization and power. The question then becomes: with what aims and for whose benefit were those partisan stories told? At first sight, it might seem obvious to suggest that the priestly legends—and particularly those which claim some natural divine inspiration as the basis of priestly status—constituted an attack on the political routinisation of priesthood and the absorption of priestly power into the political power of the Roman elite. On this reading the priestly mythology was a subversive mythology, using the myth-history of Rome to challenge the model of priesthood supported by the Roman political elite. Paradoxically, perhaps, a quite different reading is also possible: the myths can be seen not as subversive, but rather as an elite defence against the charge of ‘unreasonable monopoly’ of priestly office holding. On this interpretation, the legends served to justify the consistent exclusion from Roman priesthood of men outside the political elite by parading an ideology in which the priest had—even if in an uncertain and ambiguous form—direct links with the divine. Political challenge to the elite monopoly of religious power was warded off by claims that there was, in fact, more to priestly office holding than immediately met the eye. In this way, the absorption of priestly power into the political system of Rome was upheld and defended by a mythological image of priesthood that seemed directly at variance with that politicization of religion. Both these readings of the mythology are possible and can be held simultaneously. We do not need to suggest any crude ‘manipulation’ of the legendary inheritance of Rome to recognize both the subversive and

96

Acca Larentia Gains a Son

legitimating potential of the stories of early priests. Take the myth of Attus Naevius. On the one hand, it is clear that radical opponents of the political and religious power of the Roman elite could find in their telling or reading of this story justification for their dissatisfaction with the apparently routinized conduct of the ‘real life’ augurs. On the other, a member of the augural college could interpret the career of Naevius as confirmation of his own special status as Roman priest and of the legitimacy of his exclusion of those ‘not fit’ to take the office. And there were, no doubt, other readings that changed with the perspective of the individual concerned and with the political concerns of the time. The problem with the ‘partisan’ solution is not that it leads us to incorrect interpretations of individual stories; indeed, at its broadest, it demonstrates the variety of possible readings of the legends of early priests. It is rather that the suggestion of consistent partisan aims behind the mythology of priesthood as a whole tends to lead to two particularly misleading conclusions: first, that the mythology as a whole was invented or constructed in order to advance the claims of a particular interest group; second, that the original purpose of the stories was either to subvert or to legitimate the traditional foundations of priestly power. Both these conclusions seriously oversimplify the complex process of the formation of the mythological inheritance and the multivalent, shifting interpretations that any ‘living’ mythology necessarily allows—or even encourages.50 The next section (c) will show the importance of that complexity and multivalence.

PRIESTLY MYTHOLOGY AS DEBATE The third approach to the disjunction in the Roman images of priesthood offers a more dynamic and less rigid solution to the problem, which at the same time develops some of the important implications of the previous two sections: namely, that the separation between the mythological image of the priest and public priestly activity was not in practice as strict as I have sometimes suggested in my outline of the problem; and

50 That is to say, there is no single ‘correct’ meaning in mythology. Mythology is powerful precisely because it does allow different meanings at different times, for different groups, in different circumstances, constantly negotiated and constantly under debate. It is this fluidity of interpretation that enables individual myths, while retaining a traditional form, to remain ‘meaningful’ in a society whose concerns are radically transformed over time. This issue is fully discussed in my article, Beard 1987: 1–15.

Acca Larentia Gains a Son

97

that myth, in different ways, with different emphases, could provide a response (or challenge) to the absorption of Roman priestly activity in the political sphere. In this section, these approaches to the problem are put into a broader religious and cultural context; myth is taken together with the ‘real life’ activity of priestly officials and seen as part of an extensive debate on the nature of priesthood and on the character of Roman ‘religion’ more generally. The argument rests on seeing myth and ‘real life’ activity as parts of the same religious discourse, in dynamic interrelationship one with another. One of the functions of the Roman mythology of priesthood was cognitive and dialectical: that is, in the broadest terms, it provided a medium of discussion and debate on the nature of priesthood and priestly power. This can be seen in a simple form in the issue of the differentiation and classification of the various priestly groups. At the beginning of this paper I highlighted the very different character, duties, and privileges of the different Roman priesthoods. The mythological stories I have been discussing amplified, defined, and redefined those differences. These myths were part of what the Romans knew about their priests; and the different origins and early history ascribed in the legends to the different priestly groups was one way in which the Romans could perceive and construct the vastly different statuses of their different sets of religious officials. So, for example, to claim—as the legends did—that the Salii were founded at a different date, by a different king, under different circumstances from (say) the Arval Brethren was to offer another way of understanding those priesthoods as institutions of very different types. They did not simply have different religious duties, their difference was asserted and reasserted by their different mythological origins. This principle of debate and classification can be extended to the problem of the divergence between the different images—mythological and ‘real life’—of Roman priests. It can help us appreciate why, in a general sense, we find divergence between the various claims of myth and the various claims of everyday political power; and, with more particular point, why the mythological and political claims about Roman priestly status seems so consistently in conflict. Religions are about the unknown, about the constructed reality of a world that is both here-and-now and somewhere else, both with us and other. Religions, like other social institutions (but more so), are deeply multivalent: single meanings and single significances are not sustainable; each religious claim—whether explicit or implicit, whether ‘Christ is the Son of God’ or ‘The gods promote the well-being of the Roman state’—brings with it new questions, different interpretations,

98

Acca Larentia Gains a Son

and often its own contradictions. So, in the case of Roman priests, even if the straightforward political routinization could largely be sustained within the day-to-day practice of the Roman state, that political routinization could never provide the only answer to such questions as: What do we know about priests? What do priests do? How do they relate to the gods? Why are some people priests and other not? Such questions emerged repeatedly and in ever changing forms within the Roman religious system; and one medium in which answers to the questions were constructed was the myth-history of the city. The range of answers included those which were internally contradictory (the myth of Numa and Egeria, for example), those which ran against the political operation of priesthood in Rome (such as the story of Attus Naevius) and those which, at least occasionally, seemed to confirm the political stress (the story of Cornelius the priest). These contradictions and confirmations are equally part of the way religion talks about itself, part of a deeply evocative religious system which necessarily throws up more answers and meanings than it (or we as observers) can assimilate.51 It still remains to ask how and why, even within such a framework of debate, the mythological image seems so consistently, if not universally, discordant with the everyday images of politicized priesthood. Certainty is impossible; but, in my view, the structural tensions inherent in the political sphere must lie at the root. This tension is centred on the impossibility of religion being entirely absorbed into politics—however much it appears to be simply the plaything of the government or governing class. Religion is only ‘religion’ if it is about the other. The claims to political monopoly of religion only make sense if religion is partly about something politics is not. Much recent work on Roman religion has stressed the public political aspect of Roman state cult. In a reaction to earlier scholars who judged Rome’s religion inadequate by the standards of private and personal commitment familiar from Christianity and other world religions, it has become fashionable to endorse the sometimes strident claims of Cicero and his contemporaries that religion was ‘about’ public life and that the prime function of state cult was to ensure the success and prosperity of the Roman state.52 These claims were both true and, at the same time, an effective oversimplification. Cicero exercised himself to assert religion’s place in the central political arena of the city partly because it was so difficult to contain it there; partly because it was, as 51

For a sense of this complexity, see Gordon 1980: 1–99. My earlier claims to this effect in Beard and Crawford 1985, 31, were well intentioned, salutary at the time—but, as I now see it—too crudely stated. 52

Acca Larentia Gains a Son

99

everyone ‘knew’, not just about politics at all, but about the other world, the strange and the claims to reality that could only be upheld by reference to the divine. It was the delicate balance between the here-and-now, public stage and the Other that gave religion its continuing resonance. In the case of priesthood and its mythology, the legendary stories of the early priests (which consistently made claims to ‘otherness’) played an important part in upholding the ‘religious’ aspect of ‘religious’ officials, in necessary counterpoise to the opposing tendency to ‘secularization’.

B I B L I OG R A P H Y Beard, M. (1980). ‘The sexual status of Vestal Virgins’. JRS 70: 12–27. Beard, M. (1986). ‘Cicero and divination: the formation of a Latin discourse’. JRS 76: 33–46. Beard, M. (1987). ‘A complex of times: no more sheep on Romulus birthday’. PCPS 33: 1–15. Beard, M. and Crawford, M. (1985). Rome in the Late Republic. London. Beard, M. and North, J. (1990). Pagan Priests. London. Beard, M. (1990a). ‘Priesthood in the Roman Republic’, in Beard and North (eds.), Pagan Priests, London, 19–48. Beard, M. (1990b). ‘Introduction’, in Beard and North (eds.). London, 1–14. Bloch, R. (1963). Les prodiges dans l’antiquité classique. Paris. Brunt, P. A. (1978). ‘Laus imperii’, in P. D. A. Garnsey and C. R. Whittaker (eds.), Imperialism in the Ancient World. Cambridge, 159–91. Crawford, M. H. (1974). Roman Republican Coinage. London. Gabba, E. (1984). ‘The historians and Augustus’, in F. Millar and E. Segal (eds.), Caesar Augustus: Seven Aspects. Oxford, 61–88. Gordon, R. L. (1980). ‘Reality evocation and boundary in the Mysteries of Mithras’. Journal of Mithraic Studies 3, 19–99. Gordon, R. L. (1989). ‘From Republic to Principate: priesthood, religion and ideology’, in Beard and North (eds.), 1989, 179–98. Gordon R. L. (1989). ‘The veil of power; emperors, sacrificers and benefactors’, in Beard and North (eds.), 1989, 199–231. Hoffman Lewis, M. W. (1955). The Official Priests of Rome under the JulioClaudians: A Study of the Nobility from 44  to  68. MAAR 16. Rome. Linderski, J. (1972). ‘The aedileship of Favonius, Curio the Younger and Cicero’s election to the augurate’. HSCPh 76: 181–200 (reprinted in Linderski, J. (1995). Roman Questions, 231–50). Linderski, J. (1986). ‘The Augural Law’. ANRW II.16.3: 2146–312. Ogilvie, R. M. (1970). A Commentary on Livy, Books 1–5. Oxford. Piccaluga, G. (1969). ‘Attus Navius’. SMSR 40: 151–208.

100

Acca Larentia Gains a Son

Pomeroy, S. B. (1975). Goddesses, Whores, Wives and Slaves. New York. Sabbatucci, D. (1958). ‘Il mito di Acca Larentia’. SMSR 29: 41–76. Scheid, J. (1975). Les frères Arvales. Paris, 352–64. Scheid, J. (1978). ‘Les prêtres officiels sous les empereurs julio-claudiens’. ANRW II. 16. 1: 610–54. Scullard, H. H. (1981). Festivals and Ceremonies of the Roman Republic. London. Syme, R. (1959). ‘Livy and Augustus’. HSCPh 64: 27–87 (= Syme, Roman Papers I (1979), 400–54). Wardman, A. (1982). Religion and Statecraft among the Romans. London. Wissowa, G. (19122). Religion und Kultus der Römer. Munich.

4 Religion and War On the Relationship of a Society’s Religious and Political Systems Jörg Rüpke

INTRODUCTION At the end of the twentieth century the understanding of war as the continuation of politics by other means no longer seemed to be a phenomenon of the nineteenth century. Small, ‘sub-atomic’ wars are possible on European soil: in the play of power politics, territorial gains achieved by force—an outcome typical of how highly organized states go about war—earn a new legitimacy. Nevertheless, it is precisely when the threshold for waging war is lower that the difference between war and politics is demonstrated: ‘other means’ consist first and foremost in the activation of a military apparatus that under certain circumstances unleashes its own dynamic that slips out of the hands of the previously existing political means of control. In German politics the expansive war aims of the Prussian, soon to be Prussian-German military, after the initial success of the war against France, are a ‘good’ example of how this state of affairs comes about, even within a stable political framework.1 However, war should not be identified simply with the military, it should not be reduced to conflict pure and simple. With the increasing mechanization of warfare, the importance of long-term preparations for warfare grow; but at the political level, too, war is presented rather as just one means of resolving conflict (if also an extreme one) which can only be interpreted within the framework of the conflict as a whole, of the solutions that 1

See Nipperdey 19932: 65f., 71–3.

Jörg Rüpke, Religion and War: On the Relationship of a Society’s Religious and Political Systems In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0005

102

Religion and War

have failed and of the apparent strategies of escalation. And so too of de-escalation: wars leave traces behind and have consequences, even if the only form they take is the requirement to demobilize. Nor, against this background of the interconnection of warfare and politics, should the relationship of war and religion be investigated as a strange specimen that invites our attention by means of particularly bloody rites or the overt opposition of religion. The question to be posed in this contribution is much more along the lines of asking which are the political functions that religion observes under the stresses that war brings about in the political system, indeed in society as a whole? That there is a perception of such functions already implies a functionalistic concept of religion, in that it shows religion’s ability to deal with contingency and its integrative capacities. What also emerges is an image of religion and war as mutually reinforcing. This ought not to be taken as a definitive statement: it is a consequence of the methodical approach and it can be disproved as easily as it can be proved in plenty of individual cases. Thus, only actual historical investigation is of interest: at which points in war’s course, in what forms, and above all, with which segments of the complex system that is ‘religion’—that is, on whose behalf—are political functions to be perceived?

‘HOLY WAR’ For analyses of these kinds, the concept of ‘Holy War’, a concept that seems to denote the identification of the political and religious systems in particular ways, seems only of limited use in the case of war and religion. It is indisputable that religious motivations, and conflicts between religions and confessions, repeatedly crop up as the causes of war. But such a picture is deceptive. For historical reasons the boundaries between religions are often overlaid with social, economic, and ethnic fault lines; they are consequences of processes of subjugation or stratification. Furthermore, when there is great pressure to provide legitimation for waging war the recourse to religious symbols, with their claim to finality, offers great advantages. Above all this applies precisely where the parties to a conflict do not correspond to the dividing lines between political systems or states. ‘War of Religion’ is a more familiar concept for such conflicts: it suggests a predominance of religious motives over the interests of power politics or economics—a predominance that often does not exist in individual cases. Even where the crusades and the wars of the European

Religion and War

103

Reformation are concerned, non-religious motivations should not be underestimated.2 The Islamic concept of jihad seems to belong clearly in the field of religiously motivated wars. This kind of war is targeted at ‘unbelievers’ and conducted for the ‘strengthening of [the] religion’. But a closer look at how it was discussed in Islamic legal scholarship shows that in no way does this fit seamlessly into the category of ‘religious war’, as outlined above. To begin with it is to be established that what is at the forefront is not the justification of war but rather its promotion among the participants.3 From a historical point of view this goes back to the disputes of the founding period and of the period of early expansion; relations dating to before state formation are apparent in the neglect of war preparations on the part of both political and religious authorities, as well as in the conduct of such wars; only in the post-war phase, when the spoils were distributed, is political leadership sought for the purposes of the authority for mediation.4 Comparable phenomena can also usually be established for religiously motivated military actions in the European late Middle Ages and in the era of the Reformation: it is above all the small groups, that lack stabilized political structures, which use ‘messianic’ concepts of war that are steeped in religion.5 Even in the handful of cases cited here, the religious or power-political interests are quite varied. The concept of ‘Holy War’ contributes nothing towards explaining it. Carsten Colpe has traced its history as a concept and has shown that the term’s more widespread use in the twentieth century has its roots in the wars of liberation of the start of the nineteenth century; Ernst Moritz Arndt in particular is supposed to have been of importance to its spread within German-speaking areas.6 ‘Holy’ indicates no specific religious meaning here, it belongs far more in the field of semantics, where the nation was considered sublime and able to demand an unquestioning obligation. The semantics of the sacrifice made by the war dead, which by no means reflects ancient tradition, also belongs to this field.7 Thus it is possible to apprehend a HoffmannCurtius religious construction of politics itself, which in the context of modern Europe is nonetheless detached from institutionalized religion

2

Parker 1984; Birley 1988: 85–106; Wolgast 1990: 180–214. For the crusades: Erdmann 1935 (repr. 1955). 3 On this and the following: Noth 1995: 277–95, esp. 284f. 4 5 Noth 1995: 380. Kretschmar 1995: 297–316, esp. 313. 6 Colpe 1994: 14–22. 7 Cancik-Lindemaier 1987: 100–4, esp. 102ff. In the architectonic sphere, HoffmannCurtius (1990: 142–71) refers to parallel developments.

104

Religion and War

in particular ways. Here, any analytical separation of the political and religious system must bear this in mind. The extreme case, the identification of religion with the state waging war, does not offer a viable model for an analysis applied more widely over history. This is not connected with the potential special brutalization of ‘Holy War’: here the priority should be to explain the inadequate professionalization of the fighters mobilized in a ‘Holy War’, not the greater level of propaganda in this kind of warfare. The problem is shaped rather by social conditions, about which no generalizations are possible, and which lie behind those categories of conflict that are considered to be ‘Holy Wars’. However, a little reflection on the factors that determine war and on the areas in which religion may possibly have been influential, must precede the essential process of historical clarification.

WAR, CAUSES, INFLUENCES, AND MOTIVES Ferguson, R. Brian Investigations into war and its causes are distinguished by a growing complexity. Mono-causal theories that reduced the function of war to the controlling of the population or to the regulation of territorial size between societies at different levels of development, or which associated the frequency of war with the individual level of aggression within a society, have been replaced by theories that consider material as well as ideological factors and regional ‘constellations’ (i.e. alliances).8 In his synthesis, R. Brian Ferguson has tried to integrate the various explanatory components into a three-part model of infrastructural, structural, and super-structural factors9—a useful model for staking out areas of potential religious functions. Ferguson claims that in every society ‘infrastructural’ reasons—the environment, ecology, geography—are responsible for many of the characteristic paths to warfare.10 However he concedes that structural factors—kinship systems, economic structures, the political constitution— shape the structure of processes to resolve conflicts leading to war. Political factors such as the education of central authorities or the capacity to forge alliances, become more significant with rising social differentiation. The same applies to the ‘super-structural’ factors: the rules of warfare and the 8 Ferguson (1984: 1–81) offers an overview of materialistic theories. Bibliographies: Divale 19732; Ferguson and Farragher 1988. 9 10 Ferguson 1990: 26–55. Ferguson 1990: 34.

Religion and War

105

esteem in which bravery was held, in respect of the trophies won or institutions of vengeance that are in place, influence the behaviour in warfare and the speed of escalation; at the same time there is the oral material with which members of society are themselves able to express their own motives. However, in themselves these things are not normally sufficient for war to break out.11 Ferguson has formulated his model with reference to pre-state societies, but it can also encompass still further developments. With the ongoing process of differentiation within society the levels of conflict shift. No longer does the ecological position of the society as a whole, its share of land suitable for cultivation, of fishing grounds or of pastures, constitute the main field of conflict. It is rather in the determining of positions within the starkly differentiated society that the potential for conflict now lies. Correspondingly, where war is concerned, the expectations of victory among different members of society diverge. Hence the decision to go to war is taken according to the interests of the ruling elite or of the military leadership: force, not consensus, constitutes the basic foundation of mobilization in state-based societies—even if that is also concealed by ideological manipulation that vested interests are in a position to present as being in the interests of the whole society.12 On the whole the radical shift in the relative significance of individual factors, that is, the historical dimension of the anthropology of war, has been neglected, unfortunately.13 Furthermore, attempts to explain the characteristics of modern warfare with reference to elements of pre-state warfare predominate. This applies in particular to the analysis of the religious elements of war, which in many cases are associated with the individual motivations of those fighting. Under the circumstances of state organization, however, the cohesion and thereby the effectiveness of armies is not manufactured through collective sacrifices: in the Roman army such acts were perhaps performed only by specialists, who are at all times surrounded by officers; the soldiers as a whole might perhaps never so much as get to see the scene. This is by no means an unworthy audience. This kind of choreography in the act of sacrifice would only underline the military’s steep hierarchy, which is expressed by the extreme inequality of authority, salary (share of booty), and living conditions. Thus religion served to legitimate the military apparatus and so—once more—its efficiency, though it would not be the means of transmission to convey it to individual fighters:

11 13

12 Ferguson 1990: 46f. Ferguson 1990: 51. See Stietencron and Rüpke (eds.) 1995: 14.

106

Religion and War

harsh discipline, the effective lack of alternative choices as to how to act, ultimately the group-dynamic processes of small groups such as communal living in tents14—those are the deciding factors in the individual’s ‘motivation’. It also becomes clear from this that the ‘super-structural’ factors (ideology) do not offer anything additional, but are most strongly connected with the structural ones, with how society is actually put together. But at the same time a perception of those structural factors that is too static is to be guarded against: internal tensions can break out—or be suspended—precisely under the strain in the political system that war entails. So it is precisely by analysing war that we can lay bare the precarious cohesion of state formation,15 if the investigation is carried out systematically and takes into account not only military movements and combat action, but also the points of transition, the beginnings of wars and the end of them. The investigation of religious rituals as an intra-societal code of communication in particular contributes to expanding upon the perspective of the conflict resolution process between both of the parties to a conflict16 in order to further one of the parties’ internal vision of society. Hence the refusal to limit ourselves to the internal relationships of the military apparatus, as referred to in the example cited above. Alongside the analysis of military specialists there comes that of the political elite, on the one hand, and on the other that of the wider population. Here too it is worth maintaining a historical perspective, as a brief look at the problem of mobilization shows.17 Its significance is intrinsically bound up with the technical organization of warfare, with rates of participation.18 So it plays only a limited role in wars that are conducted by mercenary armies (as in the Hellenic east of the ancient Mediterranean or in early modern Europe), as well as in colonial, nuclear, or mechanized warfare, but it assumed tremendous significance for the levée en masse or in forms of total war in the nineteenth and twentieth centuries, with their requirement that the economic sector as a whole

14

For the moderns, Moskos 1969/70: 13–23; for the ancients, MacMullen 1984: 440–56. Mann 1986. 16 For this approach, see Ember and Ember 1981: 407–14; Koch 1979: 1–16. 17 See Wallace 1968: 173–82; van der Linden and Mergner 1991: 9–23; Ulrich 1995: 399–419. The particular situation of August 1914 with its far-reaching compromising of German intellectuals and its broad literary after-effects has, in German research, prevented the analysis of antagonistic or passive elements in society. One must point to the lack of preparedness to go to war in 1939, not as an excuse, but as the background to almost six years of outright defensive war effort for a criminal regime. 18 For the concept of a military participation rate and the social typology developed from it, see Andreski 19682. 15

Religion and War

107

move to a war footing. The analysis of religion and war must adequately reflect this complexity. Analysing religion in terms of cultural scholarship as proposed also calls for the social and historical context to be included as comprehensively as possible. Hence it has to be limited to a few of the most incisive case studies. The following investigation concentrates on two advanced cultures that belong to the pre-modern era but which are organized as states, first and foremost Rome, and then Central Mexico. Both cultures are marked by their intensive military activities and a religious apparatus with a high profile; the analysis can show how rituals that are thoroughly archaic in appearance function in societies that already have a high level of differentiation: social complexity is not a phenomenon that is restricted to the modern period.

THE RELIGIOUS ELEMENTS OF WAR: ROME

Preparation As in many ancient poleis, conscription was a defining element of ‘civic’ identity in Rome: only Roman citizens (cives Romani) might serve in Roman legions. Yet the military apparatus was not restricted to this group. Even leaving aside the baggage train, which was largely drawn from non-Roman elements, a substantial part in the conduct of the war was played by non-Roman elements among the troops—in the time of the Republic, these were the Italian confederates, in the Imperial period, auxiliary units who were only rewarded with citizenship when they were demobilized. Both Romans and non-Romans were subject to military discipline through an unconditional oath of service sworn to the field commander in person (later to the Emperor as the imaginary commander); like an (early) modern military parade, the lustratio exercitus, performed when they deployed, was supposed to have demonstrated the collective force’s hierarchical structure and the unity of its members, which was focused on the field commander. If one sets aside the muster, for individual warriors the war only began at this point. Nor as a Roman citizen, even if he lived in Rome, was he practically involved in the decision on the war. The decision to go to war at the popular assembly—which was not always sought19—was an act of

19

Rüpke 1990: 123f.

108

Religion and War

overt approval, a ‘ritual of consensus’,20 that failed on just a single occasion (200 , and only at the first attempt). In the mobilization phase the issues of controversy were the integration of Romans whose fortune was not big enough to qualify, or of non-Romans, as well as the mistrust of the already established ranks at the incorporation of individual ‘veterans’ who were drafted in once more: according to what our sources tell us the themes were questions of status, prestige, and not the decision to go to war as such—which does not necessarily mean that a unanimous consensus prevailed over this war. The religious level was only touched upon when it came to the question of who was to lead the army. A range of divinatory acts (the auspices) established the legitimacy of the departing field commander (only one or two ‘barbarian’ opponents had female commanders.) The political character of these acts is to be seen mainly in the fact that the official, who arranged the auspices for himself, could control the result to a very great extent; he did what he wanted to do and could make the people believe that he had divine blessing for it. This sort of an explanation misunderstands the possibilities and effects that are entailed in how the ritual is constructed. The rituals of the auspices practised in the later republic are distinguished by an extensive lack of transparency: all that was made public was what the serving official claimed to be the result of the act that was carried out the previous night (in cases where the flights of birds were observed) or by a specialist out of view of the public (the tripudium, the observation of the eating patterns of hens kept for that purpose). This was not like the opening the temple doors (by means of hydraulic devices) as though by hands of the gods during a public sacrifice,21 it was not at all like the white hind (released from a hiding place) appearing before the start of battle as if the gods had sent it, which was meant to eliminate the prospect of uncertainty in the face of battle.22 Instead the function of such rituals was supposed to make its mark among the political elite over which the magistrate presided as the holder of the priestly office: the lack of demonstrable ‘objectivity’ of the result of the act of divination made possible not only the claim by the official who was about to assume command, but also the obnuntiatio, the objection of the peers who perhaps sat in the college of augurs. Acting in this way the future field commander thus opened up his actions to final opportunities for objections to be raised in a complicated 20

So Flaig 1995a: 7–127, esp. 81–91. Architectural drawings for facilities of this kind are known from ancient architects. 22 Sertorius is said to have used these stratagems: Aulus Gellius, Noctes Atticae 15. 22. 1–9. 21

Religion and War

109

process of creating consensus among the ranks of leaders. In the absence of objections among the gods, only the establishment of a clear consensus among the leading ranks offered the legitimation to which the protagonist could also refer back in the case of failure or defeat.23 It is typically only the loser who has died, who is no longer represented in ongoing dialogue, whose religious legitimation is ultimately doubted.24 The defeated general was not put to death, unlike in medieval China; but the outlook for his future career was very poor.25 The system so far described demonstrates the basic pattern of oligarchic systems. While the leading group, in Republican Rome the nobility, had to maintain a high degree of authority vis-à-vis the people, the inner group was supposed to be marked out by its homogeneity. Republican Rome solved the latter problem through having rival areas of competence (collegiality, the right of veto) in the ‘executive’, amongst the few officials, and a fastidiously observed order of rank among the senators as a whole.26 The taking of the auspices belonged to the checks and balances as well as—alongside the principle of short term-limits (annuity)—a spatial criterion, in which maximum freedom to crack down in the external and military spheres contrasted with the strict limitations on the magistrate’s use of force within the city. In accordance with its political significance, this line of separation was strongly marked out in religious terms; this took the form of elaborate boundary-crossing rituals, including the changing of clothes.27 The opening of the Gates of Janus in its original form28 belonged to this system of corresponding rituals, as did the triumphal procession.

The Military Expedition (Kriegzug) A complete inventory of the rituals carried out in war or indeed all of the cults that were followed in military institutions is not being given here.29 In the normal course of events nothing spectacular was to be found among them; for the most part it was a matter of rituals of routine that the commander probably performed among his staff or even delegated to Cf. Livy 22. 42. 7–9, on the disaster at Cannae in 216 . Cf. Cicero, De divinatione 1. 37, on C. Flaminius, who died in the battle of Trasimene in 217 . 25 China: Franke 1970; Rome: Rosenstein 1990; reservations about the radical thesis of Rosenstein: Harris 1990: 288–94 (responses: 294–8). 26 On this, Flaig 1995b: 115–48. 27 For a comprehensive account, see Rüpke 1990. 28 29 Rüpke 1994: 55–78, esp. 68f. But see the tables in Rüpke 1990: at 243. 23 24

110

Religion and War

specialists. Nevertheless regulations defined these kinds of actions: above all regulations relating to space, when a river was crossed or upon reaching a camp, on occasions the regulation of time as well, when a popular feast day was to be celebrated among the troops, as far as the military situation allowed—then everyone was certainly involved, with eating and drinking in every tent [contubernium]. For such celebrations a centre for the cult such as there were in Rome, through the temple with which it was associated, or particular sites of cults that also existed outside of the city, were lacking. Of course, the commander was moved to seek publicity when he made a vow during battle. It was not just that the goddess (or god) called upon was supposed to hear him or that trusting in the divine help they expected the fighters who surrounded him would make greater efforts— both possessed great meaning, since the vow (votum) was among the most elementary and widespread of the religious forms in the ancient Mediterranean.30 From a political perspective it is more important that the vow, whose purpose was more usually the establishment of a temple (and of a cult) or the staging of games (ludi votivi), was used to make decisions: decisions about the use to which spoils were put, about celebrating the hoped-for victory—and the victor!—in the collective memory (even about a temple or some such) as well as decisions about how the Roman Pantheon was constituted, through the incorporation of a new god. The largest part of the temples in Rome went back to battle vows of this kind, decisions about the city’s architecture and topography that the rest of the nobility could change only to a limited extent. It is worth our while to linger for a moment on this point. It illuminates in exemplary fashion the religious character of Roman politics just as it does the political character of Roman religion. First of all, it is to be clarified that the temple building initiatives described (to single out the most common case) was limited to magistrates in their capacity as commanders in the course of their normally short period of office. Only within this short term of service and time-frame did a spectrum of alternative courses of action exist that was truly wide, to which the consent of fellow citizens was given or was to be induced. However, this consent was still necessary: if need be actions that remained part of the collective framework could be overturned by the ruling classes, even in retrospect, through a complicated system of legal and religious veto mechanisms and ways of forcing an annulment. When required even

30

van Straten 1981: 65–151; Linders and Nordquist (eds.) 1987.

Religion and War

111

in the very late Republic sufficient errors in the religious formalities could be found in the dedication of Cicero’s house, carried out during its owner’s exile, which allowed for the re-establishment of the status quo ante, that is, private ownership.31 This consensus was formalized in the ruling of 304, which elevated a decision made by the senate on the legal requirement to dedicate into the final act of assignment to the god. This ‘public-legal’ dedication did not constitute a necessary condition for the introduction of the cult, but it did oblige the state to assume the ensuing expenses of the new cult.32 There were a few more specific general conditions that imposed limits on how much imagination could be used when building temples. It was thus a question of the sacral-legal precepts, for instance that in a temple with several gods each had to be given his or her own cella and space for his or her cult statue, or that the augurs’ lines of observation were not to be obscured by the construction.33 Otherwise the choice of location itself seems to some extent to have been sited at the pleasure of the founder, provided that other sacral or private rights were not thereby damaged.34 In temple construction the wide distribution of power that aimed for consensus is shown in the involvement of various groups of people.35 For the preparation of the templum, the space in which the physical temple was raised, there was a need for augurs, who performed a liberatio and an effatio, an absolution from any other possible claims on religious grounds and a clear definition of the space in terms of sacred law. At the 31

See Cicero’s speeches De haruspicum responso and De domo sua. Livy 9. 46. 7. Tradition interpreted the law as the result of Flavius’ dedication of the Concordia temple; the possibility of a straightforward tribunician agreement was quite unreal. From the second quarter of the second century  onwards, it was the participation of the People that established a plebiscite (Cicero, De domo sua 127). The problem of workspace for business—with the emphasis on control by the senate—continued to apply down to the time of the Principate. For analysis, see Bardon 1955: 166–82. 33 For example, Livy 27. 25. 89 (after that, see the collection of exempla in Valerius Maximus 1. 1. 8); Festus 466. 36–468. 3 L. 34 On the implications in land-law of the consecratio, see Wissowa 19122: 477f. On the following details and evidence, see Wissowa 19122: 471–7, 527f. That the choice of place and of the resulting symbolism could result in very individual points of view, is shown by the example of the vow of M’. Acilius Glabrio as fulfilled by his son: Pietati aedem consecratam ab Acilio aiunt eo loco quo quondam mulier habitaverit, quae patrem suum inclusum carcere mammis suis clam aluerit: ob hoc factum, impunitas ei concessa est. (Acilius, they say, consecrated a temple to Pietas in the very place where a woman had once lived, who when her father was imprisoned fed him secretly from her breasts; because of this action, she was granted impunity.) 35 Jordan (1872, 229–40, esp. 233) already makes this point. However, I would not want to follow his distinction between consecratio—pontifices and dedicatio—magistrates. This separation, motivated by systematic legalism, is not supported by the sources. See Varro, De lingua Latina 6. 54 and Festus 476. 14–18 L. 32

112

Religion and War

dedication the participation of a pontifex was required, to review the formulas and to perform the demonstrative act [gestus] of touching the fasciae that marked the room in relation to the door jambs that had already been set up. The dedicator himself, if he no longer discharged any office, required a fresh official mandate through being elected as a temporary magistrate by the duoviri aedibus dedicandis. If at this time the founder was no longer alive, this office was filled, with a certain preference for the son. Any participation by other decision-makers was lacking—but only insofar as there was no need for it to be financed by any resources that the senate managed.

The Return The impression given up to now, that the political relevance of the rites of war was to be found more on the home front than on the field of battle, is confirmed by the re-entry rites, the return of the commander, and eventually that of the army. The central and most complex ritual was the triumph, which concluded a victorious campaign: an entry of the decorated army into the city, that presented a long procession, the captives and plunder, scenes of war in living images or portraits; at the centre was the triumphator, in a carriage, decorated in the manner of Jupiter Optimus Maximus, the god of the city. Insofar as it was a part of an old system of war rites involving the presentation of plunder and redeeming the vow made on departing before Jupiter Optimus Maximus, who resided on the Capitol, it is beyond dispute that the triumphal procession had to take place (inasmuch as the course of the war provided a reason to give thanks to the god). All the same it was in historical times (and until it became the monopoly of the Emperor) the subject of negotiations, long and delicate talks that dealt with the arrangement, financing, and scheduling of the celebrations. The commander’s claim was conceded in principle, but in individual cases it was not so straightforward. It would certainly not be amiss to draw attention to the antagonistic character of the situation: a victorious commander before the city, with armed troops, who as a soldier and with soldiers wished to enter into the ‘demilitarized zone’ of the urbs Romana in a ritual context, and to garner as much prestige as possible among the people of the city through the solemn presentation of his spoils. The commander’s peers, who recognized and respected the claim—perhaps they themselves would soon be in the same situation— nonetheless wanted to prevent the worst case, an unassailable rise in the candidate’s prestige.

Religion and War

113

Here religion obviously fulfils a function other than the rituals discussed so far. It now serves—comparable in this sense to the construction of the temple in fulfilment of a vow—as a medium for exchanging external military success for internal, urban prestige within society. This prestige is not amorphous but rather it is cast in fixed forms, as far as possible, allowing for similarities among the nobility. What matters is not the question of whether five thousand or ten thousand enemies were killed on the field of battle or whether five or ten enemy tribes were subjugated: all deeds that defy practicable confirmation anyway. What count are the functional values (in the sense of mathematics) of this success, as they are presented after negotiations with the senate, the representative organ of the nobility: is it enough for a triumph, or only for an entry on foot (ovatio)—or does the commander feel so cheated that he organizes a triumphal process on the Alban hill at his own expense? What will count later, with the commander’s funeral procession, and what his family will have to reckon with for all time, is whether or not the deceased might be dressed in the triumphal robes. It is clear that such strategies of differentiation could only have had limited success. What was just as important for contemporaries was the duration of the triumphal procession—one, two, even three days—the length of the procession, the number and eminence of the prisoners, the expense of the meal given to the people at the end; it was the scale of the closing ludi triumphales, the staging, length, and quality of the performances. Where the ritual of the pompa funebris, the procession of death masks in the mourning train for the most recently deceased member of the family, made the actual clothing the most important thing, in the Augustan period this was transformed into the proliferation of marks of distinction in another sphere. In these texts, it had become virtually a mania to place emphasis on the office-holder’s integrity and on his extra qualifications—such as being the youngest possible candidate chosen by the Emperor.

RITUALS WITH OR WITHOUT SPECTATORS: FOR WHOM OR WHAT IS ‘STATE RELIGION’? If one were to ask: what is the political function of religion? then the communicative aspect of religious behaviours comes to the fore. Hence there arises the question of the actors, the ‘senders’ and the ‘receivers’. In Rome there are, first and foremost, the leading political powers that we encounter among religious protagonists, magistrates who may or may

114

Religion and War

not have military responsibilities. The members of the priesthoods that are relevant here—they belong to the same class, they sit almost continuously in the senate—have the right to advise, they have a veto; they are hardly protagonists in the rites. Two groups of rituals can be clearly distinguished: public and nonpublic. The political function of the latter resides above all in the political-military leadership’s legitimation as well as in the affirmation of their own role—one thinks here of the commander changing clothes, swapping the citizen’s toga for military garb (sagum) when he crosses the city’s boundaries. In these rituals, communication occurs among the members of the leadership elite, between magistrates, senators, and the priests who come from this group. The public rituals are in the main situated after the conflict. First and foremost was the triumph. It is individualizing, emphasizing the commander, or more precisely the general acting under his own auspices on the day of his victory.36 The name of the person who on the Capitol ultimately places the laurel crown onto the lap of the statue of Jupiter is recorded in a list; in the Fasti Barberini, the oldest preserved inscribed list of the names of magistrates, not of consuls, but rather a list of triumphators (perhaps) on the Capitol. Considering the significance that was also attached to the triumph in other respects, it is conceivable that documenting office-holders in Rome on a written basis started here, with the triumphators, not with the consuls. The prestige acquired in the Triumph can also be transmitted, even if it individualizes. This is borne out in particular by those people who marched in the train without having participated in the war: the assembled senate that greeted the commander at the gates of the city and then joins the ranks, and the triumphator’s children, who could ride along on his carriage. These latter win a share of the father’s prestige, which at a later point in time—at the burial of the triumphator—will benefit them with the appearance once more of the triumphal robes. The wish to impart a lasting quality to the success demonstrated (militarily) and won (in civic terms) in the triumphal procession, defines the further arrangements made for the triumph, which transcend the fixed ritual framework: architectural monuments, which invoke ritual elements of the procession in the construction of the arches,37 preserve the memory of the victory and of the procession; the imperial monopoly 36

In the Republican construction of the army led by the two consuls, the supreme command changed every day—only one of the consuls held unlimited capacity for action, based on the auspicia. 37 On the triumphal arch, see Germann 1988: 69–81.

Religion and War

115

on the triumph, which its frequent repetition with the same people makes possible, is an attempt to express the virtus demonstrated by the uncertainty of the individual victory as an enduring characteristic of the person of the Emperor with reference to the imperial family, the domus Augusta.38 As for the actual form taken by the train, the closing games: there are practically no limits on the opportunities to drive up the expenditure. At the same time it must be stressed that a fixed structure of rituals can still be made out: the triumph remains the triumph, even when its length is doubled or stretched to three days it remains an experience restricted to a narrowly confined span of time; a later appearance in the triumphal robes, as is reported of Marius, must have unleashed profound irritation.39 Thus, in the collective memory the construction of a temple has a crucial significance. The building of votive sanctuaries does not indicate that Roman victories were almost always close calls, thus making it necessary for the commander to make a public vow as a last resort. Their construction on a regular basis is much rather a sign that in a vow to dedicate a building only a subordinate role is played the aspect of ‘crisis ritual’. The Roman general must consider how his victory will be visualized when he is still on the battlefield. It was not possible to provide day-to-day updates in newspapers with maps showing front lines (which in the pre-modern era would in any case only ever be inadequate for the documentation of conflict). First, the placards, pictures, and masses of prisoners, also the money distributed in the triumphal procession, as well as—on a permanent basis—the scale and decoration of the temple raised with the plunder, brought the extent and the meaning of the victory before the eyes of the inhabitants who had remained behind (and perhaps even of the soldiers themselves). Ritual communication between the Roman population, in particular the inhabitants of the city of Rome itself, and the elite is concentrated on the post-war phase, or more precisely: the post-victory phase—defeats were not brought up within this framework, large-scale rites for those killed in action, as in Athens, did not exist despite the system of conscription and the personal bonds thus created between fighters, that is, those who were killed, and the civilian population. The extraordinarily high level of readiness to perform military service in the mid- to late Republican period, as the documented success of mobilization attests, On the ‘theology of Victory’ and the Victor: ‘See Hölscher 1967; Heim 1992. In 105  Marius probably appeared in the senate wearing his triumphal robe: Livy, periochae 67; Plutarch, Marius 12. 2–5; see also the honorific inscription for Marius CIL 12, 195. 18. 16–19. 38 39

116

Religion and War

was not only created by the appeal for volunteers or maintained through the course of a war: it was required for each new war, the ‘state of the war’ is not specifically denoted during the conflict. To put it bluntly: in Rome war was a spatial, not a temporal phenomenon—in any case so long as the city of Rome itself was not threatened (which, on the rare occasions when it was, led to the emergency situation called the tumultus). Only the demonstration of the profitability of an (of course) victorious campaign took on the characteristics of an appeal to volunteer. A comparison with the no less elaborate post-war rituals of the Aztecs makes the Roman system clearer. These Mexican post-war rituals are shaped by the human-sacrifice-complex. Here the personal relationship between the Aztec warrior, who has taken his opponent prisoner, and this prisoner, plays a central role. The removal of the act of killing to the capital Tenochtitlan makes possible an extreme individualization that finds its ultimate expression in the victor and his family carrying along and putting on the skin that has been removed. Taken as a whole, here it is the status of the warrior that receives approbation and a sense of security through the ritual’s publicity and the presence of the trophies right up into the outskirts of the city.40 But also part of the political function of this ritual was the priesthood, which was central to society yet marginal in war, and which had a central role in the act of killing. It is they who determine and put their stamp on the choreography, they organize a context for the killing, that leaves only a narrowly confined scope for the two fighting actions, they alone perform the simple, cosmologically significant act of killing. Here the complex of human sacrifice appears as a part of a highly differentiated society, in which social mobility is channelled by success in war. It goes without saying that the choreography of the ritual underlines the leader’s central position, which emerged in the wake of the rapid expansion, consolidated in the fourteenth century, that was based on the structures of the city-state and marked by the aristocracy.41 There was a third goal that the political communication of the capital ritual aimed for. Members of the political elite of conquered or enemy states can also view the Aztec victory rites from a safe place, indeed one that was partially hidden. Perhaps this element took on great significance in the system of Aztec hegemony, which gradually acted as a stabilizing force in densely settled central Mexico with its multiplicity of political units characterized by unstable ‘dynasties’ and changing alliances.

40 41

For a detailed analysis of just this aspect: Carrasco 1995: 1–26. On this, Prem 1989: 28–30.

Religion and War

117

That the last of these elements played a role in Rome as well certainly cannot be ruled out; relevant testimony for this exists in the report by the Jewish general, prisoner of war and historian Flavius Josephus on the triumph that Vespasian and Titus celebrated over the Jewish people (Bellum Iudaicum, Book 7). But this aspect did not play an important role (as it seemingly did in Tenochtitlan). The political function was directed towards the leading strata, their inner workings, and their relationship to the people. In the rituals that accompanied the troop parade: it was a matter of legitimating the commander, who succeeded not demonstrably, but e contrario as a result of the lack of opposition the campaign commander’s control of the troop parade, which was especially problematic, and which the college, envoys of the senate [legati] or political allies [amici] had to assure; and finally keeping control of the transformation of military success and the resulting economic success into domestic political power: this was achieved through a regulation in the code, which made possible this transformation and translation.

WHAT SORT OF POLITICS? WHAT SORT OF RELIGION? The mechanisms of social behaviour have been expanded upon above to such an extent that it must in fact be stressed that Roman warfare was not a ritualized, automated system. Here, as a functional performance war depended—as in other societies—on relationships in the wider geographical region; every analysis that concentrates on the inner workings of society in societies with a high culture (as is the case with these very societies) must take this into account. Success in war was only to be predicted in conditional terms, there was no access to what would be the reaction of opponents towards border populations. Considering the enormous expense of war preparations, the prospects of success had to be assessed. One could also earn one’s spurs in other fields—though certainly not laurels like those that the triumphator wore as a crown in his hair.42 Rationality was as entrenched in the individual striving after prestige, which had as its goal the long-term improvement of the family’s position, as it was in the senate (the complete assembly of competing parties), which was concerned with the situation at the start of the following year being better rather than worse. Beyond this from the start of the first 42

For discussion of the alternatives, Rüpke 1995: 216–19.

118

Religion and War

century  up to the start of the war of confederates every activity associated with warfare took place within a system of alliances that was certainly dominated by Rome, but which was nevertheless a complex one. Within this system war represented—so far as it was not forced from the outside—an opportunity (for Rome) notably to improve its own position, on its own initiative, within the system of ‘international’ relationships. In parallel to this opportunity for society as a whole there was also the possibility of greater differentiation, improving one’s own position within the system of intra-societal relationships. In this sense war put pressure on the leading strata in particular ways, for which religion made provision, with its repertoire of behavioural forms with rules—‘rules of the game’—for communication between the political groups involved. This religion, which in the last sentence appeared as the subject, was not a unitary system. First it is to be established: taken from the point of view of the majority of people and most places at which cults were regularly kept, that was no daily religious performance associated with the majority of the rites involved. It was a matter of a single, coherent system in which its elements stood in relation to one another in many respects: the election auspices to the auspices of departure, the vows of departure with the redemption of these vows through the laying down of the laurel wreath in the same temple at the high point of the triumphal procession. ‘The religion of war’ did not characterize these connections adequately; certainly it was not a matter of ‘religious war’. A form of religion conceived specially for the needs of soldiers (above all the Roman legionaries) is not to be detected as a single system: a straightforward cult of the military standards among the masses did not exist; worship of the genius of military units in the imperial period followed the general lines of the cultic self-identification among social groups. Where there was a heightened atmosphere of crisis—beyond the individual acts of commanders seeking legitimation through divination—specialists, individual providers such as soothsayers, met the demand.43 This of course corresponds to procedures back home: for situations that went beyond this and that were not covered by institutional custom within the fundamental social structures, specialists who offered their religious services at wellknown spots became involved—traditional agents also having their place within the essentially locally structured religions of antiquity.44 43 So, during the over-long siege of the Spanish Numantia in the second century . See Appian, Iberica, 85; Valerius Maximus 2. 7. 1. 44 John Helgeland attempted a comparison of Roman and American ‘War-religion’ (Helgeland 1989: 22–44), but the embedding of the chosen phenomena in an overall analysis of the social role of religion remains inadequate. (1985/6)

Religion and War

119

On the other hand the familiar terms state religion or state cult offer a starting point for describing the religious elements associated with war and the systems that support it, not in the sense of a religious monopoly protected by the state or of a state monopoly of religion, but rather in that of interrelated forms of religious behaviour being narrowed down to the field of political action. Such a concept does not adequately bring to light the question of religion’s political dimension or the religious dimension of politics, since on that basis religion and politics can no longer be identified as being independent of one another. This calls for clarification. The system of rituals that is being analysed here in terms of a state cult does not stand in isolation—that is a first, important point. The base lines of the theological dimension correspond to local religion as a whole both in relation to the language of forms and that of sacrifice. A degree of coordination with local religion occurs in terms of both space and time, that is, in terms of the calendar: some rituals are associated with specified locations, the Capitol, the Regia, while certain days set the framework for particular political activities; after all, the temples founded as a result of the system described above constitute an important element of the sacral topography of Rome.45 On the other hand, it is precisely these conjunctures that are the source of political disputes: architectural encroachments, on a massive scale since the later Republic and above all on the Emperor’s part, change such systems of relationships. Intensive conflict, which is often just lumped together under the heading of political impasse,46 is precisely what shows that these intersections are structure-bearing elements of the system: it is the relationship to local religious structures as a whole that ensures that the terms established by state religion are not arbitrary. That is essential to their acceptance and ability Cato function. Protagonists are marked out by the same kind of connection. For the most part, magistrates become associated with the aforementioned ‘state priesthoods’ at certain points, which in fact grow more frequent as time goes on. They are, in a certain sense, religious specialists who are responsible for a self-propagating public cult (the circumstances in which it was possible to delegate are unclear). Above all they have the function of overseeing with regard to a wide scope of affairs related to the cult, among which are included political affairs once again and above all rituals deemed to be ‘state cults’.

45 46

For the conception of sacred topography, see LiberoCancik 1985/6: 250–65. See de Libero 1992.

120

Religion and War

To the extent that the state cult opens up to a wider public it acquires an integrative function for society. The games that took place in the context of the triumph or that were offered as ludi votivi after a victory offer extensive evidence for this: these are not entertainment events studded with mere religious survivals, but rituals, key components of public religion and political communication simultaneously.47 The triumphal procession celebrating the city’s own victory also has its integrative moments, which to begin with, in the order of its train and the separation of active participants and passive spectators, seem rather to make distinctions more marked: it is the triumphator himself, dressed as a god, who becomes the subject of mocking verses: ‘Caesar conquered the Gallic lands, Nicomedes [conquered] Caesar: look, Caesar, who subjugated Gaul, is now triumphant; Nicomedes, who defeated Caesar, is not [triumphant]’, sang the soldiers in Caesar’s Gallic triumph, alluding to his alleged homoerotic relationships.48 Whoever has most elevated himself out of society is at the same time brought to heel most rigorously by the society of the mockers—back to their norms, to their standards—and so he is claimed back as a member of this society. This broadly attributed capacity of religion for integration into society in the Durkheimian tradition does not characterize the whole of the system associated with the state cult. As has become clear, the primary capacities for integration reside with the leading strata: they help to constitute this as a coherent group that is thus capable of leading the people as a whole.49 That by no means excludes other religious activities by members of the group: in Rome religious pluralism is attested just as much here as among the dominant strata.

CONCLUSION The task was to define the political functions of religion: the use of the political system in the area of war as a particular situation was further 47

48 See Clavel-Lévêque 1984; Flaig 1992. Suetonius, Divus Julius 49. 4. That confirms a supposition of Gunther Kehrer’s: ‘. . . that the State-cult served the integration of the ruling authority with the administrative officials, since the occupants of positions of power and the administrators of power always have first to establish themselves anew as a Group. They form an artificial Group within society, which, whether it was originally defined as ethnic or local, also becomes functional very quickly; for that reason, it always finds itself in the situation of having a precarious sense of belonging’. See ‘Integration’ in Cancik, H. Gladigow B, and Kohl (eds.), K-H.1993: Handbuch religionswissenschaftlicher Grundbegriffe 3.Stuttgart, 269-78, esp. 276. 49

Religion and War

121

elaborated. Negative concepts were first of all excluded, as offering overly simple definitions of the relationship, reflected, for example, in the concept of ‘Holy War’. The political functions of religious activities were positive, in the context of war as elaborated for the societies of Republican Rome and of the Aztecs. But the important result to be grasped from this exploring history more deeply is not the existence of such functions—which is not surprising—but rather their specificity with regard to the public. The political itself is not a vague and diffuse aspect of society, it is made concrete much more in institutions and processes that can be thoroughly restricted to small sections of society. But the same is true for religion, which presents itself in its differentiated forms [Gestalt] in institutional multiplicity. The investigation of war leads to the claim that the relationships between religion and politics in their various parts are presented quite differently. What is relevant for Rome is above all an area that can be termed [angesprochen] ‘State religion’ or ‘State Cult’. Without the political and religious system being congruent—a ‘flat’ [flächendeckend] analysis of the relationship of both subsystems has not in fact been attempted in the above—both systems come together in the field of State Cult relatively strongly: religion, though only ‘State Religion’, makes important contributions to the constitution and the institutional arrangement of the political system. The existence and expansion of such a ‘state cult’ have nevertheless to be verified for every society, and their power to be defined precisely. Here, this has been done at least partially: war, as the continuance of politics by other means, is only clearly present as religion in the city of Rome—which is not to be estimated lightly [gering] as [a form of] cultural power. I thank the editor and Frau Dürkop of Potsdam for their [critical] remarks and suggestions for corrections.

A FT E R W O R D Research on the Roman Republic has changed the framework into which this article was written in the early 1990s. The chapter points already to the fact that ancient participants shared the academic observer’s systematizing view only to a limited degree. That certainly needs to be given even more stress today. Against the background of the discussion about the formation of statehood in ancient Italian cities, the agency of those who must be addressed as heads of clans deep into the middle Republic (and in some regards even beyond) has come into even sharper relief. Many of the rituals discussed here deal with the precariousness and the self-fulfilling strength of the military positions involved. Whereas the stress

122

Religion and War

on spatiality and the creation of specific social spaces was already one of the central results of the research presented here, the ‘material turn’ now invites to give more prominence to the material results and the history of many of the objects involved and cherished in the rituals described. A city filled by the arches and temples of triumphant generals and houses decorated with spoils (whether individually taken or allotted as part of the booty) would have shaped memories and situational identities of those involved as victors, beneficiaries, or captives of war, would have nicely illustrated narratives produced on bodily scars and losses as much as on gender roles. Above all, it will have shaped the definition of being Roman as a matter of successful participation in past or future wars, beyond differences in linguistic, regional, or any types of ethnic identities. (For further discussion, see: Rüpke 2018; 2019.)

B I B L I OG R A P H Y Andreski, S. (1968). Military Organization and Society. London. Bardon, H. (1955). ‘La naissance d’un temple’. REL 33: 166–82. Birley, R. (1988). ‘The Thirty Years’ War as Germany’s religious war’, in K. Repgen (ed.), Krieg und Politik 1618–1648: Europäische Probleme und Perspectiven. Schriften des Historischen Kollegs, Kolloquien 8. Munich, 85–106. Cancik, H. (1985). ‘Rome as a sacred Landscape: Varro and the End of Republican Religion in Rome’. Visible Religion 4/5: (1985/6) 250–65. Cancik, H., Gladigow, B. and Kohl, K.-H. (eds) (1993). Handbook of the basic principles of religious study. 3. Stuttgart. (Handbuch religionswissenschaftlicher Grundbegriffe 3.Stuttgart, 269–78, Esp. 276). Cancik-Lindemaier, H. (1987). ‘: Zur imaginären Antike der Jahrhundertwende in Deutschland und Österreich’. Altsprachlicher Unterricht 30.3: 90–104. Carrasco, D. (1995). ‘Give me some skin: the charisma of the Aztec warrior’. History of Religion 35: 1–26. Clavel-Lévêque, M. (1984). Cadastres et espace rural. Paris. Colpe, C. (1994). Der ‘Heilige Krieg’: Benennung und Wirklichkeit, Begründung und Widerstreit. Bodenheim. De Libero, L. (1992). Obstruktion: Politische Praktiken im Senat und in der Volksversammlung der ausgehenden römischen Republik (70–49 B.C.). Hermes Einzelschriften 59. Stuttgart. Divale, W. T. (19732). Warfare in Primitive Societies: A Bibliography. Santa Barbara. Ember, C. R. and M. Ember (1981). Anthropology. Englewood Cliffs, NJ. Erdmann, C. (1935). Die Entstehung des Kreuzzugsgedankens. Forschungen zur Kirchen- und Geistgesgeschichte 6. Stuttgart. Ferguson, R. B. (1984). ‘Studying war’, in R. B. Ferguson (ed.), Warfare, Culture and Environment. Orlando, 1–81.

Religion and War

123

Ferguson, R. B. (1990). ‘Explaining war’, in J. Haas (ed.), The Anthropology of War. Cambridge, 26–55. Ferguson, R. B. and L. E. Farragher (1988). The Anthropology of War: A Bibliography. Occasional Papers of the Harry Frank Guggenheim Foundation 1. New York. Flaig, E. (1992). Der Kaiser herausfordern: die Usurpation in Römischen Reich: Frankfurt. New York. Flaig, E. (1995a). ‘Entscheidung und Konsens: Zu den Feldern der politischen Kommunikation zwischen Aristokratie und Plebs’, in M. Jehne (ed.), Demokratie in Rom? Die Rolle des Volkes in der Politik der römischen Republik. Historia Einzelschriften 96. Stuttgart, 77–127. Flaig, E. (1995b). ‘Die Pompa Funebris: Adlige Konkurrenz und annalistische Erinnerung in der römischen Republik’, in O. G. Oexle (ed.), Memoria als Kultur. Göttingen, 115–48. Franke, H. (1970). Zum militärstrafrecht im chinesischen Mittelalter. Bayer. Akad. der Wiss., phil-hist Kl. 1970,5. Munich. Germann, A. (1988). ‘Zum römischen “Triumphbogen”: Der Bogen von Orange’. Altsprachlicher Unterricht 31.2: 69–81. Harris, W. V. (1990). ‘On defining the political culture of the Roman Republic. Some comments on Rosenstein, Williamson and North’. Classical Philology 85: 288–93. Helgeland J. (1989). Christians and the Military. London. Heim, F. (1992). La théologie de la victoire de Constantin à Théodose. Théologie historique 89. Paris. Hoffman-Curtius, K. (1990). ‘Altäre des Vaterlandes: Zur Genese eines neues Typus von Kriegerdenkmälern in der Weimarer Republik’. Visible Religion 7: 142–71. Hölscher, T. (1967). Victoria Romana: Archäologische Untersuchungen zur Geschichte und Wesensart der römischen Siegesgöttin von den Anfängen bis zum Ende des 3 Jhs. n. Chr. Mainz. Jordan, H. (1872). ‘De sacris quibusdam in hemerologio fratrum Arvalium commemoratis’. Ephemeris Epigraphica I: 229–40. Koch, K.-F. (1979). ‘Access to justice: an anthropological perspective’, in K.-F. Koch, Access to Justice 4: The Anthropological Perspective. Patterns of Conflict Management: Essays in the Ethnography of Law. Aalphen an den Rijn-Milan, 1–16. Kretschmar, G. (1995). ‘Der Heilige Krieg in christlicher Sicht’, in Stietencron and Rüpke, 1995, 297–316. Linden, M. van der and G. Mergner (1991). ‘Kriegsbegeisterung und mentale Kriegsvorbereitung’, in M. van der Linden and G. Mergner (eds.), Kriegsbegeisterung und mentale Kriegsvorbereitung: Interdisziplinäre Studien. Beiträge zur politischen Wissenschaft 61. Berlin, 9–23. Linders, T. and G. Nordquist, eds. (1987). Gifts to the Gods: Proceedings of the Uppsala Symposium 1985. Boreas 15. Uppsala. MacMullen, R. (1984). ‘The legion as a society’. Historia 33: 440–56.

124

Religion and War

Mann, M. (1986). The Sources of Social Power. Vol. 1: A History of Power from the Beginning to A.D. 1760. Cambridge. Moskos, C. C. (1969/70). ‘Why men fight: American combat soldiers in Vietnam’. Transaction 7: 13–23. Nipperdey, T. (1993). Deutsche Geschichte 1866–1918. Vol. 2: Machtstaat vor Demokratie. Munich. Noth, A. (1995). ‘Der a priori legitime Krieg im Islam; Hauptaspekte des islamischen Rechts zum Thema “Krieg und Frieden” ’, in Stietencron and Rüpke, 1995, 277–95. Parker, G. (1984). The Thirty Years’ Wars. London. Prem, H. J. (1989). Geschichte Alt-Amerikas. Oldenbourg Grundriss der Geschichte 23. Munich. Rosenstein, N. (1990). Imperatores Victi: Military Defeat and Aristocratic Competition in the Middle and Late Republic. Berkeley. Rüpke, J. (1990). Domi militiae: Die religiose Konstruktion des Krieges in Rom. Stuttgart. Rüpke, J. (1994). ‘Die religiose Konstruktion des Krieges in Rom’, in H. Wissmann (ed.), Krieg und Religion. Würzburg, 55–78. Rüpke, J. (1995). ‘Wege zum Töten, Wege zum Ruhm: Krieg in der römischen Republik’, in Stietencron and Rüpke, 1995, 216–19. Rüpke, J. (2018). Pantheon: A New History of Roman Religion. Princeton. Rüpke, J. (2019). Peace and War at Rome: The Religious Construction of Warfare. With an Afterword by Federico Santangelo. Transl. by David M. B. Richardson. Stuttgart. Stietencron, H. von and J. Rüpke, eds. (1995). Töten im Krieg. Veröffentlichungen des Instituts für Historische Anthropologie 6. Freiburg. Ulrich, B. (1995). ‘Kampfmotivationen und Mobilisierungsstrategien: der Beispiel Erster Weltkrieg’, in Stietencron and Rüpke 1995, 399–419. Van Straten, F. T. (1981). ‘Gifts for the gods’, in H. S. Versnel (ed.), Faith, Hope and Worship: Aspects of Religious Mentality in the Ancient World. Studies in Greek and Roman Religion 2. Leiden, 65–151. Wallace, A. F. C. (1968). ‘Psychological preparations for war’, in M. Fried, M. Harris, and R. Murphy (eds.), War: The Anthropology of Armed Combat and Aggression. Garden City, NY, 173–82. Wissowa, G. (19122). Religion und Kultus der Römer. Munich. Wolgast, E. (1990). ‘Konfessionalisiering und Religionskriege’, in J. Assmann and D. Harth (eds.), Kultur und Konflict. Frankfurt a.M., 180–214.

5 Not the One nor the Many; A Pragmatic Approach to Religious Behaviour in a Polytheistic Society The Example of Rome Andreas Bendlin

Iuppiter omnipotens, regum rerumque deumque progenitor genetrixque, deum deus, unus et omnes.1 We assert the existence not only of as many gods as all Hellenic peoples affirm, but also of many more.2

Could there be any ancient city better suited to the analysis of a complex polytheistic system than Rome? Rome is the ancient metropolis (with perhaps around a million inhabitants by the Augustan period) and is the political, social, as well as cultural centre of an empire reaching far beyond the Mediterranean. In the Imperial-period literature—both in the so-called ‘pagan’ and Christian written sources—the city of Rome is considered the ‘essence’ of the oikoumenē: she is the ‘city common to all’ and the acropolis for the inhabitants of all of the empire, her influence extending over the entire globe.3 Here, in the only true ‘world city’ of antiquity,4 all divinities, 1

Q. Valerius (Soranus) fgt. 2 FPL3 = 4 FPL2 = 2 Courtney (1993, 66f.); the text is preserved in Varro, Curio de cultu deorum fgt. II Cardauns: ‘Almighty Iuppiter, father and mother of kings, things and gods, god of gods, at the same time one and all’ (Cardauns 1960: 16–19, esp. 16). 2 Philodemus of Gadara, De pietate col. 362f. For text and translation, see Obbink 2002: 209f. 3 Astu koinon: Aelius Aristeides, Oratio 26. 61; epitomē: Athenaios 1. 20b–c. Cf. Ovid, Fasti 2. 684. 4 Morley 1996: 13–32; Edwards and Woolf 2003. Andreas Bendlin, Not the One nor the Many; A Pragmatic Approach to Religious Behaviour in a Polytheistic Society: The Example of Rome In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0006

126

Not the One nor the Many

regardless of their origin and number, come together to receive worship, as if in a ‘templum for the whole world’. Just as various ethnic groups and individual civitates elsewhere worshipped their own local gods, so in Rome honour was bestowed on every one of them jointly and without distinction.5 This is the external perspective of Rome as seen from the provinces. Authors from the city of Rome, by contrast, tend to paint a less harmonious picture: from the very beginning, Rome is a city of migrants; both humans and deities are drawn towards the city ever since its foundation. Latin literature soon thematizes fears of ‘religious infiltration’ through uncontrolled immigration of innumerable foreign gods, cults, and ritual practices, and through the proliferation of so-called superstitiones.6 For the purpose of historical analysis, how can one instil order on this many-faceted divine world in the city of Rome, when it not only confounds the modern observer, but in all likelihood appeared complex and confusing already to ancient visitors to the city? How can one make visible the ritual actions and beliefs of the social actors, which underlie the countless dedications and monuments erected to the gods in the city? How can their religious behaviour be rendered comprehensible?

THE LIMITS OF DESCRIPTION: POLYTHEISM IN ROME Until quite recently historians of Roman religion have routinely attempted to instil order on the bewilderingly diverse material with the aid of emic categories (that is, categories derived from the discourse of the social actors being investigated). Thus Georg Wissowa, in his still authoritative handbook on ‘religion and cult of the Romans’ (1902; 19122), divided Rome’s deities into autochthonous ‘di indigetes’, Italic and Greek ‘di novensides’ (‘newly created deities’), and the ‘foreign’ gods of the sacra peregrina, which included the so-called ‘oriental gods’. Wissowa’s distinction between, on the one hand, autochthonous-Roman divinities and, on the other hand, deities newly settled in Rome reflects a fundamental assumption in the nineteenth- and early twentieth-century study of religion, namely a belief in the supposed existence of an autochthonous, original Roman religion not yet corrupted by later religious forms and 5

Templum totius mundi: Ammianus 17. 4. 13; cf. Scriptores Historiae Augustae, Aurelian 20. 5. Veneration of all gods: for example, Aelius Aristeides, Oratio 26. 105; Tertullian, Apology 24. 9f.; Minucius Felix 6; Arnobius 6. 7; Prudentius, Contra Symmachum 1. 89; Themistius 13. 177d–178b. See Clarke 1974: 190–2; Fowden 1993: 45–52; Bendlin 1997: 38, 50. 6 For example, Cicero, Pro Flacco 67–9; Livy 39. 16; Tacitus, Annals 13. 32. 2; 15. 14. 4; Dio Cassius 52. 36. On this theme, see Noy 2000: 31–51; Isaac 2004: 66–7 and passim; Scheid 2005: 225–7.

Not the One nor the Many

127

ideas. Wissowa’s use of Latin terminology suggests emic categories, which the scholar merely puts to analytical use for the purpose of demarcating a core of Roman polytheism against later amplifications. However, the terminology Wissowa had reified, scholars soon realized, possesses a very different, less comprehensive meaning in Roman antiquity than that which it eventually acquired in his classificatory grid. Wissowa’s religious-historical construct, which endeavours to distinguish archaic Roman from foreign gods, is not so much owed to a unified emic perspective as it is the prisoner of the a priori nationalpolitical and cultural assumptions of the nineteenth century.7 Studies of more recent vintage cannot divorce themselves completely from dichotomous structural principles either, it seems. The distinction between deities that were peculiar to Rome and foreign divinities or that between Roman religio and non-Roman superstitio proffer another emic perspective—one that is represented in Latin literature, as previously mentioned. But the attempt to make this emic perspective a modern structuring principle for the writing the history of Roman religion8 misapprehends that it reflects one particular, elite-specific discourse (albeit an important one) about religion, rather than representing all social actors in Rome. Several recent works have demonstrated in detail that in Rome even the so-called ‘oriental cults’, against which above all the charge of superstitio was levelled in the literary texts, are products and elements of Roman religion: their deities too must be considered integral to the city of Rome’s pantheon.9 No consensus existed in ancient Rome on what was central and ‘Roman’ in the city’s religion and what was marginal and ‘foreign’. Determining Roman religious identities is more difficult, their constitution more complex, than may appear to be the case at first sight. For Rome is in constant flux: both economically and culturally the city exists (and survives) only by means of countless foreigners, migrants, and the unfree. Political, social, economic, and religious differences shape life. Old and new gods, traditional and unfamiliar religious ideas, exist side by side.10 In these circumstances the modern observer is barely able to recognize what is ‘Roman’ and

7 Wissowa 1912, 91–317. On the indefensibility of Wissowa’s construct, see the doxography in Bendlin 2000a: 1029; for more detailed discussion, see Bendlin 2006. 8 The distinction between indigenous and foreign divinities, between Roman religio and un-Roman superstitio, in short, between the Roman and the foreign still provides the structure for the—otherwise very useful—works of Beard et al. 1998, esp. 211–44; Turcan 1998, passim; and Scheid 2001, esp. 155–75 and passim. 9 Belayche 2000; Gordon 2000; Versluys 2002: esp. 385–443; Vout 2003. 10 Migration: see Jongman 2003. Religious plurality and integration: see Bendlin 2000d; Noy 2000: 183–6, 205–84; Scheid 2005: 227–36.

128

Not the One nor the Many

what is ‘foreign’. If they nevertheless prioritize the elite discourse about ‘the Roman’ in Roman religion, establishing it as a guiding criterion by which to judge the religious system as a whole, and if they make it the reified frame of their investigation, they merely affirm the claims of that particular discourse. They thus blur the boundaries between the objectlevel and the level of scholarly analysis. The alternative, etic, approach (that is, the systematic use of terminology and categories developed through scholarly discussion with no semantic equivalent in the ancient data) is also problematic, albeit for different reasons. An amorphous ancient ‘world of gods’—another modern concept whose anachronistic separation of two worlds, one human and the other divine, operating independently of one another has to my knowledge no exact correspondence in the ancient sources—is subordinated to modern classification and organization: our tools of analysis— pantheon, polytheism, indeed religion itself—are etic categories. Greek Πάνθειον and Πάνθεον as well as Latin Pantheon may designate the sacred place or shrine at which all the gods (the Θεοὶ πάντες and di deaeque omnes, respectively) receive worship, but only in the modern study of religion does the concept attain its wide-ranging, systematizing significance. Originally the Greek adjective πολύθεος merely describes that which concerns a plurality of deities, ranging from altars to the assembly of the gods (there is no direct equivalent in Latin), but it is only with Philo of Alexandria that Greek πολύθεος, πολυθεότης, and πολυθεΐα become terms of Jewish and later of Christian polemics against the idolatry and atheism of the ‘pagans’. Idololatria fulfils this function in Latin Christian apologetics at an early stage. The negative approximation of these three terms—πολυθεότης (or πολυθεΐα), idol(ol)atry, and atheism—continues in the early modern theological discourses: the mid-sixteenth-century German equivalent is called ‘Götzendienst’. In his De la démonomanie des sorciers (1580) Jean Bodin first translates πολυθεότης and ἀθεότης from Proclus as ‘polythéisme’ and ‘athéisme’, while Johann Fischart’s German translation of Bodin’s text, which is published in 1591, first deploys the terms ‘Polytheismus’ and ‘Atheismus’. For some time thereafter these words remain more or less synonymous with that of idolatry. It is only from the seventeenth century onward that scholars of religion begin to operationalize the term polytheism, together with the neologism ‘monotheism’, as a cornerstone of philosophical and historical debates on the origin of religion.11 11

On the history of these terms and their meaning, see Bendlin 2000b: 265f.; Hülsewiesche and Lorenz 1989: 1087f.; Bendlin 2001a: 80f., each with further literature. The famous ‘Pantheon’ in Rome is unlikely to have been a temple for ‘all the gods’: Godfrey and Hemsoll 1986. See also below, nn. 42–5.

Not the One nor the Many

129

Notwithstanding their ancient literary employment in polemics and critiques of religion, the words pantheon and polytheism have found their way, their significance largely transformed, into the analytical language used by historians of religion. The advantage of this etic approach is its applicability across narrow disciplinary boundaries: my tentative definition, from the perspective of history of religion, of a pantheon as ‘an ensemble of deities, their number restricted in religious practice, who are conceived of as possessing agency and are worshipped in a particular geographic and social space’, although it is formulated with a view to Graeco-Roman antiquity (it lacks a Greek or Roman equivalent, though), differs only in nuances from comparable definitions in the neighbouring disciplines of Ancient Near Eastern Studies or Egyptology.12 But it is not enough to accumulate lists, as it were, of individual deities in a given geographic and social space. If that ‘world of gods’ is not purely a matter of chance, but rather the result of the ideas and world views of the social actors operating in these spaces, there follows from this condition a question of genuine interest: how do the political, social, economic, and cultural structures in which the social actors live—in a social group, city, state, or empire—correlate to the concrete form and internal structurations of their gods? It was above all the ‘Paris School’ which demonstrated that any pantheon found in the historical record has a meaningful structure13 and thus can be investigated as a system of meanings which reflect, on the level of gods interacting with one another, basic patterns of societal organization and human social relations.14 From the very beginning, the search for internal structures and hierarchies anchored the Paris School’s heuristic endeavour, indebted as it was to the French structuralist tradition. Its exploration of the Greek pantheon was placed in the context of the study of Greek myth. On the basis of the textual evidence one may thus reconstruct familiar sociomorphic principles of structuration and classification that emphasize a fundamental hierarchy and division of labour among the gods in the pantheon: for instance, the classification of kinship relations, expressed in genealogies and patriarchally organized families of gods; the hierarchical classification of the gods as more or less powerful, greater or smaller, higher or lower; or their internal

12

Definition: see Bendlin 2000b: 265. See further Schwemer 2006, with more literature. ‘Meaningful’ (‘sinnvoll’) in the tradition of Max Weber, in so far as social action usually holds significance and meaning and religious world views reflect ideas of the social actors. 14 See, for example, Vernant 1974; Detienne 1997; Bruit Zaidman and Schmitt Pantel 1992: 176–214, esp. 183–6. See also Bremmer 1994: 11–26, esp. 14f.; Parker 2005: both authors favour a contextualized and moderately structuralist approach. 13

130

Not the One nor the Many

differentiation according to task, function, and place.15 Georges Dumézil even endeavoured to reconstruct a tri-functional hierarchy of the archaic Roman pantheon out of Indo-European roots whereby the ‘original’ pantheon, dividing the gods into the specialized functional realms of ‘political authority’, ‘war’, and ‘(agricultural) productivity’, supposedly mirrored the social structure of early Rome with its division of labour. Dumézil’s model is again indebted to the structuralist method, but it can be falsified by confrontation with the historical evidence.16 The problem with a strict application of this structuralist method is in part that it privileges the ancient literary texts, which in and of themselves undertake to establish significance inasmuch as they already proffer theological speculation and construe structures. These texts themselves merely present a selection (albeit an important one) from among the social actors’ discourses on religion; but since they only represent particular ideas and interests, it is not permissible to reify them in order that they may undergird etic accounts of a pantheon. This would blur, once again, the boundaries between the object level and the level of analysis. In the history of Greek and Roman religion in particular, the accretion of non-literary data (which frequently do not belong among the local elite discourses) results in a growing differentiation of our knowledge. For example, on the basis of a small number of antiquarian texts scholars for a long time assumed that women were as a matter of principle excluded from the Roman worship of the god Hercules; yet a survey of the literary and epigraphic data demonstrates that the restrictions, rather than being universally valid, applied only to the god’s rituals at the Ara Maxima.17 On the level of religious practice, studies of this kind reveal, a pantheon is no coherently structured reference system in which a divinity can be shown to have a clear—and above all just one single—function or task. In this way one can also falsify Dumézil’s tri-functional structural model.18 Like most purely structuralist models, his is incapable of sufficiently describing either the gods’ polyvalence in specific action situations or their historical development. The question of correlation between a pantheon’s structure and the political, social and cultural conditions under which the social actors operate remains to this day unresolved. As long ago as 1960 Angelo Brelich remarked, rather en passant, that a developed polytheistic religion presupposes at least the cultural level of an agricultural society; and it has been variously conjectured, above all with reference to ancient 15 16 18

On this topic, see Gladigow 1983: 295–301; Gladigow 1998: 324f.; Bendlin 2000b: 266. 17 In detail, see Dumézil 1970: 148–75. Schultz 2000. Cornell 1995: 75–9; Beard et al. 1998, vol. 1, 14–16. See below, n. 37.

Not the One nor the Many

131

Near Eastern civilizations, that a pronounced hierarchical classification of the realm of the gods, culminating in the establishment of a king of the gods, must go hand in hand with the existence of central rule, or at least of consolidated power relationships.19 This conclusion is not mandatory: panthea characterized by complex structures, a hierarchy, and division of labour are also known in so-called tribal religions.20 Any further claims about the relationship between hierarchically organized panthea structured by division of labour and the society that sustains them, therefore, can only ever be made with reference to concrete historical situations and contexts. But any claims about a generic correlation between political, social, or cultural structures and the divine world’s hierarchisation and internal structuring remain problematic.21 The polyvalence of that divine world and historical developments can be properly understood only if we adopt a pragmatic perspective. It must map the ensembles of gods, sanctuaries, cults, myths, and theologies in their concrete social and geographical space; it must also investigate how the gods’ availability varies according to place, time, and context. Karl Otfried Müller was the first, in 1825, to establish a scholarly horizon for this enquiry: all deities must be studied within their respective ritual context, which is conditioned by locality or regionality, and in their distinctive peculiarity, rather than primarily by means of their transregional myths. They must be interpreted as a part of ‘local religion’.22 Müller formulates a—by now accepted—scholarly perspective, namely that the religious systems of Graeco-Roman antiquity are to be contextualized as ‘local religions’, and that their pantheon must be reconstructed against the background of what conditioned the local communities worshiping these gods. In its concrete historical actualization ancient religion is first and foremost local and regional religion.23 From there some have developed the paradigm of ‘city-religion’, ‘polis-religion’, or ‘civic religion’: the enquiry into the correlation 19

Brelich 1960. In general, see Gladigow 1981a; Gladigow 1993: 37–9. Schulz 1993. 21 See Gladigow 1998: 325f. The contributions in Kratz and Spieckermann 2006 proffer numerous case studies on how the correlation between divine hierarchies and the political order may play out. In the history of Greek and Roman religion, too, one can observe how the formation of local monotheisms or pantheistic constructs occasionally goes hand in hand with very specific political or dynastic ambitions: Bendlin 2000c: 271. The problem cannot be resolved when reductive typologies are applied, as is the case with Mora 1994; Mora 2000. 22 Müller 1825: 239–45. See Schlesier 1994: 31f. 23 MacMullen 1981: 1–7; Mellor 1992. Against this background, the following methodological problem arises: Can religion that is originally situated in a local or regional context function within trans-regional territorial structures, for instance as religion for an empire? See Rüpke 1997. 20

132

Not the One nor the Many

between the pantheon and the social actors’ self-positioning within a region or locality is here reduced to the relationship of religion to the political community. The enquiry about the relationship between a pantheon and politics thus is a narrowing of the original concern with the pantheon’s relationship to society. In this perspective ancient religions, as city-religions, are tied to the respective political system and—in contrast to Christianity, for example—they cannot be understood without reference to it; religious practice does not exist beyond the parameters of religion in the political-public sphere. The concepts of state religion and Imperial religion are likewise indebted to our constructing ancient religions as entities situated in the political-public sphere inasmuch as the pantheon is analysed—as a city’s pantheon, a state’s pantheon, or an empire’s pantheon—with recourse to ancient discourse fields determined by the political-public sphere. But caution is advised when using these etic concepts: in employing them the student of religion does not simply describe the religious structures they identify on the ground—for neither these concepts nor the collective of gods they comprise constitute emic terminology and categories. Rather, it is the student of religion in the first place who gives a name and structure to groups and constellations of gods that may be observed only temporarily, are relevant only to circumscribed political and social groups, or may be pertinent only in specific contexts; it is he or she who accords them a quasi-historical legitimation they lacked in reality. The danger of this etic approach is that it reifies as a system structures that may have been fluid, and which it should properly describe in their fluidity. The critics’ main charge against the model of ‘civic religion’ or ‘state religion’ is therefore that it marginalizes religious practices and ideas that lie outside the political discourse field, and that it turns an aspect which was merely one part of ancient religious practice into the sole standard of evaluation.24

PANTHEON OR PANTHEA? LOCAL RELIGION IN THE CITY OF ROME As for pragmatic implications, the dilemma arising from these considerations calls for a different form of religious-historical description of the

24 For critiques, see Stevenson 1996: 1–10; Bendlin 1997: 47f.; Woolf 1997; Tatum 1999; Bendlin 2000e; Bendlin 2001b.

Not the One nor the Many

133

sources. I begin with a text of Latin technical writing: toward the end of the first century  Vitruvius, in the work on architecture dedicated to Augustus, discusses the selection of sites suitable for temples in a newly founded Roman city. The highest location befits the sanctuary of the Capitoline triad (Jupiter, Juno, and Minerva) as the deities upon whose protection the well-being of the citizenry is particularly dependent. Mercury, like Isis and Serapis, is to receive a temple in the market or the storage yard at the harbour. The sanctuaries to Apollo and Liber Pater must be established at the theatre, that of Hercules at the circus, unless an amphitheatre or gymnasium is available (the Greek Heracles is often associated with the ephebia and the gymnasium). The ‘other gods’ (ceteri di) are allocated their place according to the way in which sacrifices are made to them. However, to Ceres a shrine must be erected in a protected site outside the city walls; the same precaution applies to Venus (a temple at the harbour), Mars (a sanctuary ad campum), and Vulcan. In the case of these last three divinities, Vitruvius invokes the ritual directions of the disciplina Etrusca (that is, Etruscan ritual doctrine): a temple to the goddess of love within the city walls could overstimulate the veneria libido (lust) of adolescent youths or of mothers of families; one protected against the threat of fire by removing Vulcan’s power from the city; localizing the shrine of Mars outside the city walls discourages armed conflict between citizens and protects the city against external enemies.25 The placeholder reference to ‘ceteri di’ illustrates that Vitruvius’ motivation cannot have been to proffer an exhaustive list of a pantheon. His selection constitutes a summary cross-section: it is remarkable not only with respect to the deities included but also to who remains excluded. The sketching out of the pantheon’s gods could continue indefinitely—the ceteri di point to that possibility. Vitruvius’ selection does indeed name a core of ‘great’ divinities worshipped in a Roman city, but it ignores any local or regional particularities. His attribution of sites to deities operates on the basis of a simple system of correlation between the gods and their primary function, which at times betrays Greek and Mediterranean ideas: Mercury stands for trade, Apollo and Liber Pater for the theatre, Venus represents carnal love, Vulcan symbolizes fire, Mars war. The incorporation of the Etruscan haruspices’ ritual recommendations in a work of technical writing demonstrates that its author’s selection operates on the basis of Graeco-Roman typologies and theological interpretations.

25

Vitruvius 1. 7. 1–2.

134

Not the One nor the Many

Comparing the Vitruvian list with panthea attested elsewhere, either in the material record or at least epigraphically, in Rome or Roman civitates in the Latin West of the Roman Empire, one notices conspicuous differences: neither Rome nor any of the coloniae or municipia, and certainly none of the urban settlements with a status other than that of colonia or municipium, correspond, either fully or even in large part, to the topographic and ritual recommendations Vitruvius or the Haruspices make. If in these cases there exists at least some overlap, even agreement concerning particulars, it is barely present in the Greek East.26 In the Republican period only cult for Jupiter Optimus Maximus, Hercules, Vulcan, and probably Liber Pater is securely attested for the Roman colonia in Ostia supposedly founded by the king Ancus Marcius, along with another three gods who do not appear in Vitruvius’ catalogue: Castor and Pollux as well as Silvanus. The statute (lex) promulgated with the establishment of the Roman colonia of Urso in Baetica founded by C. Iulius Caesar specifically designates rituals and games just for the Capitoline triad as well as for Venus Genetrix, the divine progenitor of the colony’s founder. Her temple stood in the centre of the forum that Caesar had constructed in Rome; the goddess apparently has such strong associations with Caesar that following his death in 44  she is documented in the epigraphic data for only a restricted period of time and only to a limited extent, for example in Urso. A placeholder—‘gods and goddesses’ (dei deaeque)—are mentioned without further specification, once more marking the relative open-endedness of the pantheon to be established in the colony; its expansion was left to the local ordo decurionum. Both the small number of ritual prescriptions coming from the centre and the autonomy of the Roman colonists regarding their responsibility in religious matters stand out. Roman gods may nevertheless have been the first choice for the city fathers, in particular during the early years of existence of a newly founded colony in alien territory. In the process of acculturation within the provincial context the Roman colonies quickly developed their own religious identity, however. This is also reflected in the constitution of their panthea. In Aquilea in Gallia Cisalpina, a colonia since 181 , from the Vitruvian list cult for Apollo, Hercules and Minerva (without Jupiter and Juno), and possibly also for Serapis, can be substantiated in the material record or epigraphically. Bona Mens, Diovis, Fortuna, the Lares compitales, and Atamen (an otherwise unknown, perhaps local god), together with Timavus (an indigenous river god), Belanus, Borea the god of wind,

26

See Tosi 1984; Bullo 1992.

Not the One nor the Many

135

who was worshipped together with Apollo, and a deity by the name of Attis Papas complement the local pantheon. Unfortunately, no unequivocal identification of any further ritual activity from the Republican period, which could provide an even more diverse picture of the colony’s local pantheon, is possible.27 The Capitoline triad, which Vitruvius mentions in first place, is attested epigraphically for Urso, but a corresponding Capitolium must be a matter of conjecture since no material remains have been discovered so far. Similarly, Capitolia can indeed be attested for numerous coloniae founded in the Republican period, but certainly not for all of them, while still other coloniae only had Capitolia constructed some considerable time after their foundation. In Cosa, a Latin colonia from 273  onwards, a Capitolium providing cult for the Capitoline triad replaces an earlier temple to Jupiter around 160 . In the old Roman colony of Ostia, the building preceding the Capitolium constructed in Hadrian’s reign can only be dated back to the time of Augustus. In Paestum, a Latin colony also founded in 273 , construction of a Capitolium was started at the beginning of the second century, though it was never completed; a Temple of Peace took its place around 100 . In contrast to these instances there exist in the second and third centuries  the examples of several North African municipia as well as of settlements that pursued municipal status but did not yet possess it: the construction of a Capitolium and the establishment of cult for the Capitoline triad— that is, the use of a status symbol in anticipation of a political status that previously only went to coloniae—in these communities casts a revealing light on their heightened political and social ambitions within a provincial context.28 Turning our view to the city of Rome’s pantheon in the Republican period, does it exhibit comparable diversification? Or is it possible to identify at least a Roman core of the city’s pantheon? For this question the sacred topography of Rome may be put aside for the moment. For beyond providing a collection of the data, what would be the value of a more or less complete inventory of the gods worshipped in Rome at a particular point in time, of their shrines and altars (although such an inventory would result in an astonishing number of deities of the utmost diversity), so long as the religious expectations of the social actors and the city of Rome’s religious communication system remain fuzzy? Can 27

Ostia: Meiggs 1973: 337–51. Urso: Bendlin 1997: 48–50; see also Rives 1994. Aquileia/ Gallia Cisalpina: Fontana 1997. 28 See Meiggs 1973: 351–3, 380f.; Todd 1985; Rives 1995: 40–2 and passim; Beard et al. 1998: vol. 1, 333–6.

136

Not the One nor the Many

one infer from the social actors’ communication the possibilities in structuring the pantheon in the city of Rome? In the Roman Republic laws could be expanded by an additional passage that bound executive bodies by oath to comply with the respective law. Thus the implementation of the so-called Lex Latina tabulae Bantinae from the late second century  was guaranteed by an oath that incumbent and any future Roman magistrates had to swear ‘right before the Temple of Castor [and Pollux] in public view in the light of day facing the Forum [Romanum], within five days of the promulgation of the law, in the presence of a quaestor’; the same injunction applied to senators ‘within ten days, in the presence of a quaestor at the Aerarium in public view in the light of day.’ Roman magistrates generally had to swear to follow the existing laws within five days of their entry into office. A refusal would have meant the end of their political career: the oathclause could thus be used as a means of exerting political pressure in the event of controversial laws. All the more pertinent to our enquiry are the gods addressed in these oaths: they are Jupiter Optimus Maximus and the Di Penates.29 In the Late Republican Lex Ursonensis, referred to previously, the scriba tasked with the financial administration of the colonia also has to take an oath to Jupiter and the Di Penates. The same addressees are named in the oath contained in the Lex Irnitana, a statute enacted in Rome for the benefit of the Hispanic municipium of Irni in the time of Domitian. This legal text from the Imperial period, however, inserts additional oath-gods between Jupiter and the Di Penates: Divus Augustus, Divus Claudius, Divus Vespasianus Augustus, Divus Titus Augustus, and the Genius of the emperor Caesar Domitianus Augustus—that is, those deified emperors through whom the ruling dynasty of the Flavians in general and the emperor Domitian in particular legitimized themselves. The text demonstrates the way that historically motivated pantheon-formations adapted under the actual conditions of the late first century of the Imperial period.30 The incorporation of Jupiter as oath-god is not surprising. Cicero can make fun of those orators who think they are best prepared to appear at court if they are only able to recite the words ‘Iovem ego Optimum Maximum’ from any old speech. The elevated position of Jupiter Optimus Maximus as the god of oaths stands in correlation to his significance 29 Lex Latina tabulae Bantinae: Crawford 1996: vol. 1, no. 7, lines 14–18, 24. Further examples are Crawford 1996: vol. 1, no. 8, lines 20–3; no. 12, Delphi copy, Block C, lines 10–15; no. 23. 30 Lex Ursonensis: Crawford 1996: vol. 1, no. 25, cap. LXXXI, lines 17–19. Lex Irnitana: Cap. 26, IIIB, lines 40–3; text and commentary: Lamberti 1993. See also Bendlin 1997: 50f.

Not the One nor the Many

137

for the city’s political rituals. Jupiter as oath-god is paired with the Di Penates, or rather: with the Di Penates publici Populi Romani Quiritium (thus ILS 108.) As the protectors of the res publica the latter were worshipped on the Via Sacra, at the south end of the Forum Romanum.31 The choice of Jupiter and the Di Penates publici as the paradigmatic gods of protection and of oaths points to the specific political and administrative context of the documents in which they are invoked: as with the case of Vitruvius, here too a selection occurs—and hence a rejection of other candidates—on the basis of attribution of a primary function. However, the exclusive functional contextualization of the Di Penates in the public-political sphere—they are worshipped by the Populus Romanus collectively, that is, by the magistrates as its representatives—entails that these specific deities are not represented in other areas of public and private life. As regards the religious views of the urban population, these deities apparently were associated so strongly with the protection of the res publica Populi Romani that worshipping them appeared not to be an attractive proposition either in the daily lives of free citizens or to inhabitants without citizenship, the unfree and migrants. Moreover, the free Roman citizen and paterfamilias worshipped his own Di Penates in domestic ritual, while his familia of unfree dependents sacrificed to the Lares or to the Genius of the master of the household.32 The formation of the Roman pantheon, the presence of the gods in the cityscape and last but not least the personal preferences of the city’s inhabitants as regards their worshiping individual deities are conditioned by the economic, political, and social factors prevailing in Roman society. These various factors reflect the diversification of Rome the ‘world-city’, resulting not in a homogeneous religion in the city, but rather in a plurality of religious ideas and divine worlds competing with one another within the urban environment. Cicero’s political and forensic speeches document the way that the selection of gods depended on the situation, even in the context of political communication. Following his return from exile, he names before the Roman priestly college of the Pontifices—a wider audience is excluded—the gods once more central to the political self-identity of the ruling class in the Republic: the Capitoline triad with Jupiter Capitolinus (sic!), Juno Regina and Minerva, the Di Penates Publici and 31 Cicero, Divinatio in Caecilium 43. Di Penates: Linderski 2000. Cicero repeatedly exploits this public-political protective character of Iuppiter and the Di Penates publici: see, for instance, Cicero, De haruspicum responso 37; Pro Sestio 45; Pro Sulla 86. 1; In Verrem 2. 4. 17. 32 On this pattern of differentiation, see Fröhlich 1991: 30f., 40, 178f., 261.

138

Not the One nor the Many

Vesta. The list concluding Cicero’s final speech against Verres presents a contrast: alongside Jupiter Optimus Maximus, Juno Regina and Minerva, the orator names before the judges Latona, Apollo and Diana, Mercury, Hercules, the Mater magna, Castor and Pollux, Ceres, as well as ‘all the gods’ (omnes di), including those local gods whose rights C. Verres had violated in his capacity as Roman governor of Sicily—a list appropriate, given Cicero’s subject, to the affairs of Sicily.33 In both cases it is first and foremost the speech’s communicative context that dictates the selection: although the inclusion and exclusion of deities does not occur randomly, the data nevertheless demonstrate a great amount of variability—a variability conditioned more by context-dependent factors than by essentialist, unchanging categories. Cicero’s final speech against Verres is revealing in yet another respect. It illustrates in exemplary fashion a principle of religious communication in Rome—a communication that in numerous other texts also addresses the gods in their entirety, that is, as di immortales or di deaeque. While modern students of Roman religion have often tried to define in quasi-essentialist fashion the ‘Roman’ pantheon by a process of exclusion, ancient logic clearly worked inclusively: a salient principle not only of ancient Roman polytheism is the attempt to incorporate all the gods—both the collective gods worshipped in one’s own city and all others.34 Last but not least, the Imperial-period commentarii of the Fratres Arvales (‘Arval Brothers’), a Roman priesthood in the grove of the goddess Dea Dia, document the degree of variability that went into compiling lists of gods in a polytheistic system. In the year 183  the master of the college performs the sacrifice of Suovetaurilia (a bull, a ram, and a boar) before removing a fig tree that had grown on the roof of the goddess’s temple and undertaking the roof ’s repair—an expiatory ritual (piaculum) in anticipation of the intervention that is about to occur. Thereupon (in what follows, the little word ‘item’, ‘likewise’, is each time marking a logical break) he sacrifices different sacrificial animals (the differentiation revealing of the hierarchical internal structuring of those honoured) to the following divinities at the sanctuary of Dea Dia: Dea Dia, Janus Pater, Jupiter, Mars, the Juno of Dea Dia, sive deus sive dea (‘whether it be a god or a goddess’), the Divine Virgines, Fons, Flora,

33 Cicero, De domo 144–6; In Verrem 2. 5. 184–9. Heibges 1969 is still useful for the analysis of these and comparable passages. 34 On this principle, see Bendlin 1997: 53f. See Livy 42. 3. 9: ‘. . . tamquam non iidem ubique di immortales sint’ (‘. . . as if the immortal gods were not the same everywhere’, in the sense of ‘they must be honoured as gods everywhere, including outside their own catchment area’).

Not the One nor the Many

139

Vesta, and Vesta Mater. There follow (item) sacrifices to Adolenda Conmolenda Deferunda. Finally (item) sacrifices are performed to the sixteen Divi on the altar in front of the Caesareum, the shrine to the imperial cult. Once the slaves have cut down the tree, chopped it up, and burned it (that divinity’s name, Adolenda Conmolenda Deferunda, is related to these three activities) and have repaired the roof, the Suovetaurilia sacrifices are repeated. Then (item) there follow sacrifices to Dea Dia, Janus Pater, Jupiter, Mars, the Juno of Dea Dia, the god or goddess, the Divine Virgins, the Divine Assistants, the Lares, the Mother of the Lares, the god or the goddess ‘under whose protection the grove or site is placed’, Fons, Flora, Vesta, and Vesta mater. Thereupon (item) further sacrifices to Adolenda Conmolenda Deferunda and (item) to the sixteen Divi follow.35 The second of these two sequences of sacrifices already expands the list of divine addressees through the addition of ‘lesser’ deities. In the commentary of 224  the deity whose name is associated with the slaves’ activities in the grove is suddenly called Adolenda Coinquenda. A thunderbolt had hit the trees in the grove: these must now be dug up, cut down in size, burned, and replaced with new trees ( . . . earum[ . . . ] arborum eruendarum ferro ferendarum adolendarum commolendarum item aliarum restituendarum causa). Before the work begins, Suovetaurilia are again performed. Likewise (item) Dea Dia again receives sacrifice, and then (item), this time apparently in separation from the sacrifice to the goddess of the grove, another sacrifice occurs to Janus Pater, Jupiter, Mars, the god or the goddess, the Juno of Dea Dia, the Divine Virgins, the Divine Assistants, the Lares, the Mother of the Lares, Fons, Flora, Summanus Pater, Vesta Mater, and Vesta of the gods and goddesses. Finally (item) a sacrifice is offered to Adolenda Coinquenda and—this time there is no item—in front of the Caesareum to the Genius of the reigning emperor Severus Alexander Augustus ‘our Lord’, and only then (item) to the Divi. The sequence of sacrifices is repeated, beginning with Suovetaurilia, once work in the grove has been finished: ‘the rest as above’ (cetera quae supra), the epigraphic text says lapidarily. In the text from 240  Mars, the Juno of Dea Dia, Fons, Summanus Pater (whose name first appeared in the list of 224 ), Vesta of the gods and goddesses, and Adolenda Coinquenda are absent, while the internal structuring by means of item, which occurred in 183 and 224, is largely abandoned.36 The list of ritual addressees in the commentarii of the Arval 35 Commentarii fratrum Arvalium (hereafter, CFA) no. 94, col. I, line 20–col. II, line 14 (= Scheid 1998: 263f.). 36 224 : CFA no. 105b, lines 5–19 (Scheid); 240 : CFA no. 114, col. I (Scheid).

140

Not the One nor the Many

Brothers is subject to constant revision. Its fundamental principle of classification, rather than following strict logical rules, appears indebted to the attempt to include the widest range of divinities believed to display their agency: recipients of sacrifice such as the Genius of the emperor are supplemented since 224 , whereas both ‘major’ and ‘minor’ gods are removed or change their names down the years, as is the case with Adolenda Conmolenda Deferunda, the ‘functional divinity’ created in the context of the commentarii.37

PLURALITY AND UNITY The existence of a plurality of deities perceived as displaying agency and interacting with others perforce generates, it appears, claims on the part of social actors about the nature of these deities and their relationship with one another. The pantheon’s internal structure thus becomes a (theo-)logical problem, emerging as the object of endeavours to systematize or entertain philosophical speculation. In Rome, one encounters such processes in text genres and discursive fields as diverse as the commentarii of a Roman priesthood and an architect’s technical writing. In the discursive field of civic religion in cities and city-states, the systematization of rituals and internal structuring of the divine addressees of ritual in a local religion occur at an early stage, namely when local elites start to exert control over the urban settlement context; they go hand in hand with the emergence of religious specialists, the development of a religious infrastructure of cult, temples, and festivals, and the formulation of a ritual calendar. These processes are typical not just of the wider Mediterranean world, but of other pre-modern societies as well.38 Beginning with Homer and Hesiod, religious-philosophical—or, theological—speculation about the nature of the gods constitutes a complementary discursive field in the Greek and Roman Mediterranean. Systematizing the relationship among a plurality of gods and speculating about the many and the one soon leads to the philosophical critique of

37 For which, see Scheid 2003: 171, 179–87. Both here (Scheid 2003: 176–80) and in Scheid and Svenbro 1997: 297–302, Scheid chooses Dumézil’s trifunctional model in order to explain the classification adopted in the Arval Brothers’ lists of sacrifices, but he must concede that the Dumézilian model is not consistently applicable. 38 For more details, see Whitehouse 1992; Cornell 1995; for a comparativist perspective, see Trigger 2003: 472–521, 564–82.

Not the One nor the Many

141

the poets’ material conception of the divine. They also result in thought experiments about the one supreme god—who sometimes may be conceived of as being transcendent.39 The purpose of the quotations that preface this contribution by way of a motto is to illustrate that this intellectual confrontation between the many gods and the one god does not represent a development peculiar to the ancient Near East or Greece, but that it has a home in Roman polytheism too. Philodemus of Gadara defends the Epicurean version of the traditional idea of the divine against Stoic philosophizing in Italy in the first century . Q. Valerius of Sora—who is likely to be identical with the Roman tribune of the people executed in Sicily in 82  at the behest of the Roman Senate— composes an invocation of Jupiter in the form of a hymn that defends the same Stoic pantheism (albeit one built upon Orphic foundations) Philodemus is attacking, and which celebrates the highest god as a union of all logical opposites, transcending the traditional plurality of gods.40 But the one who is placed above all the other gods thus also stands at a remove from the realm of the gods and is no longer accessible. The philosophical hymn is a logical thought experiment; it does not directly reflect personal religiosity, and no immediate consequences for ritual practice arise from it.41 The title of this contribution is a variation on the title of a 1971 book by Erik Hornung, subsequently translated as Conceptions of God in Ancient Egypt: The One and the Many. In this book Hornung demonstrates conclusively that ‘monotheism does not arise within polytheism by way of a slow accumulation of “monotheistic tendencies”, but requires a complete transformation of thought patterns. Tendencies to classify the pantheon should not be equated with an inclination toward monotheism.’42 Since the seventeenth century scholarship on religion has often pursued a different conceptual model, which describes the historical development of religion as an evolution from polytheism (or 39

For a detailed discussion, see Versnel 2000; Bendlin 2005: 217–28. On Q. Valerius, see Suerbaum 2002. On the Orphic concept of a cosmic god, which appears as early as in the author of the so-called Derveni Papyrus, see Betegh 2004: 182–223; for its adaptation by the Stoics, see Courtney 1993: 67f. In the hymn, the addressee must receive a name: in hymns in general since Aeschylus (Agamemnon 160–83; fgt. 70 TrGF) and in Stoic hymns since Cleanthes, Zeus or Jupiter in particular (but not exclusively so: Bendlin 2000c: 271) stand for ‘the highest god’. 41 Versnel 2000: 156: ‘To explain the inexorable, divinity must be depersonalized: the One god is nameless and is not conceived as an approachable personal authority, hence is neither worshipped nor invoked through prayer . . . . Though belonging to the same polythetic class of “gods”, the One and the Many are separate categories. In this respect, they are incomparable, hence do not compete . . .’; Bendlin 2005: 219. 42 Hornung 1982: 243 (for the original, see Hornung 1971: 239). 40

142

Not the One nor the Many

even more primitive stages) toward monotheism. More often than not, scholars have thus adopted a position capable of interpreting a polytheistic religious system only through the lens of a monotheistic perspective.43 But if we narrowly focus on the dichotomy of polytheism versus monotheism, we fail to come any nearer to understanding a polytheistic system: for this focus places the (monotheistic) framing of the enquiry into how to conceptualize the divine—the one or the many?—at the centre of its classificatory typology of religion.44 Long into the Roman Imperial period, however, that question does not constitute a central category of ‘pagan’ religious self-definition; it achieves centrality only in confrontation with early Christian apologetics.45 In order to achieve a methodologically satisfactory understanding of polytheism, the title of my contribution proposes, we must dissolve the misleading dichotomy of the one and the many. For the modern focus on how to conceptualize god, and its reduction to the dichotomy of oneness versus plurality, cannot adequately describe the actual structural properties of polytheistic systems or the problems their structure generates. In ‘pagan’ antiquity the debate over the correct perception of the divine in which Philodemus and Q. Valerius engage, each of them representative of a wider debate, in fact remains a dispute among heterodoxies in ‘pagan’ antiquity. For those engaging in this dispute and their audiences, it circumscribes different theological and religious systems of meaning while at the same time referencing the existence of the respective other system of meaning. Hence, we may rather say that the structural properties conditioning religious action and reflection in a polytheistic system permit ‘insular’ philosophical ideas of one god, even monotheisms (in the plural), which compete but may also be reconciled with opposing systems of meaning. Augustine already put words to the ostensible paradox, averring that the various ‘pagan’ philosophical schools, although they developed contrarian perceptions of the divine, nonetheless participated in the same rituals for the same gods.46 Religious action and philosophical reflection may well belong to two distinct fields of discourse, but in polytheism’s logic they are not incompatible. As tribune of the Roman people, Q. Valerius of Sora, author of the hymn to the one god above all the other gods, also participated in the religious and political rituals of the res publica. In this sense too, the polytheistic system of ancient Rome exemplifies the normalcy of religious pluralism. 43

For a critique of this perspective, see Gladigow 1983: 293; Ahn 1993: 5–12; Gladigow 1998: 322f. 44 45 Ahn 1993: 15–21. On which, see Momigliano 1986. 46 Augustine, De vera religione 2.

Not the One nor the Many

143

THE ETHNOLOGIST’S GAZE For an adequate description of a religious system predicated on plurality, variability, and polyvalence, we require a ‘thick description’ of the religious actions, ideas, and world views of the social actors. Such a description must consider their religious self-representations across highly diverse contexts, but it must also account for the way that place, time, and the social actors’ origin and status determine their varied representations of the gods. The heuristic notion of ‘thick description’ is borrowed from the work of the American cultural anthropologist Clifford Geertz. As is generally known, Geertz, developing earlier theoretical approaches, identifies the task of ethnology, including the ethnology of religion, as ‘thick description’: the goal of the ethnologist (of religion) is not merely to collect and catalogue the material, with reference to which Geertz speaks disparagingly of ‘thin description’. ‘Thick description’ rather aims to observe and explain (religious) behaviour in its cultural context. Actions are public and (following Max Weber) meaningful; but actions communicate meaning on the basis of a common cultural (or religious) symbol system. Hence ‘thick description’ puts the ethnologist in a position to observe and understand the social actors’ actions, verbal propositions, and ideas (which are varied, complex, and may at first seem alien and contradictory) as forms of symbolic communication, and to reconstruct on their basis the cultural (or religious) symbol system represented (symbolically) in them.47 The theoretical insufficiency of the symbolic theory of culture underlying Geertz’s model48 need not interest us here; nor do we have to be concerned about the circular argument that results when a cultural symbol system is first constructed on the basis of actions, and when that symbol system is then deployed to gauge the meaning of these actions.49 What is attractive about ‘thick description’ is the claim that the process of (ethnographical) understanding is not the outcome of psychologizing or phenomenological intuition but rather a result of observation. Also attractive is the heuristic potential entailed in the emic approach of ‘thick description—which, in the special case of the ethnologist engaging with their object through observation, is invariably empathic.50 If the history of ancient religion is to go beyond mere ‘thin description’ it must 47 Geertz 1973: 3–30. For Geertz religion too is a ‘cultural symbol system’, just as religious actions are a symbolic communication: Geertz 1973: 87–125. 48 See Bühl 1987: 59–71; Schulz 1993: 196–214. 49 For a more detailed discussion, with further bibliography, see Bendlin 2002: 24–7. 50 On this emic perspective, see Geertz 1983: 58f.

144

Not the One nor the Many

embrace the role of descriptive and interpretive ethnologist: in addition to a critical historical analysis of the religious-historical data, this task requires a suitable measure of—historically conditioned—empathic imagination. Where the religious and social history of the ancient world currently instrumentalizes post-modern alternatives to the traditional modes of representation, such as fictitious historical narratives or documentary collage, it is with the purpose of getting closer to the object of one’s study through quasi-ethnological observation.51 The differences between modern ethnology and the history of ancient religion are readily apparent, of course: in contrast to the ethnographical field researcher, who visits, observes, and, when he or she so wishes, asks questions of a live object, historians of religion work under the heavy constraints determined by the state of their sources. Even in the case of the comparatively rich documentation about religion in the city of Rome, the fragmentarily transmitted evidence privileges the religious actions and ideas of an educated and wealthy minority, whereas we can scarcely gain access to the religious practices and world views of the vast majority of those living in Rome, or can no longer do so directly. The ethnologist socialized in the Western tradition, like Geertz, may at least entertain the illusion of understanding and describing their Asian, Arabic, or African interlocutor through recourse to emic categories mediated by the cultural encounter. Scholars of the ancient world, by contrast, are by default separated from their object of research not only by millennia, but above and beyond by far more than mere distance between ‘first’ and ‘third’ worlds. Is it possible, despite these differences, to capture through ‘thick description’ the religious actions, ideas, and structures in ancient Rome with the historical ethnologist’s gaze, as it were? This is only ever possible where the transmitted data allow more than the odd snapshot, or where those snapshots might be assembled into a larger whole. The Capitoline Hill in Rome is one such place. On 1 September 22 , the emperor Augustus dedicated a temple to Jupiter Tonans, the ‘Thunderer’, at the entrance to the Area Capitolina. The cult image, a fourth-century bronze statue by the Attic artist Leochares, showed a nude bearded Zeus, stepping forward slightly, holding a sceptre in his raised right hand and in the left a bolt of lightning. This sanctuary, with its marble walls and several statues, was among 51

See, for example, the claim that (religious)-historical reconstruction demands both critical analysis and empathic imagination, in Hopkins 1999: 2 and Morley 1996: 157. In the case of Hopkins (1999), however, the experiment with fictitious historical narrative must nevertheless be deemed a failure: see Bowersock 2000; Kinzig 2000.

Not the One nor the Many

145

Augustus’ most lavish and magnificent building projects. The temple is mentioned for its importance both in the Res Gestae, published posthumously in front of the Princeps’ mausoleum in the Field of Mars, and by Suetonius.52 The Princeps had vowed this temple to the god in the year 26  in Cantabria as a gesture of gratitude: a bolt of lightning (fulgur) is said to have grazed his litter, but instead of him it struck and killed a slave who was lighting the way. In Roman religion such fulgura were often taken as bad omens: once a spot had been struck by lightning, it became a locus religiosus; the person killed by lightning was denied the iusta, the burial rites; the lightning strike itself had to be ritually expiated. Augustus, however, interprets the incident as a positive sign, for which he too is able to invoke religious tradition: Etruscan doctrine on lightning apparently argued that a political leader, or a king, who had been struck by lightning but survived the incident was destined for political success and eternal fame, as was his offspring. The temple dedication ritual on the Capitoline Hill on 1 September 22  was reportedly accompanied by a rumble of thunder with which the god expressed his satisfaction.53 As a matter of fact Jupiter Tonans, because he is intimately connected with the emperor, is a dynastic, well-nigh ‘royal’ god: while thundering [tonans] Jupiter, master of lightning and thunder, presides on Mount Olympus, Horace writes in the third book of his Odes in 23 , Augustus is considered a praesens divus already during his lifetime.54 After the temple dedication Augustus (whom we know to have paid much attention to his own dreams and to those of others) has a dream:55 when he frequents the temple of Jupiter Tonans, Jupiter Optimus Maximus appears to him to complain that the rituals for Jupiter Tonans on the Capitoline Hill relegate him to second place inasmuch as they deprive him of visitors. The threat was real: in the Augustan period the ancient cult statue of a seated Jupiter in the Temple of Jupiter in the Forum in Pompeii, which had been wrought after the image of Jupiter Optimus Maximus in the city of Rome, was replaced by a new cult image that took the form of the statue type of Jupiter Tonans.56 In Rome it must not 52 Res Gestae 19; Suetonius, Divus Augustus 29; Dio Cassius 54. 4. 2; Inscriptiones Italiae 13. 2, p. 504; Wissowa 1912: 122; Platner and Ashby 1929: 305f.; Martin 1988: 254f. with cat. no. 253; Gros 1996: 159f. with fig. 107. 53 Suetonius, Divus Augustus 29. 3; Dio Cassius 54. 4. 2. On bolts of lightning in Roman religion, see Cicero, De legibus 2. 21; Festus 190L; Festus (P) 82L; Le Bourdellès 1973; Schilling 1974. For the Etruscan doctrine on lightning, see Thulin 1905–9: 78, 90f. 54 Horace, Carmina 3. 5. 1–3; Martin 1988: 254. 55 Suetonius, Divus Augustus 91.2; Dio Cassius 54. 4. 2–4. On Augustus’ dreams, see in general Suetonius, Divus Augustus 91. 1; Weber 2003. 56 Martin 1988: 255.

146

Not the One nor the Many

come to this: in his dream Augustus responds to the ‘Best and Greatest’ that Jupiter Tonans was placed at his, the god’s, side to serve as gatekeeper (ianitor) of the Area Capitolina. Subsequently he has small bells fixed to the pediment of the Temple of Jupiter Tonans (so Suetonius reports; Cassius Dio links them to the god’s cult image). Afterwards he personally requests a monetary donation of the urban population once a year, apparently earmarked for the cultic benefit of Jupiter Optimus Maximus. Cassius Dio explains the small bells with reference to the Roman urban custom of night watchmen carrying bells so that they could warn the people of danger. On this explanation the small bells would symbolize how the status of Jupiter Tonans was now subordinate to that of Jupiter Optimus Maximus: as a rule, door- and gatekeepers as well as night watchmen were of lower rank or slaves. Once a year Augustus too abases himself and begs the population of Rome for money. Bells can indeed have such a meaning in the Roman cultural symbol system. But is it conceivable that Augustus humiliated Jupiter Tonans in such a way? Should one imagine Augustus to have created such a stark hierarchy within the pantheon? By rights would he not then have to reckon with the wrath of Jupiter Tonans? An alternative interpretation, drawing on the function of small bells in ritual contexts where their ringing was supposed to calm the anger of the gods, provides a solution. Against the background of this symbolic meaning of small bells, their installation above the entrance to the temple of the lord of lightning and thunder is quite a sensible precaution against future calamity.57 Cassius Dio appears to have had access, imparted to him by other sources, to Augustus’ records for the period after 1 September 22 , since he quotes Augustus’ own words: the anger of Jupiter Optimus Maximus is understandable, Augustus avers, since the urban population also, like Augustus himself, preferred Jupiter Tonans, firstly because of the novelty of the god’s name and his cult statue; secondly because cult was established by the Princeps; but above all because anyone entering the Area Capitolina first had to pass by the Temple of Jupiter Tonans. Augustus’ dream and the conflicts fought in it could be spun further: there was a Republican Temple of Jupiter Fulgur in Rome, whose foundation date fell on 7 October.58 What might this more ancient god of lightning have had to say about the competition of Jupiter Tonans’, had he appeared to Augustus in a dream?

57 For the range of meanings (small) bells had in Roman profane and religious contexts, see Hinard and Dumont 2003: 122. 58 Schilling 1974; Manacorda 1996: 136–8.

Not the One nor the Many

147

The cult for Jupiter Tonans and the dream appearance of Jupiter Optimus Maximus confront the Princeps with the problems that arise from the possibilities inherent in structuring the Roman pantheon, but also from the rules that apply to it. The various epithets (toponyms are another possibility) serve as instruments to achieve a more extensive differentiation among the gods in this pantheon, which quickly exceeds the manageable number of only a few ‘major gods’;59 thus no limits are set to the Roman pantheon’s expansion. I would like to demonstrate this very briefly just for Jupiter. Alongside Jupiter Optimus Maximus, Jupiter Tonans, and Jupiter Fulgur, a Jupiter Africus is attested in the Area Capitolina, his shrine dating perhaps to the second century . Jupiter Pistor, who was taken by Wissowa to be another god of lightning, has an altar in the same area during the Augustan period; the epithet ‘baker’, though, should make us rather consider a connection of the god with this profession. Domitian first had a shrine to Jupiter Conservator (the ‘Protector’) established near the Capitoline Hill, which he later replaced with a larger temple to Jupiter Custos (the ‘Guardian’). Under Claudius an altar to Jupiter Depulsor (the ‘Averter’) was established on the Capitoline Hill itself. The old Republican temple to Jupiter Feretrius in the Area Capitolina was among the shrines Augustus restored. Jupiter Elicius, whose exact nature remains mysterious, had his altar on the Aventine. The restoration of a temple to Jupiter Liber, which should also be located on the Aventine, is again mentioned in Augustus’ Res Gestae. Jupiter Viminus was worshipped at an altar on the Viminal and the god is believed to have lent his name to that hill. The shrine of Jupiter Fagutalis, which stood in the Lucus Fagutalis, gave this god’s name to a neighbourhood (vicus) of Rome. A Jupiter Tragoedus is found among the statues dedicated by Augustus in the Roman vici. The foundation day of the Temple of Jupiter Victor was 13 April. The temple of another god of victory, Jupiter Invictus (the ‘Undefeated’) on the Palatine, dates back to as early as the third century . Jupiter Propugnator (the ‘Protector’) is not identical with that god: his shrine, likewise on the Palatine, dates to the High Imperial period. A god of oaths, Jupiter Iurarius, was worshipped on the Isola Tiberina as early as in the Republican period. Jupiter Stator (the ‘Steadfast’) even possessed two places of worship in Republican Rome. The dedication of a shrine to Jupiter Ultor (the ‘Avenger’) by the emperor Alexander Severus in 222  only added to this plurality.60

59

On this topic, see also Gladigow 1981: 1202–37; Parker 2003: 173–83. Jupiter Africus: Chioffi 1996—Jupiter Pistor: Ovid, Fasti 6. 349–94; Wissowa 1912: 122; Aronen 1996a—Jupiter Conservator: Reusser 1996—Jupiter Depulsor: Chioffi 1996a— 60

148

Not the One nor the Many

How can we interpret these internal differentiations within the pantheon, all of them the result of epithetizing or toponymity? For the Paris School these gods are ‘powers’ that lack any more wide-ranging individuality, but who, because they are not persons, exist only within the logic of their pantheon’s structuralist frame of reference. The epithet substantiates their place in that frame of reference.61 With reference to the list of gods in the commentarii of the Fratres Arvales, John Scheid interprets the ‘epithetization’ of the gods as a deity’s progressive atomization into her or his different functional aspects and narrowly confined spheres of action, which supplement one another in a relational system of complementary activities—a system again understood to operate along structural(ist) lines.62 For Erik Hornung too the differentiation of the divine world is a logical problem, which must be resolved by way of a polyvalent logic: ‘the one and the many’ is recast as a matter of god as ‘a unity in worship and revelation, and multiple in nature and manifestation’. In using this formulation Hornung merely perpetuates one particular theological discourse on the object level, which can be found in this form in the ancient Egyptian sources or among Stoics and Platonists in the Hellenistic and Roman Imperial periods.63 Surely, the Epicurean Philodemus would have vehemently objected. We must reach a different verdict when we set these claims against Augustus’ dream, where we encounter two gods to whom personality is attributed and who, in competition with one another, display subjective agency. Although both carry the name of Jupiter and ostensibly there is ‘only’ an epithet to distinguish them (an epithet that, in the case of Jupiter Tonans, indeed denotes a primary function), we are not dealing with a case of functional additions in a logical system of meaning and

Jupiter Feretrius: Res Gestae 19; Coarelli 1996—Jupiter Elicius: Bendlin 1999—Jupiter Liber: Res Gestae 19; ‘. . . Iovis Libertatis’; on this name, see Andreussi 1996—Jupiter Viminus: Aronen 1996b—Jupiter Fagutalis: Varro, De lingua Latina 5. 152; Pliny, Natural History 16. 37; Festus (P) 77L—Jupiter Tragoedus: Suetonius, Divus Augustus 57—Jupiter Victor: Coarelli 1996b—Jupiter Invictus: Ziolkowski 1992: 80–5—Jupiter Propugnator: Chioffi 1996b—Jupiter Iurarius: D. Degrassi 1996—Jupiter Stator: Coarelli 1996a; Viscogliosi 1996—Jupiter Ultor: Ziolkowski 1992: 80–5 on the sanctuary’s relation to the temple for Jupiter Victor. 61

Bruit Zaidman and Schmitt Pantel 1992: 177f., 184–6. Scheid 2003: 185–7, esp. 186, who views the epithets as ‘. . . aspects of the action of a deity to whom they are linked according to need or context . . . . [A]ll of these divine figures and all of these polytheistic structures work together to express, within a given context, one sole concept and one sole action’. 63 Hornung 1982: 242 (= Hornung 1971: 239). For ancient variations on Hornung’s formulation, see Assmann 2004: 24–8 (in a section misleadingly entitled ‘evolutionary monotheism’). 62

Not the One nor the Many

149

action. Rather, the two gods are ‘persons’ differentiated by way of their individuality, who appear before our eyes as autonomous social actors and, in the case of Jupiter Optimus Maximus, display emotions. Neither is merely a variable in a relational system characterized by structure; one experiences both in their complex individuality. The process of ‘epithetization’ contributes to such an attribution of personality and individuality. Moreover, both gods differ from one another in their appearance; for this reason, too, they are considered distinct by the population of Rome. One approaches Jupiter Tonans because the god is novel and different, and because his worship has been established by the most powerful man in Rome. The fact that the temple can be reached even more quickly than the sanctuary of the Capitoline triad further increases the attractivity of Jupiter Tonans in an open religious market of supply and demand.64 If we grant these divine social actors a personality, we may resolve another problem—one that is probably more of a logical than a pragmatic kind: in religious-philosophical or theological discourse it can be argued that there exists one Jupiter behind Jupiter Optimus Maximus and Jupiter Tonans (and behind all Jupiter-gods in Rome). From a ritualpragmatic perspective the differentiation of the divine world by name or place (that is, by means of epithets or toponyms) and the anchoring of the gods in a local space (with their own temple, festival, and cult image) guarantees their distinctiveness; were it not for this condition, the social actors would be incapable of any meaningful religious agency on the ground. At the same time the god’s first name allows them to reconnect any particularized rituals to more widely shared theologies or world views. Jupiter Tonans competes so successfully against Jupiter Optimus Maximus because of his difference—but also because the alikeness of their names establishes a system of mutual relations and hence generates compatibility. Attributing personality and individuality to gods, which can be observed here, must not occasion surprise. From the perspective of behavioural psychology, ascribing these two character traits to deities is the rule regardless of the religious systems in which the phenomenon can be observed, be they monotheistic or polytheistic.65 One’s interaction with gods perceived as persons follows the same behavioural patterns that social actors display when they interact with one another. Augustus explains to himself why Jupiter Optimus Maximus has a right to be angry: anyone else, including Augustus, would react with as much frustration as the god if they were in a comparable situation. Only our

64 65

For the analysis of polytheistic religions as a religious market, see Bendlin 2000e: 134f. See Barrett and Keil 1996.

150

Not the One nor the Many

attributing personality and individuality to the gods enables us to entertain a unique relationship with them. In a personal crisis situation Augustus chooses Jupiter Tonans as his personal divine protector, thereby also deciding not to privilege Jupiter Optimus Maximus.66 A polytheistic system does not only allow for but apparently generates processes of selection and privileging, which reduces the plurality of deities on supply from among a theoretically inexhaustible reservoir: polytheism does not simply mean that one worships a plurality of gods, but it implies that one selects from among such plurality. The alternative between the one and the many is again misleading. In light of a pluralistic market of supply, establishing interpersonal relationships to deities does not usually lead to ‘monotheistic’ decisions;67 but meaningful relationships with an indefinite number of gods are as unsustainable as they would be with an unlimited number of social actors. This explains why in the reality of ritual practice an individual’s pantheon must be strictly limited in size, even when it features a plurality of divine addressees (by the same token, maintaining a relationship to just one such addressee in monotheistic religions is also exceptional, at least on the level of religious practice).68

In addition to Iuppiter Tonans, Augustus privileged two more ‘dynastic’ gods, and their temples became centres of political as well as religious communication in the Augustan period, in competition with the Temple of the Capitoline triad: as early as in 42 , during the Civil War, the young Octavian vowed to Mars Ultor (the ‘Avenger’) a temple, which had originally been planned on the Capitoline Hill, but which was eventually dedicated in 2  as the focus of Augustus’ new Forum (Rich 1998: 79–89; Spannagel 1999: 41–72). The Temple of Apollo Palatinus (‘on the Palatine’), vowed during the war in 36 , was dedicated in 28 : the magnificent building complex included, inter alia, a shrine of Vesta and the House of Augustus (Platner and Ashby 1929: 16–19). 67 The conflict between Jupiter Optimus Maximus and Jupiter Tonans also explores how personal religious loyalties can be lived amid the religious plurality of a polytheistic system. In the Euripidean tragedy of the same name, Hippolytus neglects Aphrodite and the other gods in favour of Artemis and suffers death because of his contempt of the other deities (on this conflict, see Gladigow 1990). The conflict and its resolution dramatize, on the level of the mythological thought-experiment, the necessity or rather expectation within a polytheistic system to weigh the available religious options and alternatives and find a compromise. Augustus’ resolution of this loyalty conflict also demonstrates that conflicts with the gods can be settled by ritual-pragmatic means (an outcome frustrated in the myth’s narrative structure) and personal preferences can be respected, and that compromises are possible even with the gods. 68 See Gladigow 1998: 324; Bendlin 2001a: 81f. These considerations find indirect confirmation in the archaeological evidence of house-shrines in Campania and in the northwestern provinces of the Roman Empire: not only the Lares and the Genius of the master of the household but, in addition, a plurality of different divinities were the recipients of worship in the (so-called) Lararia. Their selection was dictated by local, regional, or current trends, but it was not predicated on any binding rules; their number varied from one up to ten with on 66

Not the One nor the Many

151

What are the parameters according to which selections are made here? Which ideas, wishes, political, social, and cultural expectations condition the social actors’ choices? Employing the ethnologist’s gaze and ‘thick description’, let us turn one last time to the Capitoline Hill. The Area Capitolina was not merely the stage of competition between Jupiter Optimus Maximus and Jupiter Tonans. Alongside the sanctuaries to the deities of the Capitoline triad, Jupiter Tonans and Jupiter Feretrius, there existed further temples to other gods as well as numerous cults whose organisation levels were less institutionalized, among them cult for Isis Capitolina early in the first century .69 At the same time the Area Capitolina and the Temple of Jupiter Optimus Maximus, Juno, and Minerva was the location par excellence of political ritual in the city. The consuls and praetors sacrificed to Jupiter Optimus Maximus on the day they entered office and offered vows for the duration of their term of office. The cella of the ‘highest and greatest’ Jupiter in that temple, whose architecture reflected the Roman republic’s political ambitions, was the final destination of the triumphal procession of Roman generals. The temple served as the place where treaties were kept and also accommodated the Sibylline Books, before Augustus had these transferred to the Temple of Apollo on the Palatine. Dating to the second and first centuries , numerous dedications to Jupiter Optimus Maximus (and the Roman People) by kings and allied communities are preserved on the Capitoline Hill, serving as indication of the god’s central political role both domestically and in the relationship with foreigners.70 But the Area Capitolina was littered with other dedications too, as well as with altars and statues. Young men are believed to have marked their receiving of the toga virilis and acceptance into the Roman citizens community with a sacrifice to Jupiter Optimus Maximus. Even the Fratres Arvales left the grove of Dea Dia on occasion for sacrifices to average four to six divinities. For the data, see Kaufmann-Heinimann 1998: esp. 184–8, 191–5, 315–18. The studies by Stark and Finke (2000: esp. 218–28) on the relation of religious pluralism and religious commitment in the United States may be read as being complementary to this evidence: a pluralistic situation, that is, the availability of several (Christian) churches competing with one another, stimulates religious commitment, rather than being detrimental to it, as the members of a ‘congregation’ also join other congregations. At a certain point, however, saturation of the market is reached: the religious commitment does not diminish, but neither does it increase any further. 69 Versluys 2004. For a helpful survey of the monuments of the Area Capitolina’s (sacred) topography (including a map), see Reusser 1993: 32–51. 70 Sacrifices and vows: Orlin 1997: 36–45. Architecture: Martin 1983. Triumph: Künzl 1988: 85–108. Dedications: ILLRP 174–81 = A. Degrassi 1962–71: vol. 1, 415–44; Beard et al. 1998: vol. 1, 158. Reusser (1993: 138–58) proposes a date for these dedications in the Sullan period.

152

Not the One nor the Many

the Capitoline triad. We learn from the Acta Diurna of Augustus, who knew how to exploit the event for propagandistic purposes, that a certain C. Crispinius Hilarus of the Plebs Faesulana offered sacrifice to Jupiter Optimus Maximus on the Capitoline Hill in the Augustan period, accompanied by his eight children, twenty-seven grandsons, eighteen great-grandsons, and eight granddaughters.71 Seneca enriches our picture of religious life on the Capitoline Hill with another facet:72 All the same there is a fixed time for this madness [a reference to an annual festival in the cult of Isis]. It is tolerable to lose one’s senses once a year. But go to the Capitoline Hill: you will feel shame at the public display of insanity [there]: with what service does vain frenzy charge itself! One person supplies to the god the names [of visitors], another announces to Jupiter the hours [of the day], a third is his {lictor and a fourth his anointer, who with vain movement of the arms only mimics one who anoints. There are women who dress the hair of Juno and Minerva (they stand some distance away not only from the image, but from the temple, and move their fingers in the manner of coiffeurs), while other women hold up the mirror. There are people who, because they are bound to appear in court, call upon the gods for aid, while others present their petitions and expound to them their legal cases. A learned chief mime, an old man, long decrepit, used to perform a mime every day on the Capitoline Hill—as if the gods were glad to provide an audience for him, whom his human spectators had deserted. All sorts of craftspeople hang around there so as to provide their labour to the immortal gods. All the same these people promise the god something that though it is of no use, is at least not unseemly or disreputable. But on the Capitoline Hill there sit around some women who believe that they are loved by Jupiter; they are not even 71

Fratres Arvales: Beard et al. 1998: vol. 1, 195f. Crispinius Hilarus: Pliny, Natural History 7. 60. 72 Seneca, De superstitione fgt. 35–6 (Haase) = 69–70 (Vottero), preserved in Augustine, De civitate Dei 6. 10: Huic tamen . . . furori certum tempus est; tolerabile est semel anno insanire. In Capitolium perveni: pudebit publicatae dementiae, quod sibi vanus furor adtribuit officii. Alius nomina deo subicit, alius horas Iovi nuntiat, alius {lictor est, alius unctor qui vano motu bracchiorum imitatur unguentem. Sunt quae Iunoni ac Minervae capillos disponant (longe a templo non tantum a simulacro stantes digitos movent ornantium modo), sunt quae speculum teneant. Sunt qui ad vadimonia sua deos advocent, sunt qui libellos offerant et illos causam suam doceant. Doctus archimimus, senex iam decrepitus, cotidie in Capitolio mimum agebat, quasi dii libenter spectarent quem homines desierant. Omne illic artificum genus operatum diis immortalibus desidet. Hi tamen . . . etiamsi supervauum usum, non turpem nec infamem deo promittunt. Sedent quaedam in Capitolio quae se a Iove amari putant: ne Iunonis quidem, si credere poetis velis, iracundissimae respectu terrentur.

Not the One nor the Many

153

frightened by the thought of Juno who, if one wanted to believe the poets, is terribly irascible. The addressees of these ‘rituals’ are the three deities of the Capitoline triad:73 Jupiter Optimus Maximus, sitting in the temple’s central cella with a mantle on his lower body and over his left shoulder, and holding a bolt of lightning in the right hand and a sceptre in the raised left hand; Juno occupies the cella to his left, Minerva the one to his right. The social actors, men and women, among them quite likely the free and the unfree, Roman citizens and foreigners, strive to make personal contact with the deities, but they are incapable of advancing into the temple, let alone its cellae. From afar some of them perform labours that are the tasks of slaves in the houses of the prosperous and the powerful. From a distance, others pay their respects to the gods in the same way as a client pays his early matutinal respects to a patron in order to ensure he has his support in court.74 Actors and craftspeople offer their services. Women even think that the god has fallen in love with them, a priori not an implausible hope by any means, in view of the mythological past of Zeus-Jupiter. As Seneca, the Stoic philosopher, suggests to us, all of them abase themselves before their fellow humans and the gods—an example of emotionally excessive and ridiculous religious behaviour (superstitio).75 If one wishes to venture beyond the philosophical critique of religion and show respect for the social actors’ intention to get closer to ‘their’ gods, win their favour and profit from their power, another picture emerges. In order to put their wishes into effect, the social actors make use of behavioural patterns, social rituals, and services that no longer have much to do with conventional, traditional religious ritual. Their ‘rituals’ are tried and tested in everyday life. As they interact with the Capitoline triad, these rituals promise success because the gods too are conceptualized as social actors in the everyday lives of their worshipers, that is, as a dominus and patron or the recipient of services. In the presence of these social rituals the unbridgeable distance between god and human has already been abandoned. For this reason, the ritually correct form of worshiping the gods—through prayer and sacrifice at the altar in front of the temple—is no longer required since the gods are already embedded in their worshippers’ social interaction structures.

73

On this passage, see Gladigow 1994; Estienne 2001. For the division of the social actors into ‘slaves’ and ‘clients’, see the parallel passage, Seneca, Epistles 95. 47. 75 The work of Seneca from which this passage is taken carries the title De superstitione. For philosophical evaluations of such behaviour as superstitio, see also Theophrastus, De pietate fgt. 584D (Fortenbaugh); Plutarch, De superstitione 12. 171A–B; 2. 165B. 74

154

Not the One nor the Many

Religious pluralism in a polytheistic system is not merely the result of a quantitative expansion of a divine world that is anchored locally or regionally. More gods do not necessarily amount to a greater degree of religious pluralism. Even when there exists a limited number of traditional divine addressees—and which god could be more traditional than Jupiter Optimus Maximus?—the personalization and individualization of people’s relationships with ‘their’ gods result in pluralistic patterns of action and meaning; these we can only observe in their polyvalence. ‘Thick description’ therefore yields no unified cultural symbol system. Nor does it provide structure where structures are heterogeneous, and the principles of classification are ambivalent. But as a snapshot it can throw an instructive light on the variability of religious behaviour in an ancient polytheistic system. Religious pluralism is no recent, postmodern phenomenon.

A FT E R W O R D Several of the topics I sketched in this chapter have since received a more detailed treatment. For example, projects such as the corpus of ‘cult places’ in ancient Italy, Fana, templa, delubra (2008–), proffer useful repertories of local ritual activity, redeeming the investigation of ‘local religion’. A number of studies provide a fuller picture, complementary to the claims I make, of the religious agency of Roman colonies and other Roman communities, highlighting also how the religious landscapes in these nominally Roman communities emerged in autonomy from Rome. Others have examined alternative religious landscapes in the city of Rome— albeit usually with a focus on cults and religions rather than the social actors. Besides, neither ‘cults’ nor ‘religions’ are innocent analytical tools (cf. Gordon 2014: 676); their modern deployment tends to essentialize what, this chapter argues, were often fluid structures of local religious practice. The less savoury aspects of local religion must be scrutinized more critically. I am thinking, for example, of the religion of migrants, foreigners, and diasporic religious networks in the Roman city. The current claim that different deities cohabitated by and large consensually in the polytheistic marketplace (before late antiquity, at any rate), because it is predicated on problematic definitions of religious pluralism in antiquity, prevents us from recognizing polytheism’s dark side. Selecting one’s gods from among a wide range of possible choices, this chapter suggests, implies rejecting other gods, as well as exclusion of, and occasional violence against, the religious Other (see now Bendlin 2020). The civic-religion approach—that is, any attempts to view Roman religion to a significant extent through the lens of urban religion’s embeddedness in the city’s political institutions—has also been critiqued by others. I am doubtful, however,

Not the One nor the Many

155

whether the alternative focus on individual religious agency—as advocated for by those endeavouring to reconstruct what they call ‘lived ancient religion’ (for example, Rüpke 2011; Albrecht et al. 2018)—proffers a viable methodological alternative. Shifting the focus to the individual’s appropriation of shared religious practices no doubt offers a helpful corrective to the study of institutionalized religion in the city-state. The obvious attractiveness of this corrective, however, must not mask its methodological shortcomings. From the opposite end, the privileging of individual agency faces (yet does not confront) the same sociological challenge as the paradigms it professes to have overcome, namely, how to calibrate the relation between structure and agency. In devoting one’s attention to individual religious agency, one quickly loses sight of the structures and strictures framing religious activity on the ground. In the course of my outline of polytheism’s structural properties, I note how polytheism and monotheism are inventions of an early modern religious taxonomy. The intellectual tradition underlying this taxonomy is still with us, not least in the evolutionist model of a ‘pagan monotheism’ prevalent, it is averred, among many cultured elites of the Imperial period. I argue that the creation of monotheisms (note the plural) is not a radically distinct way of religious thinking, but one of many variables written into polytheism’s structural properties. Understanding these properties thus means that we better understand religion in antiquity. This implies that we retire the dichotomy of polytheism versus monotheism—that is, that we leave behind the routine mode of thinking scholarship has employed to model religious evolution.

B I B L I OG R A P H Y Ahn, G. (1993). ‘ “Monotheismus”—“Polytheismus”. Grenzen und Möglichkeiten einer Klassifikation von Gottesvorstellungen’, in M. Dietrich and O. Loretz (eds.), Mesopotamica—Ugaritica—Biblica. Neukirchen-Vluyn, 1–24. Albrecht, J. et al. (2018). ‘Religion in the making: the Lived Ancient Religion approach’. Religion 48.1: 1–26. Andreussi, M. (1996). ‘Iuppiter Libertas, Aedes’. LTUR 3: 144. Aronen, J. (1996a). ‘Iuppiter Pistor, Ara’. LTUR 3: 154f. Aronen, J. (1996b). ‘Iuppiter Viminus, Ara’. LTUR 3: 162. Assmann, J. (2004). ‘Monotheism and polytheism’, in S. I. Johnston (ed.), Religions of the Ancient World: A Guide. Cambridge, MA—London, 17–31. Barrett, J. L. and F. C. Keil (1996). ‘Conceptualising a non-natural entity: anthropomorphism in god concepts’. Cognitive Psychology 31: 219–47. Beard, M., J. N. North, and S. R. F. Price (1998). Religions of Rome. 2 vols. Cambridge. Belayche, N. (2000). ‘L’Oronte et le Tibre: l’Orient des cultes “orientaux” de l’Empire romain’, in M. A. Amir-Moezzi and J. Scheid (eds.), L’Orient dans l’histoire religieuse de l’Europe: l’invention des origins. Turnhout, 1–35.

156

Not the One nor the Many

Bendlin, A. (1997). ‘Peripheral centres—central peripheries: religious communication in the Roman Empire’, in Cancik and Rüpke, 1997, 35–68. Bendlin, A. (1999). ‘Manalis lapis’. DNP 7: 789 (tr. in Brill’s New Pauly 8 [2006]). Bendlin, A. (2000a). ‘Novensides Di’. DNP 8: 1028f. (tr. in Brill’s New Pauly 9 [2006]) Bendlin, A. (2000b). ‘Pantheon III: Klassische Antike’. DNP 9: 265–68 (tr. in Brill’s New Pauly 10 [2007]). Bendlin, A. (2000c). ‘Pantheos, Pantheios’. DNP 9: 270–2. Bendlin, A. (2000d). ‘Patrii Di’. DNP 9: 410–12 (tr. In Brill’s New Pauly 10 [2007]). Bendlin, A. (2000e). ‘Looking beyond the civic compromise: religious pluralism in Late Republican Rome’, in E. Bispham and C. Smith (eds.), Religion in Archaic and Republican Rome. Evidence and Experience. Edinburgh, 115–35, 167–71. Bendlin, A. (2001a). ‘Polytheismus’. DNP 10: 80–3 (tr. in Brill’s New Pauly 11 [2007]). Bendlin, A. (2001b). ‘Rituals or beliefs? “Religion” and the religious life of Rome’. Scripta Classica Israelica 20: 191–208. Bendlin, A. (2002). ‘Gemeinschaft, Öffentlichkeit und Identität. Forschungsgeschichliche Anmerkungen zu den Mustern sozialer Ordnung in Rome’, in U. Egelhaaf-Gaiser and A. Schäfer (eds.), Religiöse Vereine in der römischen Antike. Tübingen, 9–40. Bendlin, A. (2005). ‘Wer braucht “Heilige Schriften”? Die Textbezogenheit der Religionsgeschichte und das “Reden über die Götter” in der griechischrömischen Antike’, in Ch. Bultmann, C.-P. März, and V. N. Makrides (eds.), Heilige Schriften. Ursprung, Geltung und Gebrauch. Münster, 205–28, 251–4. Bendlin, A. (2006). ‘ “Eine wenig Sinn für Religiosität verratende Betrachtungsweise”: Emotion und Orient in der römischen Religionsgeschichtsschreibung der Moderne’. Archiv für Religionsgeschichte 8: 227–56. Bendlin, A. (2020). ‘Religion, violence and the diasporic experience: the Jewish diaspora in Flavian Rome and Puteoli’. in J. H. F. Dijkstra and C. R. Raschle, (eds.), Religious Violence in the Ancient World. Cambridge, 133–76. Betegh, G. (2004). The Derveni Papyrus. Cosmology, Theology and Interpretation. Cambridge. Bowersock, G. (2000). ‘Early Christianity as historical fiction’. JRA 13.2: 763–6. Brelich, A. (1960). ‘Der Polytheismus’. Numen 7: 123–36. Bremmer, J. N. (1994). Greek Religion. Oxford. Bruit Zaidman, L. and P. Schmitt Pantel (1992). Religion in the Ancient Greek City. Cambridge. (Originally published as La religion grecque dans les cités à l’époque Classique. Paris [19912].) Bühl, W. L. (1987). Kulturwandel. Für eine dynamische Kultursoziologie. Darmstadt. Bullo, S. (1992). ‘Le indicazioni di Vitruvio sulla localizzazione dei templi urbani (de arch 1.7.1): il caso africano’. L’Africa Romana 10: 515–58. Cancik, H. and J. Rüpke, eds. (1997). Römische Reichsreligion und Provinzialreligion. Tübingen. Cardauns, B., ed. (1960). Varros Logistoricus über die Götterverehrung. Würzburg.

Not the One nor the Many

157

Chioffi, L. (1996). ‘Iuppiter Africus’. LTUR 3: 130. Chioffi, L. (1996a). ‘Iuppiter Depulsor, Ara’. LTUR 3: 132. Chioffi, L. (1996b). ‘Iuppiter Propugnator, Aedes’. LTUR 3: 155. Clarke, G. W. (1974). The Octavius of Marcus Minucius Felix. New York. Coarelli, F. (1996). ‘Iuppiter Feretrius, Aedes’. LTUR 3: 135f. Coarelli, F. (1996a). ‘Iuppiter Stator, Aedes’. LTUR 3: 155–7. Coarelli, F. (1996b). ‘Iuppiter Victor, Templum’. LTUR 3: 161. Cornell, T. J. (1995). The Beginnings of Rome. Italy and Rome from the Bronze Age to the Punic Wars (c. 1000–264 ). London. Courtney, E. (1993). The Fragmentary Latin Poets. Oxford. Crawford, M. H., ed. (1996). Romans Statutes. 2 vols. London. Degrassi, A. (1962–71). Scritti vari di antiquità. 4 vols. Venice—Trieste. Degrassi, D. (1996). ‘Iuppiter Iurarius’. LTUR 3: 143f. Detienne, M. (1997). ‘Expérimenter dans le champ des polythéismes’. Kernos 10: 57–72. Dumézil, G. (1970). Archaic Roman Religion. Chicago. (Originally published as La religion romaine archaïque. Paris [1966].) Edwards, C. and G. D. Woolf (2003). ‘Cosmopolis: Rome as world city’, in Edwards and Woolf, 2003, 1–20. Edwards, C. and G. D. Woolf, eds. (2003). Rome the Cosmopolis. Cambridge. Estienne, S. (2001). ‘Les “dévots” du Capitole. Le “culte des images” dans la Rome impériale, entre rites et superstition’. MEFRA 113: 189–210. Fontana, F. (1997). I culti di Aquileia repubblicana. Aspetti della politica religiosa in Gallia Cisalpina tra il III e il II sec. a.c. Rome. Fowden, G. (1993). Empire to Commonwealth: Consequences of Monotheism in Late Antiquity. Princeton. Fröhlich, Th. (1991). Lararien und Fassadenbilder in den Vesuvstädten. Untersuchungen zur volkstümlichen pompejanischen Malerei. Mainz. Geertz, C. (1973). The Interpretation of Cultures. New York. Geertz, C. (1983). Local Knowledge. Further Essays in Interpretive Anthropology. New York. Gladigow, B. (1981). ‘Gottesnamen (Gottesepitheta)’. RAC 11: 1202–37. Gladigow, B. (1981a). ‘Kraft, Macht, Herrschaft. Zur Religionsgeschichte politischer Begriffe’, in B. Gladigow (ed.), Staat und Religion. Düsseldorf, 7–22. Gladigow, B. (1983). ‘Strukturprobleme polytheistischer Religionen’. Saeculum 34: 292–304 (repr. in his Religionswissenschaft als Kulturwissenschaft. Stuttgart [2005], 125–37). Gladigow, B. (1990). ‘Chrēsthai Theois. Orientierungs- und Loyalitätskonflikte in der griechischen Religion’, in Ch. Elsas and H. G. Kippenberg (eds.), Loyalitätskonflikte in der Religionsgeschichte. Würzburg, 237–51 (repr. in his Religionswissenschaft als Kulturwissenschaft. Stuttgart [2005], 138–48). Gladigow, B. (1993). ‘Gottesvorstellungen’. HrwG 3: 43–9. Gladigow, B. (1994). ‘Zur Ikonographie und Pragmatik römischer Kultbilder’, in H. Keller and N. Staubach (eds.), Iconologia sacra. Mythos, Bildkunst und Dichtung in der Religions- und Sozialgeschichte Alteuropas. Berlin, 9–24. Gladigow, B. (1998). ‘Polytheismus’. HrwG 4: 321–30.

158

Not the One nor the Many

Godfrey, P. and D. Hemsoll (1986). ‘The Pantheon: temple or rotunda?’, in M. Henig and A. King (eds.), Pagan Gods and Shrines of the Roman Empire. Oxford, 195–209. Gordon, R. (2000). ‘Griechisch-Orientalische Kulte’. RGG4 3: 1292f. Gordon, R. (2014). ‘Monotheism, henotheism, megatheism: debating preConstantinian religious change’. JRA 27.2: 665–76. Gros, P. (1996). ‘Iuppiter Tonans, Aedes’. LTUR 3: 159f. Heibges, U. (1969). ‘Religion and rhetoric in Cicero’s speeches’. Latomus 28: 833–49. Hinard, F. and J. Ch. Dumont, eds. (2003). Libitina. Pompes funèbres et supplices en Companie à l’époque d’Auguste. Paris. Hopkins, M. K. (1999). A World Full of Gods. The Strange Triumph of Christianity. London. Hornung, E. (1971). Der Eine und die Vielen. Ägyptische Gottesvorstellungen. Darmstadt. Hornung, E. (1982). Conceptions of God in Ancient Egypt. The One and the Many. Ithaca. Hülsewiesche, R. and S. Lorenz (1989). ‘Polytheismus I’. HWdPH 7: 1087–93. Isaac, B. (2004). The Invention of Racism in Classical Antiquity. Princeton. Jongman, W. (2003). ‘Slavery and the growth of Rome: the transformation of Italy in the second and first centuries ’, in Edwards and Woolf, 2003, 100–22. Kaufmann-Heinimann, A. (1998). Götter und Lararien aus Augusta Raurica. Augst. Kinzig, W. (2000). ‘Letter to Keith Hopkins’. ZAC 4: 219–24. Kratz, R. G. and H. Spieckermann, eds. (2006). Götterbilder—Gottesbilder— Weltbilder: Polytheismus und Monotheismus in der Welt der Antike. 2 vols. Tübingen (2nd edn [2009]). Künzl, E. (1988). Der römische Triumph. Siegesfeiern im antiken Rom. Munich. Lamberti, F. (1993). Tabulae Irnitanae. Municipalità e ius Romanorum. Naples. Le Bourdellès, H. (1973). ‘La loi du foudroyé’. REL 51: 62–76. Linderski, J. (2000). ‘Penates (Di Penates)’. DNP 9: 514–16. MacMullen, R. (1981). Paganism in the Roman Empire. New Haven—London. Manacorda, D. (1996). ‘Iuppiter Fulgur, Aedificium’. LTUR 3: 136–8. Martin, H. G. (1983). ‘Architecture et politique. Le temple de Jupiter Capitolin’. Caesarodunum 18: 9–29. Martin, H. G. (1988). ‘Die Tempelkultbilder’, in M. Hofter et al. (eds.), Kaiser Augustus und die verlorene Republik. Mainz, 251–63. Meiggs, R. (1973). Roman Ostia. 2nd edn. Oxford. Mellor, R. (1992). ‘The local character of Roman imperial religion’. Athenaeum 80: 385–400. Momigliano, A. D. (1986). ‘The disadvantages of monotheism for a universal state’. ClPh 81: 285–97 (repr. in his On Pagans, Jews and Christians. Middletown [1987], 142–58). Mora, F. (1994). ‘Verso una tipologia del politeismo’, in U. Bianchi (ed.), The Notion of Religion in Comparative Research. Rome, 823–30.

Not the One nor the Many

159

Mora, F. (2000). ‘Verso una tipologia delle religioni classiche’. Kernos 13: 9–33. Morley, N. (1996). Metropolis and Hinterland. The City of Rome and the Italian Economy (200 ..–.. 200). Cambridge. Morley, N. (2003). ‘Migration and the metropolis’, in Edwards and Woolf, 2003, 147–57. Müller, K. O. (1825). Prolegomena zu einer wissenschaftlichen Mythologie. Göttingen. Noy, D. (2000). Foreigners at Rome: Citizens and Strangers. London. Obbink, D. (2002). ‘All gods are true’, in D. Frede and A. Laks (eds.), Traditions of Theology: Studies in Hellenistic Theology, Its Background and Aftermath. Leiden, 184–221. Orlin, E. M. (1997). Temples, Religion and Politics in the Roman Republic. Leiden. Parker, R. (2003). ‘The problem of the Greek cult epithet’. Opuscula Atheniensia 28: 173–83. Parker, R. (2005). Polytheism and Society at Athens. Oxford. Platner, S. B. and Th. Ashby (1929). A Topographical Dictionary of Ancient Rome. London. Reusser, Ch. (1993). Der Fidestempel auf dem Capitol in Rom und seine Ausstattung. Rome. Reusser, Ch. (1996). ‘Iuppiter Conservator’. LTUR 3: 131f. Rich, J. W. (1998). ‘Augustus’ Parthian honours: the temple of Mars Ultor and the arch in the Forum Romanum’. PBSR 56: 71–128. Rives, J. (1994). ‘Venus Genetrix outside Rome’. Phoenix 48: 294–306. Rives, J. (1995). Religion and Authority in Roman Carthage from Augustus to Constantine. Oxford. Rüpke, J. (1997). ‘Römische Religion und Reichsreligion. Begriffsgeschichtliche und methodische Bemerkungen’, in Cancik and Rüpke, 1997, 3–23. Rüpke, J. (2011 [2012]). ‘Lived ancient religion: questioning “cults” and “polis religion” ’. Mythos n.s. 5: 191–204. Scheid, J. and J. Svenbro (1997). ‘Le comparatisme, point de départ ou point d’arivée?’, in F. Boespflug and F. Dunant (eds.), Le comparatisme en histoire des religions. Paris, 295–312. Scheid, J., ed. (1998). Commentarii fratrum Arvalium qui supersunt. Rome. Scheid, J. (2001). Religion et piété à Rome. 2nd edn. Paris. Scheid, J. (2003). ‘Hierarchy and structure in Roman polytheism: Roman methods of conceiving action’, in C. Ando (ed.), Roman Religion. Edinburgh, 164–89 (originally published as ‘Hiérarchie et structure dans le polythéisme romain. Façons romaines de penser l’action’, ARG 1.2 [1999]: 184–203). Scheid, J. (2005). ‘Fremde Kulte in Rom: Nachbarn oder Feinde?’, in U. Riemer and P. Riemer (eds.), Xenophobie—Philoxenie. Vom Umgang mit Fremden in der Antike. Stuttgart, 225–40. Schilling, R. (1974). ‘Iuppiter fulgur. À propos de deux lois archaïques’, in Mélanges de philosophie, de littérature et d’histoire ancienne offerts à Pierre Boyancé. Rome, 681–9 (repr. in his Rites, cultes, dieux de Rome. Paris [1979], 358–65).

160

Not the One nor the Many

Schlesier, R. (1994). Mythen und Gelehrte. Anthropologie der Antike seit 1800. Frankfurt am Main. Schultz, C. E. (2000). ‘Modern prejudice and ancient praxis: female worship of Hercules in Rome’. ZPE 133: 291–7. Schulz, H. (1993). Stammesreligionen. Zur Kreativität des kulturellen Bewusstseins. Stuttgart—Berlin—Cologne. Schwemer, D. (2006). ‘Das hethitische Reichspantheon. Überlegungen zu Struktur und Genese’, in Kratz and Spieckermann, 2006, vol. 2, 241–65. Spannagel, M. (1999). Exemplaria Principis. Untersuchungen zu Entstehung und Ausstattung des Augustusforums. Heidelberg. Stark, R. and R. Finke (2000). Acts of Faith. Explaining the Human Side of Religion. Berkeley—Los Angeles—London. Stevenson, T. R. (1996). ‘Social and psychological interpretations of GraecoRoman religions. Some thoughts on the ideal benefactor’. Antichthon 30: 1–18. Suerbaum, W. (2002). ‘Q. Valerius aus Sora’, in W. Suerbaum (ed.), Handbuch der lateinischen Literatur der Antike I. Munich. Tatum, W. J. (1999). ‘Roman religion: fragments and further questions’, in S. N. Byrne and E. P. Cueva (eds.), Veritatis amicitiae causa. Essays in Honour of A. L. Matto and J. R. Clarke. Wauconda, 273–91. Thulin, C. (1905–9). Die etruskische Disziplin. Göteborg (repr. Darmstadt [1968]). Todd, M. (1985). ‘Forum and Capitolium in the early Empire’, in R. F. Grew and B. Hobley (eds.), Roman Urban Topography in Britain and the Western Empire. London, 56–66. Tosi, G. (1984). ‘La città antica e la religio nel de architectura di Vitruvio’, in Religione e città nel mondo antico. Atti (Centro ricerche e documentazione sull’ antiquità classica) V 11, NS 1. Rome, 425–39. Trigger, B. G. (2003). Understanding Early Civilisations. A Comparative Study. Cambridge. Turcan, R. (1998). Rome et ses dieux. Paris (tr. as The Gods of Ancient Rome. Religion in Everyday Life from Archaic to Imperial Times. Edinburgh [2000]). Vernant, J.-P. (1974). Mythe et société en Grèce ancienne. Paris. Versluys, M. J. (2002). Aegyptiaca Romana. Nilotic Scenes and the Roman Views of Egypt. Leiden. Versluys, M. J. (2004). ‘Isis Capitolina and the Egyptian cults in Late Republican Rome’, in L. Bricault (ed.), Isis en Occident. Leiden, 421–48. Versnel, H. S. (2000). ‘Thrice One: three Greek experiments in oneness’, in B. N. Porter (ed.), One God or Many? Concepts of Divinity in the Ancient World. Chebeague, 79–164. Viscogliosi, A. (1996). ‘Iuppiter Stator, Aedes ad Circum’. LTUR 3: 157–9. Vout, C. (2003). ‘Embracing Egypt’, in Edwards and Woolf, 2003, 177–202. Weber, G. (2003). ‘Augustus und die Träume’, in G. Weber and M. Zimmermann (eds.), Propaganda—Selbstdarstellung—Repräsentation im römischen Kaiserreich des 1. Jhs. n. Chr. Stuttgart, 297–316.

Not the One nor the Many

161

Whitehouse, R. D. (1992). Underground Religion. Cult and Culture in Prehistoric Italy. London. Wissowa, G. (1912). Religion und Kultus der Römer. 2nd edn. Munich. Woolf, G. D. (1997). ‘Polis-religion and its alternatives in the Roman provinces’, in Cancik and Rüpke, 1997, 71–84. Ziolkowski, A. (1992). The Temples of Mid-Republican Rome. Rome.

6 Action and Ritual in Roman Historians Or How Horatius Held the Door-Post John North

INTRODUCTION For the most part, the Roman historians of the Republic and early Principate do not provide us in their narratives with very satisfactory accounts of ritual activities. Just occasionally we get a detailed account of the triumph of a particular commander,1 but even in that case there is usually more emphasis on the display of spoils and prisoners than on the more specifically religious elements of the procession. Occasionally again, we are supplied with the formula for the taking of a state vow, or with an account of debate about the finer points of ritual procedure, what modern commentators have tended to call ‘legalistic’ points.2 On the whole, however, the business of historians in antiquity was to produce narratives telling what happened, explaining the causes of events or reflecting on the moral issues raised by events. They did not produce background information about procedures, laws, or social customs unless there was a definite need to do so. They might pause, for very specific reasons, to describe institutions, as Polybius does to describe the Roman political and military system or the 1 See e.g. the triumph of L. Aemilius Paullus (Plutarch, Aemilius Paullus 32–4; Livy’s account is mostly lost in the lacuna between 45. 29 and 30); on triumphs in general, Versnel 1970; Künzl 1988 and see 141–50 for the texts (in German) of the main accounts; Auliard 2001; Itgenshorst 2005; Beard 2007. 2 Livy 31. 9. 7–10; see Briscoe 1973–81: ad loc., 79–82. The debate here turned on whether a closed fund (lucar) had to be set up in advance for the fulfilling of the vow or whether the amount could be fixed when the time came for fulfilling the vow. The pontifex maximus stood out for the closed fund; the college when consulted formally over-ruled his view.

John North, Action and Ritual in Roman Historians: Or How Horatius Held the Door-Post In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0007

Action and Ritual in Roman Historians

163

development of the Achaean League;3 but for the most part, even constitutional arrangements tend to be assumed rather than explained at any length, unless there is a reform to be reported or a constitutional point that is essential to the narrative in progress. Ritual action very seldom comes into this category. It is not normally a determinative of action as historians in antiquity conceived it. If a decision to attack or not to attack was taken wholly by some method of divination, we would expect to be told of that; but on the whole we are not so told. There are exceptions, when a ritual event gets more attention than usual: such, for instance, is the case of the ver sacrum vowed in 216  and finally celebrated in 195/4 .4 But predominantly, what was recorded in the Livian tradition, and in later historians as well, was mention of, or reference to, a ritual taking place, not a description of the ritual procedure itself. So, to take the most familiar example, the regular prodigy-lists in the third, fourth, and fifth decades of Livy’s History mention very large numbers of recommended sacrifices, lustrations, supplications, and processions; but (with odd exceptions) there are few if any details of the ritual proceedings themselves.5 For Livy, of course, this is partly a matter of style: the prodigy lists have to be clipped, dry, and archaic in tone, the very markers of the annalistic manner. It is remarkable anyway that in these contexts Livy makes no attempt to elaborate, to use inventio or to tell anecdotes, as he might have chosen to do. The proprieties of Roman religious discourse limit the historian’s loquacity. The result, whatever the reason, is that we do not get much of a story.6 The most significant exceptions occur when rituals go seriously wrong—or are believed or alleged to have done so. This is most obvious in the case of military defeats, when the suggestion is sometimes made that a ritual procedure had failed or been omitted or never reached its final completion.7 In Livy Book 41 there is a series of such allegations,8

3 Polybius 6 passim; 2. 41–72; the Oxyrhynchus historian, here as elsewhere, provides an exception to the rule, e.g. on the Boeotian constitution, Hellenica Oxyrhynchia, ch. 11. 4 Livy 22. 10 = Beard et al. 1998: vol. 2, 6. 5 (for the vow in 217 ); 33. 44 (for the first celebration, 195 ); 34. 44 (for the repeat 194 ); Heurgon 1957; Beard et al. 1998: vol. 1. 80; Dench 1995: 185–6, 203–12. 5 Discussion in Wülker 1903; MacBain 1982; Levene 1993; Rosenberger 1998; North 2000; Rasmussen 2003; Davies 2004; Engels 2007. 6 For an index of Livy’s lists, see MacBain 1982: 82–106. 7 For the conception of litatio, see e.g. Livy 27. 23. 4 (per dies aliquot hostiae maiores sine litatione caesae); Plautus, Poenulus 489 ( . . . me Iuppiter faciat ut semper sacrificem nec umquam litem); Wissowa 1912, 418–9. The term perlitatio for sacrificing to completion seems to be modern, though the verb ‘perlitare’ is used. 8 Livy 41. 14. 7; 15. 1–4; 16. 1–7; 18. 8–16.

164

Action and Ritual in Roman Historians

alas in a defective section of the manuscript, with the result that the full sequence of events and its final outcome are lost. In what survives we have a number of references to mistakes and evidence of a disagreement; and perhaps of a constitutional crisis.9 There are also, of course, more famous occasions on which defeats were retrospectively connected with ignoring the auspices or with other vitia in ritual performance; but the details are seldom provided.10 One case known in a little detail is not in fact connected with defeat at all; Cicero tells the story11 of how Ti. Sempronius Gracchus, consul of 162  and the father of the tribunes, admitted (or perhaps rather claimed) that he had forgotten to take the auspices when he crossed the amnis Petronia12 when on his way into the Campus Martius to hold the elections. He thus compromised the consuls he had himself elected, who were both induced to resign. In cases of military success—and in Books 31–45 Livy is for the most part telling a story of military success—the historian has no real occasion to mention the details or niceties of ritual action, unless he wants to parade his expertise in this area. Mostly, the rituals can be assumed to have been successful, since the actions of which they formed an essential part had themselves a successful outcome. Notoriously, for instance, the Roman historians never give us any sustained account of the elaborate ritual of animal sacrifice.13 We have two such accounts in surviving literature: one is from Dionysius of Halicarnassus,14 because he wanted to prove that Roman sacrifice was imitated from the Greeks and thought that the ritual helped his case; the other is from Arnobius of Sicca, who wanted to attack pagan sacrifice on behalf of the Christians.15 The result is that we are routinely told that a sacrifice was to be made, but never in any detail how it was actually done. Modern ideas come from art and archaeology, from poets and antiquarians, but historians writing in antiquity have little to offer us. Nor is this at all surprising: they simply did not contemplate having readers who would be so ignorant as to need 9 Livy 41. 18. 16 is a fragment quoted by Priscian Institutio de arte grammatica 17. 29 and (plausibly) placed at this point in the text by Sigonius; it implies a constitutional conflict about whether a suffect consul could hold the consular elections; we do not know the outcome. 10 On defeated Roman generals, see Rosenstein 1990. 11 Cicero, De divinatione 1. 33; 36; 2. 74; Cicero, De natura deorum 2. 74; cf. Valerius Maximus 1. 1. 3; Plutarch, Marcellus 5. 1–3; Broughton 1951–84: vol. 1, 441–2. 12 On which see, Pease 1968: on Cicero, De natura deorum, cit. n. 11. 13 For which see Beard et al. 1998: vol. 2, ch. 6. 14 Dionysius of Halicarnassus (Roman Antiquities 7. 72. 15–18) claimed to have seen the ritual in Rome himself: see ibid. 18. 15 Arnobius, Adversus nationes 7; for recent discussion, Simmons 1995: 304–18; North 2007.

Action and Ritual in Roman Historians

165

telling how a sacrifice should be conducted. It would, therefore, be a mistake of the first order to infer from this situation that the historians were taking such matters lightly or without interest. We should never forget the delicacies of religious speech in the Roman tradition; it would have been vulgar and even dangerous to assume simple connections between ritual performance and the outcome of actions.16 The rituals I am referring to here are to be thought of as interactions between gods or goddesses and men or women, related to the narrative flow of the historian’s account, but not in themselves constituting part of the sequence of reported events. They might be said to be contingent presuppositions of the narrative: ‘of course’ the consul could not have left Rome for the campaign without taking the right vows, sacrifices and processions or without the auspices being taken at the appropriate moments; ‘of course’ you could not celebrate a triumph without the usual ritual sequence (procession, victim, prayer, sacrifice, etc.); in just the same way, you cannot launch a ship (at least not in the UK) without using a bottle of champagne; but if you were to check a history of shipbuilding, you would scarcely expect to find ‘champagne’ in the index. The fact of a valid law, or of a successful battle, or a ship on the ocean, all imply that the necessary ritual had occurred. If such things are not mentioned that is because they do not need to be specified. Some particular rituals, however, are not mere accompaniments to action in the sense I have defined, but themselves constitute the action.17 Such are the rituals that transform persons or places into some state different from their state before the ritual took place: So the ritual of marriage creates a married couple; and the ritual of consecration creates an altar or a temple where there was no altar or temple before. The text to be discussed here18 concerns precisely the act of dedication of a temple, but the attention it gets is not simply the result of the significance of the occasion, great though that was; here too the story becomes a story precisely because of what went wrong, not just of what went right.

THE CAPITOLINE DEDICATION The action in question was not the dedication of any ordinary temple. We are told that in the very first year of the new republican regime, the

16 17

For a relevant anecdote, see Douglas 1966: 73ff. 18 Austin 1970; Rappaport 1999: 107–17. Livy 2. 8.

166

Action and Ritual in Roman Historians

Romans dedicated their great triple temple on the Capitoline Hill to Iuppiter Optimus Maximus, to Iuno, and to Minerva. It would scarcely be possible to exaggerate the historic importance of this occasion. The Capitoline temple was by far the biggest and most dominant temple throughout the republican era.19 The three cults it housed were central to the religious life of the Republic and, in particular, Iuppiter was the god to whom Roman generals made their vows before they went out to war from Rome and to whom they returned in triumph.20 He was the patron of the auspices which were supposed to have guided every step in Roman history. As colonies were founded throughout Italy and overseas, it was the Capitoline structure in one form or another that came to symbolize the omnipresence of Rome and its gods and goddesses.21 It is an inference from the scale of the great platform on which the temple was built that it was already a massive presence in the late sixth century .22 The evidence of Livy and others can leave us in no doubt of its importance in later periods. It is perhaps paradoxical that the building of the supreme symbol of the Republic should be ascribed to the final monarch, otherwise represented in the tradition as a villain; but of course if the dedication ritual was going to be co-incident with the founding year of the Republic, then the building had to be finished and ready for dedication when the first consuls took office.23 Even more dignity and weight ought to be resting on the accounts of this essential ritual moment. Livy gives an account of what happened: Nondum dedicata erat in Capitolio Iovis aedes; Valerius Horatiusque consules sortiti uter dedi-caret. Horatio sorte evenit: Publicola ad Veientium bellum profectus. Aegrius quam dignum erat tulere Valeri necessarii dedicationem tam incliti templi Horatio dari. Id omnibus modis impedire conati, postquam alia frustra temptata erant, postem iam tenenti consuli foedum inter precationem deum nuntium incutiunt, mortuum eius filium esse, funestaque familia dedicare eum templum non posse. Non crediderit factum an tantum animo roboris fuerit, nec traditur certum nec interpretatio est facilis. Nihil aliud ad eum nuntium a proposito aversus quam ut cadaver efferri iuberet, tenens postem precationem peragit et dedicat templum. (Livy 2. 8) The temple of Iuppiter on the Capitol had not yet been dedicated; the consuls, Valerius (Publicola) and Horatius, drew lots as to which of them 19 Boethius 1978: 46–8; Colonna 1987: 64–6; Steinby (ed.) 1993–2000: III. 144–8 (G. Tagliamonte); Cornell 1995: 96 and n. 48; Beard et al. 1998: vol. 1, 59–60; vol. 2, 23–5. 20 For the Capitoline temple and the triumph, see Latte 1960: 149–54. 21 Beard et al. 1998: vol. 1, 334–7; vol. 2, 244–5. 22 So Tagliamonte, cit. n. 19; for an alternative view, Castagnoli 1984. 23 Livy 2. 8. 6 implies that the temple already existed but had not yet been dedicated.

Action and Ritual in Roman Historians

167

should perform the dedication. The lot fell to Horatius, so Publicola set off for the war against Veii. The supporters of Publicola took it harder than was proper that Horatius should be awarded the dedication of such a famous temple. In every way, they tried to impede him. They made some unsuccessful attempts; but then, when the consul was already holding the door-post, they sent, in the very course of the prayer to the gods, a terrible message that his son had died and that, the family being now in mourning, he could not perform the dedication. It has not been transmitted with certainty, and would not be easy to conjecture, whether he believed that this message was untrue or whether he had great strength of mind; but he in no way diverted his attention to the messenger and away from his task, except to give an order that the body should be carried out. He held on to the door-post, completed the prayer and dedicated the temple.

The story is surprising in many ways; but we should note as particularly in need of explanation: the rivalrous behaviour of the two consuls and their supporters; the credit at stake for the man who carried out the ceremony; the apparently dramatic violation of the ritual by the awful message brought in; the behaviour of the famous dedicator in holding on to the door-post through thick and thin. None of this is what might be expected of a narrative to be located at a sensitive moment at the very beginning of Livy’s account of the early Republic. The essential issue is of course not what really happened on this occasion, which I assume is beyond recovery—nec facilis interpretatio—but why Livy should have received and included this story and what is the historiographic significance of its presence here. It is already quite surprising that the republican tradition should accept so cheerfully that its greatest and, in religious terms, most significant monument should be owed to the efforts of the evil foreign tyrants they had expelled from the city. Granted that this was too deeply engrained a part of the story to be removed, and that the ritual dedication was the only part for which the new regime itself would ever be able to claim the credit, the least one would expect would be that it should have been carried out with exemplary ritual precision and success, more Romano.24 In fact everything goes wrong: first, the performance of the ritual is interrupted by the announcing of the bad news, so that the dedicator actually speaks to the messenger;25 secondly, the story as Livy reports it 24

For a discussion of the other accounts of the incident, see Peña 1981. Livy reports his words as: ut cadaver efferre iuberet, which ought to be the instruction for the funeral to take place (cf. e.g. Livy 2. 33. 11 for this sense of the verb). Perhaps, however, there is some reminiscence of the action of Sulla, as reported by Plutarch, Sulla 35. 2, who also persisted in the completion of a ritual, despite the death of his wife, through the device of divorcing her and having her removed from his house. But the circumstances are 25

168

Action and Ritual in Roman Historians

does not resolve the issue of whether the dedicator’s family was ‘funesta’ or not;26 thirdly, there seem to be ritual errors both in the decision as to who should conduct the ceremony and in the actual performance, to both of which we shall come later.27 Surely, even a not too well-informed reader of Livy’s text would be shocked by these serious objections to the success of the ritual. Why should an account with so many flaws have been accepted in the tradition? In order to put the mistakes in their right context, they should be compared with the balancing stories of rituals either abandoned or repeated when tiny errors were discovered in the performance of the rituals. If the hat of the flamen Dialis fell off his head during a ritual, if there was an error in the saying of the prayer, if the squeak of a mouse was heard at an inauspicious moment, a whole ceremony might have to be repeated.28 So we are told. Perhaps there is room for a little scepticism about the literal truth of these tales, which are recorded precisely to demonstrate the extreme religiosity of Roman public life in the time of the maiores. But this is an issue of ideology not of literal truth: Romans expected absolute precision in the performance; Livy’s account dramatically defies that expectation. How could it?

WHO HELD THE DOOR-POST? The most important ritual issue is the role of Horatius himself and I propose to concentrate first on that aspect of the story. In Livy’s account, Horatius is consul;29 chosen by lot for the task, he performs the dedication and it is he who holds on to the door-post. Indeed, the holding is the central point: it is because he hangs on through thick and thin that he can claim afterwards that the ritual was not compromised by the interruption. Valerius Maximus has a significantly different version of events:

quite different here, and the words seem simply to accept the fact of the son’s death and hence of his own status as polluted by it. They give the game away, whereas Sulla was preserving the letter of the law. 26

The claim that his family was funesta is attributed to his enemies; Livy expresses his own uncertainty. Plutarch (Poplicola 14. 8) states explicitly that the announcement was a lie. 27 See below, pp. 27–30. 28 Plutarch, Marcellus 4–5; Valerius Maximus 1. 1. 4–5. 29 For the various consuls of the first year of the Republic, Broughton 1951–84: vol. 1, 1–5; Hanell 1946; Cornell 1995: 218–23. Dionysius of Halicarnassus, Roman Antiquities 5. 35. 3 implies that the temple bore the name of Horatius, but does not say in what capacity.

Action and Ritual in Roman Historians

169

Horatius Pulvillus, cum in Capitolio Iovi Optimo Maximo aedem pontifex dedicaret interque nuncupationem sollemnium verborum postem tenens mortuum esse filium suum audisset, neque manum a poste removit, ne tanti templi dedicationem interrumperet, neque vultum a publica religione ad privatum dolorem deflexit, ne patris magis quam pontificis partes egisse videretur. (Valerius Maximus 5.10.1) When Horatius Pulvillus as pontifex was dedicating the temple of Iuppiter Optimus Maximus on the Capitol and heard, while actually performing the sacred formulae and holding the door-post, that his son had died, he did not move his hand from the door-post, in case he might interrupt the dedication of so great a temple, nor did he turn his face away from the public ritual towards his private grief, in case he might seem to be playing the role of father rather than that of pontifex.

It would be a mistake to assume that Valerius’ version here was radically different from Livy’s; he omits the whole tale of intrigue and the possibility that the son was not dead at all, but he may have done so deliberately, in order to pare the story down to the barest and most elegant minimum, not because he was unaware of these elements of the tradition. He is almost certainly wrong, as we shall see, in making Horatius both the dedicator and the pontifex: the performance of the dedication was indeed the function of a magistrate, not of a priest. But the essence of the story, and the point at which he differs from Livy’s version, is in saying that it was as pontifex not as consul that Horatius was holding the door-post. There are substantial reasons for believing that it was not the dedicator himself, but the pontifex in attendance who in fact held the door-post. Only one other passage might seem to offer support to Livy’s view.30 Servius,31 when commenting on a line of Virgil: in medio mihi Caesar erit templumque tenebit, connects the word ‘tenebit’ with the formula postem tenere: et verbo usus est pontificali; nam qui templum dicabat postem tenens dare se dicebat numini, quod ab illo necesse fuerat iam teneri et ab humano iure discedere. He is using a pontifical term: for the dedicator of the temple held the doorpost and said that he was handing the temple over to the deity, because it was essential that it should be possessed by the deity and cut off from human legality. 30

It has been suggested (e.g. in Broughton 1951–84: vol. 1, 3–4 and n. 5 and by Ogilvie 1965: 253) that Cicero, De domo sua 139, also implies that Horatius was acting as pontifex not as consul, but this is a misinterpretation of the text: Cicero is describing the ideal conditions for a dedication, for which purpose Coruncanius is the wisest of all pontifices, Horatius the strongest-minded of all dedicators. The distinction of role is, in fact, being carefully respected. 31 Servius, On Virgil, Georgics 3.16.

170

Action and Ritual in Roman Historians

There can be no doubt that Servius is at this point talking about the dedicator not about the attendant pontifex. He also offers a definite understanding of the ritual: the dedicator on this view holds the doorpost to symbolize the possession of the temple that he is giving away from the state to the deity. However, he seems quite wrong about the line of Virgil: the temple in question is not a literal one, but an allegory of his own poetical achievements which the poet dedicates to Augustus,32 who is invoked not as the dedicator, but as the god who will possess this temple and whose statue will stand in the centre of it. There must be suspicion that Servius’ account of the ritual derives from a misunderstanding of the line. In this case, we have in fact better evidence to follow up, whose implications have not been appreciated. The crucial evidence comes from Cicero’s speech De domo sua, the only surviving speech addressed directly to the pontifices, which concerns the alleged dedication of part of his own house. The situation was that Clodius had taken advantage of Cicero’s absence in exile to have his house destroyed; part of the ambulatio was then ritually dedicated to the goddess Libertas, whose principles Cicero was held to have violated by his actions as consul in 63 .33 When Cicero was restored from exile in 57 , he had to prove to the college of pontifices that this act of dedication had not been valid; otherwise, he would have been unable to resume possession of the house. The central point in dispute was whether Clodius had in fact performed an act of dedication valid in religious law or whether the ceremony, as carried out, had been in some way flawed. Cicero has a good deal to say about the pontifex who officiated at the ceremony: he was L. Pinarius Natta, the stepson of Licinius Murena whom Cicero had defended against a charge of corruption in the pro Murena;34 Pinarius had been co-opted as the youngest and newest member of the college, according to Cicero, just a few days before the dedication ceremony took place.35 Cicero was able to have a good deal of knockabout fun with his inexperience and incompetence, but there is no serious suggestion in the speech that the fact of his youth in any way invalidated the ritual performance. The speech as a whole therefore makes it unmistakeably clear that Clodius himself was the magistrate who performed the dedication

32

Mynors 1990: ad loc. For an account of the situation: Rawson 1975: 122–3; Mitchell 1991: 158–61; Beard et al. 1998: vol. 1, 114–15; on the speech itself: Nisbet 1939; Bergemann 1992. 34 For Pinarius as pontifex, Broughton 1951–84: vol. 2, 199, 213; Szemler 1972: pontifex no. 64; for the connection with Murena, cf. Cicero, Pro Murena 73; Münzer 1950: s.v. Pinarius no. 19. For his death in 56 , apparently, Cicero, ad Atticum 4. 8a. 3. 35 Cicero, De domo sua 118. 33

Action and Ritual in Roman Historians

171

ceremony, his young friend Pinarius Natta, the pontifex who was there to assist him. Most importantly of all, Cicero’s text includes what is quite certainly the formula of invitation from the presiding magistrate to the pontifex whom he needed to take part in the ritual.36 Admittedly the formula occurs in a parodied form, but the point of the joke must depend on Cicero’s use of the right formula as a basis: Ades, Luculle, ades, Servili, dum dedico domum Ciceronis, ut mihi praeeatis postemque teneatis. Attend, Lucullus; attend, Servilius, while I am dedicating Cicero’s house, so that you may dictate the formula for me and hold the door-post.

What Cicero is doing here is deriding what really happened at the dedication-ceremony by imagining for satirical purposes that Clodius had done the unimaginable and instead of picking the most junior of the college, on whom he could most easily lean, picked the two most senior and conservative figures, on whom he could not have leaned at all.37 The joke is further loaded by Cicero’s characteristically twisting the facts and implying (as he often does) that Clodius sought to dedicate his whole house, not a shrine of Libertas in part of the house.38 The more extreme the action, the more absurd the idea that these very senior pontifices would ever have agreed to endorse it. What is important for this argument is that the joke only has any point if the formula used is precisely the correct one. We can therefore be quite certain that when the dedicating magistrate summoned a pontifex to officiate, he said: Ades, (name of the pontifex), dum dedico (name of the temple), ut mihi praeatis postemque teneatis.

The pontifex so summoned had then two functions to perform in the ritual of dedication. First, to dictate the correct words to constitute the dedication;39 secondly, to hold the door-post during the ritual. The phrase comes up again more than once in the course of Cicero’s speech,

36

Cicero, De domo sua 133; the point was taken by Wissowa 19122, 394–5 and n. 7. For the two senior pontifices mentioned, P. Servilius Vatia Isauricus (RE Servilius no. 93) and M. Terentius Varro Lucullus (RE Licinius no. 169) see Broughton 1951–84: vol. 2, 206; Szemler 1972: pontifices nos. 44 and 45. They were the consuls of 79  and 73 , respectively. Note that Cicero for obvious reasons avoids mentioning the pontifex maximus of the day, away in Gaul at the time. 38 On the issues, see Bergemann 1992; Beard et al. 1998: vol. 1, 114–15; van Haeperen 2002: 251–2. 39 For the formula, verba praeire, see Varro, De lingua Latina 6.95; Pliny, Natural History 28; TLL vol. X, 2 fasc. IV, 595–6; Wissowa 1912: 394–5. The usage is often found in the Arval Acta; there is a list of instances at Scheid 1998, 403 s.v. praeeo. 37

172

Action and Ritual in Roman Historians

always with reference to the pontifex, never with reference to Clodius himself. For instance,40 when Cicero is playing up the fear that what had happened to him might later happen to anybody else, he conjures up the vision of a rascally pontifex turning up unannounced on anybody’s doorstep; would any house be duly consecrated: . . . si is postem tenuerit et aliquid dixerit . . . ( . . . if he held the door-post and spoke some words . . . ). Clodius himself is never credited with holding the post, though Cicero at one point does make him claim in so many words that the pontifex had in fact held it: . . . quoniam pontificem postem tenuisse dixisti ( . . . since you have stated that the pontifex held the post).41 Cicero has a certain amount of fun at the expense of the young Pinarius Natta and at his difficulties in carrying out his role and reciting the prayer formulae.42 But this teasing does not affect the point: however badly Natta may have performed it, the ritual demanded of him that he should do the holding. Nothing in the speech suggests that Clodius did so. It may have been ridiculous that so new a priest should have performed the ritual or at least Cicero thought it was an opportunity for derision. But it is evident that this inexperience was not an issue in itself threatening the validity of Clodius’ dedication. Cicero himself, the pontifices and the senate when it reached its decision later, all attached importance to other issues altogether.43 So far as the evidence of this speech goes, the pontifex held the post and the magistrate did not. It is also the best evidence we could ever ask for. In theory there are three possibilities to be considered: (a) that the priest held the post and the magistrate did not; (b) that the magistrate held the post and the priest did not; (c) that they both did. (b) is totally ruled out by the evidence of the De domo; (a) is strongly supported by the evidence of the De domo; (c) remains just a possibility, though nothing in the De domo supports it and there seems no reason why Cicero should be deliberately excluding any mention of Clodius’ having performed this particular ritual act, when he so signally emphasizes Natta’s performance 40

Cicero, De domo sua 119. Cicero, De domo sua 121. Cicero’s point here is not entirely clear, but he is quoting Clodius as emphasizing the holding of the post by the pontifex to show that the ‘sacredness’ cannot be moved, as it could in the case of a consecrated statue. Clodius will have been making the point that the shrine of Libertas had to stay where it was, in Cicero’s ambulatio. 42 Cicero, De domo sua 139–40. 43 For the basis of the decision of the case, see Cicero, ad Atticum 4. 2 (= 74 S.B.), which gives Cicero’s account of his speech (4), the decree of the pontifices (3), Clodius’ response at a contio (3), and the debate on the issue in the senate (4–5). Note that the pontifices’ ruling is not a decision on what should be done, or even on the facts of the case, but on the state of the sacred law. Facts and actions are the senate’s business, to be decided on the priests’ advice, but not under their control. See Bergemann 1992. 41

Action and Ritual in Roman Historians

173

of it. If in fact both priest and magistrate performed it, presumably they would have done so with different symbolic meanings: the magistrate as part of his performative actions, the priest as part of his demonstrative actions; not only, on this view, was he dictating what the magistrate was supposed to say, but also showing him what he should do. There is surely a far stronger view to be found if we concentrate on the meaning of the gesture itself. The magistrate’s role in the ceremony is to act as the representative of the Roman people in surrendering the land and the temple to the deity; he recites the form of words that constitute the handing over as well as the lex (statute) that defines the temple’s rights and duties for the future. The magistrate does the giving up, saying ‘do, dico, dedico’, a formula we find recorded in the inscriptions;44 the pontifex does the accepting by his symbolic gesture of holding the doorpost. The pontifex, on this view, plays a dual role: in relation to the magistrate, he acts as a guide; in relation to the deity, he acts as a representative or agent. In fact, we might say that, in taking over the property, he acts as if he were himself the god. John Scheid has shown45 that, in the case of certain Roman rituals, the priest or priestess performing them becomes the deity whose priest or priestess he or she is; this admittedly applies to flamines and Vestals, sometimes conceived as the most ancient of priesthoods; but, even if this perception of priesthood is to be thought of as archaic, the pontifical rituals may well derive from the same conceptions as the others. We do not actually know that the pontifex was any less ancient than the flamen. However this ritual should be understood, there is little room for doubt that it was the priest who held the door-post: if Livy’s story about Horatius Pulvillus really belongs to him, then it must have been as pontifex not as consul that he acted in the Capitoline dedication.

THE CHARACTER OF LIVY’S REPORT The flaws in Livy’s account, to which I am drawing attention, are not necessarily all caused by the same features of the tradition. So far as the error in procedure is concerned, that may well result from the story being 44 For this formula see e.g. the lex arae from Salona ILS, 4907, ll. 9–10: hisce legibus hisce regionibus sic, uti dixi, hanc tibi aram, Iuppiter optime maxime, do dico dedicoque uti sis volens propitius mihi collegisque meis, decurionibus, colonis incolis coloniae Martiae Iuliae Salonae, coniugibus liberisque nostris. For discussion, Palmer 1974, 57–78; Magdelain 1978. 45 Scheid 1986.

174

Action and Ritual in Roman Historians

strongly attached to the name Horatius in the tradition, originally as a pontifex; while Livy is following a version that made him a consul and the actual dedicator—that is, the man who generally received all the credit from posterity. The pontifex, on the other hand, is not usually remembered at all in our records of later dedications. The magistrate is the man who has a name and the honour of the ritual role;46 but to perform the role of the pontifex, any member of the college would do; on the theory just discussed above, that might be explained because his role is to be the god in the ritual, so he is not there as an individual agent at all, but only as a realization of the god. Another, just as puzzling, element in the story is the ruthless struggle to monopolize the credit, particularly by the supporters of Valerius. A not dissimilar story is to be found a few chapters later in Livy’s narrative: Certamen consulibus inciderat, uter dedicaret Mercurii aedem. Senatus a se rem ad populum reiecit; utri eorum dedicatio iussu populi data esset, eum praeesse annonae, mercatorum collegium instituere; sollemnia pro pontifice iussit suscipere. Populus dedicationem aedis dat M. Laetorio, primi pili centurioni, quod facile appareret non tam ad honorem eius cui curatio altior fastigio suo data esset factum quam ad consulum ignominiam. (Livy 2.27.4) At this time the consuls quarrelled as to which of them should dedicate the temple of Mercury. The senate referred the decision to the people, but decided that whichever of the two should be awarded the dedication by order of the people, the same man should take charge of the corn-supply, set up a guild of merchants and undertake the dedication in place of the pontifex. The people awarded the dedication of the temple to M. Laetorius, a senior centurion, evidently as a reproach to the consuls, rather than an honour to a man whose rank was so inappropriate for the task.

To some extent, this tale reflects the same set of ideas as does the story of Horatius and Valerius. In the same way, the privilege of dedicating the temple to Mercury is so important that it becomes the centre of a major political struggle.47 In this case, there seems to be no question of using the lot to settle the issue, but rather a reference to the senate, who in turn refer it on to the popular assembly. On the other hand, there is a similar confusion over the role in the ritual of the pontifex, since, as we have seen, the dedicator should be collaborating with him not replacing him, as the words pro pontifice seem to imply. But in this case the means of

46 47

For this role, see Orlin 1997, 162–89. For full discussion, Combet-Farnoux 1980: 7–53. On Mercury: Radke 1979: 213–14.

Action and Ritual in Roman Historians

175

resolving the dispute are quite different and the people’s resolution of the problem strikes modern commentators as totally anachronistic, an anticipation of the conditions of the late Republic.48 But the idea that this is a retrojection of later situations does not in this case answer all the problems. Struggles of this kind in the late Republic are about access to power, commands, and resources, not about who should carry out a ritual. Perhaps, one might think that this is a reflection of the belief, so strongly expressed by Varro and Cicero,49 that the generation of the late Republic had lost the religious traditions of their ancestors, who had once placed piety towards the gods and goddesses at the very centre of their public life. It would then be a kind of romantic fiction to attribute to early Roman heroes the competitiveness of late republican politicians, but transposed on to the level of piety, rather than that of avarice or thirst for power. This discussion has so far raised a number of problems about Livy’s account: 1. The ritual, so far from being consensual, is marked by bitter conflict and rivalry; 2. The narrative contains elements that must cast doubt in any reader’s mind on the ritual validity of the whole transaction; 3. The ritual procedure, even before the intervention occurs, is quite at variance with our other information about the sacred law. To these questions and doubts a number of responses seem possible. We might quite simply argue that the story is pure fiction and therefore the mistakes unsurprising; but of course, it is precisely the fact that it is a fiction that makes us expect the story to be unproblematic. Who would ever have composed a foundation story with so many flaws in it? The answer to that might be that it reflects the views of a radical opponent of religion; it demonstrates after all that the famous care and precision characteristic of the Roman religious tradition was in fact unnecessary, since such an irregular episode made no difference to the successful completion of the dedication ritual. To develop this point, this postulated anti-religionist might be arguing implicitly that moral issues are more important than ritual ones; that it mattered more to the gods that Horatius should have the self-control to place his public duty above his family affections, than that he should or should not have conducted the dedication according to the ritual rules.

48 49

For the history of the Laetorii: Combet-Farnoux 1980: 27–35. On whose views see Beard et al. 1998: vol. 1, 120–1.

176

Action and Ritual in Roman Historians

Behold, to obey is better than sacrifice and to hearken than the fat of rams.50 But, to say the least, this is not a typically Roman attitude and it is impossible to believe that this was in fact the idea that Livy was seeking to convey to his readers. We must look elsewhere to find parallels for the story.

MYTHICAL PARALLELS One particular feature of the Horatius story that we have been examining is its marginal status in the tradition as a whole; not merely that it occurs on the chronological margin between regal and republican Rome, but also on the conceptual margin between mythical time in the full sense and historical time as represented by the annual structures of republican life. The temple itself, of course, was also on the margins, because tradition was perfectly unambiguous in claiming that it was devised, planned and built by the very Etruscan kings characterized as tyrants and foreigners. It is only the story of its dedication that marks it off as suitable to be the most sacred place and the most famous symbol of the new republican order. If we now go backwards into the purely mythical realm, a close parallel to what we have been examining is at hand. After the famous taking of the auspices had picked Romulus as the founder, the very first event was the killing of Remus when he contemptuously jumped over the foundations of the new walls. Notoriously, the details of the story vary widely in the different extant versions, but the underlying structure remains: 1. The founding act was interrupted by rivalry and ill-feeling. 2. The death of a close family relative ought to have rendered the whole proceeding ill-omened and flawed. 3. The conductor of the foundation carries on and completes the action despite the interruption. 4. The incident, however unfortunate, does not seem to affect the final acceptability of the foundation ritual. Structurally, the two stories are composed of the same elements, rearranged for the different situations.51 50

1 Samuel XV. 22. For theorizing on the death of Remus: Wagenvoort 1956: 169–83; Ogilvie 1965: 54; Versnel 1970: 132–63; Bremmer and Horsfall 1987: 34–8; Wiseman 1995: 117–25; Bannon 1997: 158–73. 51

Action and Ritual in Roman Historians

177

Most of all, it is essential to note that both versions of the story carry the same essentially Roman message. The Roman founders, like later Roman heroes, put the interests of the Roman state above all else; they personify almost fascist attitudes. Personal emotion or grief, family attachments, private obligations must all be subordinated to the good of the city. So, if your son betrays Rome and you are the responsible official at the time, you have him put to death.52 If your brother shows contempt for the new foundation, you kill him yourself or support the killer. If your son dies in the middle of an important ritual transaction, you hang on to temple door-post unflinchingly. When Livy tells the story of Romulus, he does so very briefly but does not try to white-wash the event. Ovid in the Fasti53 has Romulus behave much like Livy’s Horatius: he suppresses his grief and holds back his tears until the building is finished, and only then gives way to personal emotion. There is of course variation as to whether Romulus did the killing himself, just as to whether Horatius’ son was really dead or not, but in neither case does that variation affect the underlying structure of the story.54 My second parallel is very different and concerns a different but related part of the episode, but it also concerns a mythical king of Rome. Numa was also a founder, but this time the religious founder of Rome.55 His crucial role was to have left in writing the books that contained all the rules and regulations through which the college of pontifices directed the religious life of the Romans.56 The great paradox of Numa is that the maker of the rules, the creator of the rituals that all later generations had to follow with the most scrupulous care, was himself a quite different order of figure. While the Romans get the rules from his writings, he gets them from his semi-divine concubine; the Romans carry out rituals of worship, Numa can coerce Iuppiter from the sky and make him disgorge information by trickery; he can even in some versions oppose animal sacrifice, on which the whole ritual system in fact rested.57 52 Livy 2. 3–5 is the classic account of a father not sparing his own children: for the story type as typically Roman, see Polybius 6. 54. 5. 53 5. 469–85. 54 For the many variants of the story, see the elegant account in Wiseman 1995: 43–88. 55 For the story of Numa: Livy 1. 18–21; Dionysius of Halicarnassus, Roman Antiquities 2. 58–76; Plutarch, Numa, passim; Cicero, De republica 2. 25–30. Grant 1971: 134–46; Glaser 1936, 1245–8: in RE XVII.1.1245–8, s.v. Numa Pompilius; Ogilvie 1965: 88–91; Beard et al. 1998: vol. 2, 4–7. 56 So Livy 1. 20. 5–6. 57 For Egeria, Livy 1. 19. 5; 21. 3; Plutarch, Numa 15: dealings with Iuppiter, Valerius Antias fgt. 6 P = Arnobius, Adversus Nationes 5.1; Ovid, Fasti 3. 285–360; rejection of sacrifices, explicit in Cassius Hemina fgt. 12 P = 15 Ch. = Pliny, Natural History 18. 7; but the idea of vegetarianism must be implicit in the Pythagorean connection.

178

Action and Ritual in Roman Historians

Of course, the founder must be perceived as standing outside the order he (or she) founds and therefore is able to do what his descendants cannot do. There is a most striking demonstration of this principle in the historical tradition itself. Numa’s successor King Tullus found, among the papers that Numa left behind at his death, the ritual by which Numa had successfully brought Iuppiter down to earth. Tullus, seeking a remedy for a destructive plague that was killing his people, decided to try Numa’s methods. The result was catastrophic: his house was struck by lightning and he and his sons were burned to death.58 The apparent moral, for Tullus but also for all later Romans, was that you should follow what the founder wrote, not what he did. The same theme arises, centuries later, in the incident when the books of King Numa were excavated on the Janiculan in Rome by farmers working in their fields.59 The books had been buried with him and the reaction of the authorities, in the form of the praetor of 181 , to their discovery was to confiscate them and have them publicly burned in the forum.60 The publicity of this action is important: they were not hidden away, or secretly destroyed, or handed over for preservation to the pontifices, but destroyed as openly as possible, un-read even by the praetor himself, who claimed only to have read the chapter-headings.61 Modern discussions have tended to assume automatically that the meaning of this action was that the books were treated as forgeries, and therefore dangerous attempts to hi-jack the authority of Numa.62 In some sense, forgeries they may well have been; but that is not stated in the sources and Pliny, who had studied the record carefully and supplies most of the information that has come down to us, discusses the issue precisely because it provides him with evidence for the early use of papyrus, which they would do only if they were genuine ancient documents.63 There seems to be no escaping the conclusion that the books were destroyed not as forgeries, but as genuine works of King Numa, dating from the time of his burial. 58

Livy 1. 31. 8; cf. Pliny, Natural History 28.14. Sources for this incident: Livy 40. 29; Plutarch, Numa 22. 2–5; Pliny, Natural History 13. 84–6; Pliny quotes Cassius Hemina (fgt. 40 Ch. = 37 P) and cites on various points Piso (fgt. 13 Ch. = 11 P), Tuditanus (fgt. 7 Ch. = 3P), and Valerius Antias (fgts. 8 and 15 P). For bibliography, see Gruen 1990: 164 n. 28 (though with a strongly polemical agenda); Beard et al. 1998: vol. 2. 11. 1. 60 For the praetor, Q. Petillius Spurinus (RE no. 4, 11), Broughton 1951–84: vol. 1, 384. 61 Livy loc. cit. n. 58. 62 Gruen (1990, 158–70) has a brilliantly ingenious theory of his own about the episode, based on the assumption that the papers were fakes and known to be fakes, which I cannot accept. 63 Pliny, Natural History 13. 84. 59

Action and Ritual in Roman Historians

179

A possible way out might be to argue that in the second century , Numa had not yet become a major figure in the historical tradition. The invention of the tradition that he was the religious founder would, on this hypothesis, be the work of the later annalists or possibly even of the Augustan period itself. Consequently, the second-century  Numa would have been a figure so obscure that his writings, even if genuine, could have been destroyed out-of-hand. But this hypothesis, even if possible in itself, cannot solve the problem here: any attempt to make sense of the controversy must imply that Numa was for some reason a desirable precedent; the whole point of arguing about the significance and fate of the books (whatever theory one has about that) must have been that it mattered to the two sides whether the books derived from Numa or not, and what they said or did not say. But if he was a nugatory figure in the tradition, why should it have mattered what he said? The view that he was a Pythagorean, however misguided, and the repeated refutations of that belief must in any case imply that he was an important figure and that his place in the tradition was bitterly contested.64 It looks very much as if both traditionalists and reformers had their own Numa and used him in their different ways.65 The second possible answer brings us full circle. Numa was known to have been a powerful, founding king of Rome, the possessor of strange powers that made him almost equal to a deity himself. But the books, while reflecting this power, were not seen as suitable for the ordinary run of mankind. Plutarch hints at this in his account in the Life of Numa when he says that the praetor judged the writings unsuitable for the many.66 He therefore had them burned not because they lacked religious power or possible revelations, but because they were too dangerous. So he was not attacking the memory of Numa, but relying on and reinforcing Numa’s own judgement; whatever Numa judged it safe and expedient for the Romans to know, he had put into writing and handed down to the pontifices to be used for the benefit of the Roman people during the whole of their history. Anything that he had not handed over, but had buried with him when he died, was evidently too dangerous to be preserved or even read. The implication of this seems to be that Roman religion should be seen, not as an attempt to explore ultimate truths or to seek to make contact with the supreme powers of the universe, but as a more limited 64 For this tradition see esp.: Ovid, Fasti 3. 153f.; Plutarch, Numa 8. 5–21; for discussion, Delatte 1936; Gagé 1955: 328–38; Gabba 1967; Ferrero 1955: 142–7; Skutsch 1985: 263–4; Gruen 1990: 158–62. 65 For very speculative, but intelligent, thoughts, see Gagé 1955: 307–47. 66 Plutarch, Numa 22. 5.

180

Action and Ritual in Roman Historians

system tailored to human limitations and originally set up by a superior religious authority. The praetor does not seek to reject or assess that authority, only to replicate it. Numa provides, as it were, the frame within which the role of Roman religion was defined. The story of Tullus provides the model: Numa was strong enough to play with dangerous powers; but, like Tullus himself, we mere mortals have to operate within more restrictive limits, or we would come to harm. It would be misleading to interpret this as just a manipulation of the ignorant masses by the sceptical elite, because the story emphasizes that neither magistrates, nor priests, nor senate actually read or made their own judgements, while even the praetor only saw the chapter-headings before recommending the destruction of the books. Through the paradigmatic role of Numa, the Romans implicitly accept that there are powers and possibilities lying outside their own religious boundaries, about which they know, but which they do not seek to explore; they thus show awareness that their system, of which they are so proud, did not exhaust religious truth, but was set up within a selfconscious frame.67 As we have seen, the story of Numa, so far from providing a model as to how Roman leaders were supposed to behave, is virtually a mirror-image of that model, telling you what Roman leaders should not do. The founder has to have religious powers and capacities outside the system that he himself is in the process of creating. The discovery of his books threatened to breach the boundaries fixed by the initial act of definition. To put it another way, what is raised here is the question of how you can have an originator of a ritual. Since rituals are by definition the repetition of actions inherited from numberless previous performances, there should never be an occasion on which a ritual was invented or first performed.68 If there was to be an inventor, he has necessarily to be outside the ritual order, powerful and therefore dangerous. Numa’s books should never have been found and, if they were found, they had to be destroyed—in public and un-read.

CONCLUSION The problems that arise from Livy’s account of the dedication of the Capitoline temple cannot be resolved by any simple account of the events themselves or of the tradition that handed the story down. Whatever can be said about these, it remains a puzzle why the story had not received 67

For a similar point cf. Beard 1989.

68

See Rappaport 1999: 107–38.

Action and Ritual in Roman Historians

181

the regularization that might have been expected long before Livy came to write his version of it. Part of the answer is, of course, that it was such a good story; part is provided by one of the parallels discussed above— Horatius, like Romulus, drove home the moral of putting loyalty to the state above all other emotions, even family piety. What more appropriate moment for emphasising this message than at the very beginning of the republican order. But Horatius, minor figure in the narrative though he is, shares also in the special character of the founding fathers of Rome: he is only responsible for a single ritual action, but it is the founding ritual of the supreme sacred place of the city. So, echoing the status of the great founders, he stands outside the normal rules that govern the religious activities of the Roman citizen in later ages. His strength is such that he can defy, not only the news of his son’s death, but the breaking of the rules of religio and still carry through without flaw the dedication of Iuppiter’s own temple. Such aberrations could never have been tolerated in later ages.

A FT E R W O R D This essay stretches widely in time from Rome’s founders (Romulus and Numa) via the dedicator of the Capitoline temple (the rather less well-known Horatius Pulvillus) down as far as Clodius, the attempted dedicator of a temple to Libertas within the site of Cicero’s house. The parts played by Clodius himself and by his young friend Pinarius Natta provide crucial information for us as to the significance of the ritual of dedication. A linking theme between the stories is provided by the occasions on which religious rituals go disastrously wrong, but are accepted as valid, none the less. Strangest of all is the foundation-ritual of the Capitoline temple, which stands throughout Roman history, despite the major errors and misfortunes of its initiation. Connoisseurs of Alan Cameron’s assessment (Cameron 2011: 567–626) of the limitations of Servius’ understanding of Virgil, as he sought to comment on the text, will not be very surprised to meet him here in an entertaining, if negative, context.

B I B L I OG R A P H Y Auliard, C. (2001). Victoires et triomphes à Rome: droit et réalités sous la République. Besançon. Austin, J. L. (1970). ‘Performative utterances’, in J. Urmson and G. Warnock (eds.), Philosophical Papers of J. L. Austin. 2nd edn. Oxford, 233–52. Bannon, C. J. (1997). The Brothers of Romulus: Fraternal pietas in Roman Law, Literature and Society. Princeton.

182

Action and Ritual in Roman Historians

Beard, M. (1989). ‘Acca Larentia gains a son: myths and priesthood at Rome’, in M. M. Mackenzie and C. Roueché (eds.), Images of Authority. Cambridge Phil. Soc. Suppl. 16. Cambridge, 41–61. Beard, M. (2007). The Roman Triumph. Cambridge, MA—London. Beard, M. et al. (1998). Religions of Rome. 2 vols. Cambridge. Bergemann, C. (1992). Politik und Religion im spätrepublikanischen Rom. Palingenesia 38. Stuttgart. Boëthius, A. (1978). Etruscan and Early Roman Architecture. Harmondsworth. Bremmer, J. and N. Horsfall (1987). Roman Myth and Mythography. BICS Supplement 52. London. Briscoe, J. (1973–81). A Commentary on Livy. Oxford. Broughton, T. R. S. (1951–84). The Magistrates of the Roman Republic. 3 vols. Chico, CA. Cameron, A. D. E. (2011). The Last Pagans of Rome. Oxford. Castagnoli, F. (1984). ‘Il tempio romano. Questioni di terminologia e di tipologia’. PBSR 52: 3–20. Colonna, G. (1987). ‘Etruria e Lazio nell’età dei Tarquini’, in Etruria e Lazio arcaico. Atti dell’incontro di studio (10–11 novembre 1986) a cura di Cristofani Mauro, (Quad. del Centro di studio per l’archeol. etrusco-italica 15). Rome, 55–66. Combet-Farnoux, B. (1980). Mercure romain: le culte public de Mercure et la function mercantile à Rome de la république archaïque à l’époque augustéenne. BÉFAR 238. Rome. Cornell, T. J. (1995). The Beginnings of Rome: Italy and Rome from the Bronze Age to the Punic Wars (c.1000–264 ). London. Davies, J. P. (2004). Rome’s Religious History: Livy, Tacitus and Ammianus on Their Gods. Cambridge. Delatte, A. (1936). ‘Les doctrines pythagoriciennes des livres de Numa’. BAB 22: 19–40. Dench, E. (1995). From Barbarians to New Men: Greek, Roman and Modern Perceptions of the Peoples of the Central Apennines. Oxford. Douglas, M. (1966). Purity and Danger: An Analysis of Concepts of Pollution and Taboo. Harmondsworth. Engels, D. (2007). Das römische Vorzeichenwesen (753–27 v. Chr.): Quellen, Terminologie, Kommentar, historische Entwicklung. PaWB 22. Stuttgart. Ferrero L. (1955). Storia del pitagorismo nel mondo romano: dalle origini alla fine della repubblica. Turin. Gabba, E. (1967). ‘Considerazioni sulla tradizione letteraria sulle origini della Repubblica’, in Les origines de la République romaine. Entretiens sur l’antiquité class. 13. Vandoevres—Geneva, 133–74. Gabba, E. (1991). Dionysius and the History of Archaic Rome. Berkeley—Los Angeles—Oxford. Gagé, J. (1955). s.v. romain. Essai sur le culte d’Apollon et le développement du ‘ritus Graecus’ à Rome des origines à Auguste. BÉFAR 182. Paris. Glaser, K. (1936). RE s.v. Numa Pompilius, 1242–52. Grant, M. (1971). Roman Myths. Harmondsworth.

Action and Ritual in Roman Historians

183

Gruen, E. S. (1990). Studies in Greek Culture and Roman Policy. Leiden—New York—Copenhagen—Cologne. Haeperen, F. van (2002). Le collège pontifical. Brussels—Rome. Hanell, K. (1946). Das Altrömische Eponyme Amt. Lund. Heurgon, J. (1957). Trois études sur le ‘ver sacrum’. Collection Latomus 26. Brussels. Itgenshorst, T. (2005). Tota illa pompa: der Triumph in der römischen Republik. Hypomnemata 161. Göttingen. Köves-Zulauf, T. (1972). Reden und Schweigen. Munich. Künzl, E. (1988). Das römische Triumph: Siegesfeiern im antiken Rom. Munich. Latte, K. (1960). Römische Religionsgeschichte. Munich. Levene, D. S. (1993). Religion in Livy. Mnemosyne Suppl. 127. Leiden. MacBain, B. (1982). Prodigy and Expiation: A Study in Religion and Politics in Republican Rome. Collection Latomus 177. Brussels. Magdelain, A. (1978). La Loi à Rome: Histoire d’un concept. Collection d’études latines, fasc. XXXIV. Paris. Mitchell, T. N. (1991). Cicero the Senior Statesman. New Haven. Münzer, F. (1950). ‘Pinarius no. 19’. RE 20. 2, cols. 1402–3. Mynors, R. A. B. (1990). Virgil’s Georgics. Oxford. Nisbet, R. G., ed. (1939). Cicero De domo sua. Oxford. North, J. A. (2000). ‘Prophet and text in the third century ’, in E. Bispham and C. Smith (eds.), Religion in Republican Rome and Italy: Evidence and Experience. Edinburgh. North, J. A. (2007). ‘Arnobius on sacrifice’, in J. Drinkwater and R. Salway (eds.), Wolf Liebeschuetz Reflected. BICS Supplement 91. London, 27–36. Ogilvie, R. M. (1965). A Commentary on Livy Books I–V. Oxford. Orlin, E. (1997). Temples, Religion and Politics in the Roman Republic. Leiden— New York. Palmer, R. E. A. (1974). Roman Religion and Roman Empire. Philadelphia. Pease, A. S. (1968). M. Tulli Ciceronis De natura deorum libri III. Darmstadt (reprint). Peña, M. J. (1981). ‘La dedición yel dedicante del templo de Jupiter Capitolino’, Faventia 3: 149–70. Radke, G. (1979). Die Götter Altitaliens. Fontes et Commentationes 3. Münster. Rappaport, R. (1999). Ritual and Religion in the Making of Humanity. Cambridge Studies in Social and Cultural Anthropology 110. Cambridge. Rasmussen, S. W. (2003). Public Portents in Republican Rome. ARID Suppl. 34. Rome. Rawson, E. (1975). Cicero: A Portrait. London. Rosenberger, V. (1998). Gezähmte Götter: das Prodigienwesen der römischen Republik. Stuttgart. Rosenstein, N. (1990). Imperatores Victi: Military Defeat and Aristocratic Competition in the Middle and Late Republic. Berkeley—Los Angeles—Oxford. Ryberg, I. S. (1955). Rites of the State Religion in Roman Art. MAAR 23. New Haven.

184

Action and Ritual in Roman Historians

Scheid, J. (1986). ‘Le flamine de Jupiter, les Vestales at le général triomphant’, in C. Malamud and J.-P. Vernant (eds.), Corps des dieux. Le temps de la réflexion 7. Paris, 213–30. Scheid, J., ed. (1998). Commentarii Fratrum Arvalium qui supersunt. Roma Antica 4. Rome. Simmons, M. B. (1995). Arnobius of Sicca: Religious Conflict and Competition in the Age of Diocletian. Oxford. Skutsch, O. (1985). The Annals of Quintus Ennius. Oxford. Steinby, E. M., ed. (1993–2000). Lexicon Topographicum Urbis Romae. Rome. Szemler, G. J. (1972). The Priests of the Roman Republic: A Study of Interactions between Priesthoods and Magistracies. Collection Latomus 127. Brussels. Versnel, H. S. (1970). Triumphus. Leiden. Wagenvoort, H. (1956). Studies in Roman Literature, Culture and Religion. Leiden. Wiseman, P. (1995). Remus: A Roman Myth. Cambridge. Wissowa, G. (19122). Religion und Kultus der Römer. Munich. Wülker, L. (1903). Die geschichtliche Entwicklung des Prodigienwesens bei den Römern: Studien zur Geschichte und Überlieferung der Staatsprodigien. Leipzig.

7 Rites and Practices of Warfare in Italy between Romans and Samnites Going under the Yoke and the Samnite Legio Linteata Olivier de Cazanove

In this chapter we shall discuss rites, but also, more broadly, certain practices and procedures surrounding the practice of war in Italy in the fifth to third centuries , that is, at the time when Rome was gradually coming into conflict with many of the peoples of the peninsula. Even more than before, war necessarily became a common language, a common culture, with widely shared codes and techniques. This was true initially of the Italic peoples on the frontier of Latium: the Sabines, the Aequi, and the Volsci in the first place, then the Samnites, after the first treaty with Rome in 343 , then the first Samnite War in 326, and the third in 298.1 I should immediately make it clear that I shall adopt the traditional chronology, (i.e.) Livy’s, without discussing it. Even if it raises problems of detail and contains some unlikely elements, all in all it is less artificial than the arbitrary reconstructions suggested by the moderns.2 When the Romans and Samnites clashed once more after the Naples affair of 327–326—just before the beginning of the second Samnite War—the Samnites suggested consulting ‘Mars, common to both adversaries’, i.e. ‘Mars communis’ or ‘communis Mars belli’. This expression, 1 On the Samnites and the Samnite Wars, the fundamental work is still that of Salmon (1967). The excellent manual of G. Tagliamonte (20062) comments very usefully on recent debates. See besides, from among a vast bibliography, the collective works in Studi sull’Italia dei Sanniti 2000; Briquel and Thuillier 2001; Caiazza 2004; Jones 2004. 2 See esp. e.g. Sordi 1960; Sordi 1965.

Olivier de Cazanove, Rites and Practices of Warfare in Italy between Romans and Samnites: Going under the Yoke and the Samnite Legio Linteata In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0008

186

Rites and Practices of Warfare in Italy

frequently found in Cicero, Tacitus, and above all in Livy3 should not be over-interpreted. It simply means that the fortunes of war are shared more or less equally between the warring nations. The shared gods, as Servius explains in his commentary on Virgil,4 are Mars, Bellona, and Victoria, because they can favour either adversary. In the Aeneid,5 we do indeed see Rutuli and Trojans preparing the lists where Turnus and Aeneas are to fight, and placing temporary altars to the ‘shared gods’ on the grass (dis communibus aras gramineas); a young pig and a 2-yearold sheep were sacrificed on them, and Aeneas and Turnus both pray there to ‘the same gods’—the Sun and the Moon, Jupiter and Juno, the fountains and rivers. This mythical confrontation foreshadows the Roman wars against the peoples of Italy. Another element in this shared language of war is the weaponry. Not only because the colours of the clothing, the lustre of the metal in the armour and the plumes on the helmets are meaningful, but also, on a more practical note, because items of weaponry were liable to be borrowed from the adversary and copied from him, if they had proved their worth. The question of the borrowing of Samnite arms by the Romans has already been widely discussed, as has the problem of the appearance of the Samnite armies, in a volume edited by A.-M. Adam and A. Rouveret, Guerre et societies en Italie aux Ve et IVe s. av. J.-C., a book that is twenty years old (1986) but has lost nothing of its relevance.6 It is therefore pointless to return to this subject, except perhaps in order to point out that the recent publication of M. Humm’s thesis puts forward new arguments for dating the manipular reform and the tactical adjustments that went with it, including matters concerning weaponry, to the censorship of 312.7 For my part, I shall simply concentrate on two episodes that demonstrate that the rites and practices of war for both Romans and Samnites 3 Livy 5. 12. 1; 7. 8. 1; 8. 11. 6; 8. 23; 8. 31. 5; 10. 28. 1; 28. 19. 11; 28. 41. 14; 30. 30. 20; 37. 45. 13; 42. 14. 4; 42. 49. 5; Cicero, ad familiares 6. 4. 1; II Verrine. 5. 132; pro Sestio 12; pro Milone 56; Philippic 10. 20; De Oratore 3. 167; Seneca, De Ira 1. 12. 5; Tacitus, Histories 4. 64. 4. 4 Servius, ad Aeneid 12.118 (Thilo, 1884): ET DIS COMMUNIBUS ARAS Di communes sunt, ut alii dicunt, Mars Bellona Victoria, quod hi in bello utrique parti possunt favere. (‘The di communes are, as others tell us, Mars, Bellona, Victoria, because it is these who can give their favour to either side.’) Servius continues with other exegeses of the same expression: see Servius, ad Aeneid 3. 35; 12. 714. 5 Virgil, Aeneid 12.116–19: campum ad certamen magnae sub moenibus urbis / dimensi Rutulique viri Teucrique parabant / in medioque focos et dis communibus aras / gramineas. (‘Beneath the walls of the great city, after measuring the field for the conflict, the Rutulians and Trojans were preparing fires and grassy altars for their common gods.’) 6 7 Briquel 1986. Humm 2005: 268–308. See also Fraccaro 1975: 41–2.

Rites and Practices of Warfare in Italy

187

were similar or even the same. This similarity, however, does not carry the same significance in both cases. In 321, the notorious episode of the Caudine Forks, which I shall discuss first, shows that the adversaries actually had a shared custom, passing under the yoke, used by either of them depending on the vicissitudes of war. In 293 the recruiting of the Samnite Legio Linteata takes place in a setting so Roman that we may well ask whether it is an antiquarian-type reinvention by Livy or his annalistic sources.

SENDING UNDER THE YOKE First, a clarification, but one that goes almost without saying: the wellattested practice in Italy of mittere sub jugum has of course nothing to do with the Punic, and apparently more broadly Semitic, ritual of intrare sub jugum, under the yoke this time in the case of a divinity. On this last point, I shall simply refer the reader to a specific bibliography.8 The defeat at the Caudine Forks, in or at the entry of the Caudine Valley in 320 , during the second Samnite War,9 is still considered by both ancients and moderns to be one of the greatest Roman military disasters, comparable to the Cremera, the Allia, and later disasters at Heracleia, Cannae, Carrhae, and the Teutoburg Forest. Yet this reversal of fortunes was much less costly in terms of human lives than the disasters that occurred some years later, such as those at Lautulae10 or the Aornian Woods.11 The Roman defeat was mainly symbolic and can be summed up in the shame of being sent under the yoke. For the ancients, the defeat remained notorious precisely because of this unheard-of humiliation. For the moderns one event was completely subsumed into the other. The ultimate going under the yoke was at the Caudine Forks. Consequently, this practice seems to us to be principally Samnite and Italic. Now, an examination of the sources shows beyond any doubt that this rite was above all Roman (and only secondarily Latin, used once by the Tusculans). 8

Nock 1926; Berthier and Tayeb 1970; Bénichou-Safar 1993. Livy 9. 2–11; Appian, Samnite Wars 5–11; on the identification of the site, see de Sanctis 1907: 309–10 (= ed. 1960, 293–5); Salmon 1967: 225–6; Horsfall 1982. 10 At the strategic trap of Lautulae, near Terracina, the defeat was accompanied in 315  by the death of Q. Aulius Cerretanus, the commander of the cavalry, but also by an uprising of the Ausonians and even by the beginnings of a defection at Capua: Livy 9. 23. 4–5; Diodorus Siculus 19. 72. 5–8. 11 In 311, the Samnites massacred a Roman army in the ‘Aornian Woods’: Dio Cassius, in Zonaras 8. 1; the opposite view in Livy 9. 31. 6ff. Cf. Libourel 1973; Briquel 2001. 9

188

Rites and Practices of Warfare in Italy

Going under the yoke is mentioned on seven occasions by Livy (sometimes repeated, rather than confirmed by other sources). In 459, the Latins of Tusculum made the Aequi (who had had their citadel seized and laid low by famine) go under the yoke unarmed and naked, inermes et nudi. They fled in this state, endeavouring to go home, but the consul Q. Fabius Vibulanus intercepted them on the Algidus and massacred them all.12 In 454, Cincinnatus, called the dictator, sent the Aequi under the yoke after cornering them in their retrenchment on the Algidus: (The Aequi) switched from weapons to prayers and begged the dictator as well as the consul not to make the victory into a massacre, but to disarm them and let them go. The consul sent them back to the dictator who, in his anger, wanted to dishonour them further. He told them to bring their general Gracchus Cloelius and the other senior officers in chains to him and to hand over to him the town of Corbio: ‘he had no need of the blood of the Aequi: they could go. But in order to wrest from them the admission of their defeat and submission they would not leave without going under the yoke’. A yoke was made with three lances, two planted in the earth and one attached horizontally above. It was under this yoke that the dictator sent the Aequi.13

In 438, the Volsci besieging Ardea were themselves encircled and in their turn went under the yoke. The consul M. Geganius then let them go, each man dressed in a simple garment and unarmed, but when their disbanded troops halted not far from Tusculum, the Tusculans massacred them, motivated by an old hatred (vetere odio).14

12 Livy 3. 23. 5: inermes nudique omnes sub iugum ab Tusculanis missi. Hos ignominiosa fuga se recipientis Romanus consul in Algido consecutus ad unum omnes occidit. (‘All of them, disarmed and naked, were sent under the yoke by the men of Tusculum. When they were retreating in ignominious flight, the Roman consul pursued them on to Mount Algidus and massacred them to a man.’) 13 Livy 3. 28. 9–11: Tum, ancipiti malo urgente, a proelio ad preces versi hinc dictatorem, hinc consulem orare ‘ne in occidione victoriam ponerent, ut inermes se inde abire sinerent’. Ab consule ad dictatorem ire iussi; is ignominiam infensus addidit: Gracchum Cloelium ducem principesque alios vinctos ad se adduci iubet, oppido Corbione decedi: ‘sanguinis se Aequorum non egere; licere abire. Sed, ut exprimatur tandem confessio subactam domitamque esse gentem, sub iugum abituros’. Tribus hastis iugum fit, humi fixis duabus superque eas transversa una deligata. Sub hoc iugo dictator Aequos misit (trans. G. Baillet, CUF 1954). Cf. Florus 10. 40; Valerius Maximus 2. 7. 7. For the location of Mons Algidus by the via Latina on the crossing of the Alban Hills, see de Sanctis 1907: 112. 14 Volsci . . . cum ab omni parte caederentur, ad preces a certamine versi, dedito imperatore traditique armis sub iugum missi, cum singulis vestimentis ignominiae cladisque pleni dimittuntur; et cum haud procul urbe Tusculo consedissent, vetere Tusculanorum odio

Rites and Practices of Warfare in Italy

189

It is likely that, as well as these three pieces of evidence, other episodes occurred that we do not know about. In 441, another Quinctius, the consul T. Quinctius Capitolinus, spoke of the enemies of Rome—the context shows that these were the Aequi and the Volsci—‘so many times put to flight and scattered, forced to abandon their camps, sent under the yoke’.15 We cannot however be sure that this is not a rhetorical exaggeration. In any case, after these pieces of evidence, very close in time and space, distributed over about twenty years and within a radius of about 35 km to the south-south-east of Rome—we could even say that the hub Tusculum-Algidus is the focus in every case—after that, nothing happens for over a century. In fact, the next case of going under the yoke is nothing other than the Caudine Forks in 321 . The Romans wanted to open up a way into Apulia, to support their ‘good and loyal allies’ in Luceria. The consuls chose the shortest route (the future Via Appia) and found themselves trapped with their legions in the valle Caudina. The Samnites nevertheless agreed to let them go, safe and sound, but only after they had solemnly acknowledged their defeat, both by going half-naked and unarmed under the yoke, exposed to the derision of the Samnite warriors,16 and

inermes oppressi dederunt poenas vix nuntiis caedis relictis . . . . Consul triumphans in Urbem redit, Cluilio duce Volscorum ante currum ducto praelatisque spoliis quibus dearmatum exercitum hostium sub iugum miserat. (‘The Volsci, being slaughtered on every side, turned from fighting to praying and after surrendering their general and giving up their arms, they were sent under the yoke and then, dressed in a single vestment and full of the disgrace of the disaster, dismissed; when they rested not far away from the city of Tusculum and because the Tusculans had an ancient grievance against them, they were attacked while unarmed, so that there was scarcely a survivor to tell the story . . . . The consul returned in triumph. Cluilius, leader of the Volsci was led before his chariot and the spoils from the enemy’s army paraded as a show—the spoils, of which they had been stripped, when sent under the yoke.’) 15 Livy 3. 67. 5: quippe quotiens fusi fugatique, castris exuti, agro multati, sub iugum missi, et se et vos novere. 16 Livy 9. 4–8: Pontius replies to the Roman envoys: ‘since, even when defeated and captured, they do not know how to admit their ill-fortune, he would send them under the yoke disarmed and in a single vestment’ . . . . The conditions are accepted: ‘The consuls are first sent under the yoke, half-naked; then as each man is the next in rank, so he is exposed to ignominy; then the individual legions in succession. Armed enemies stood around, accusing and jeering: many Romans had swords turned on them and some were wounded and killed, if their expression, bolder than the situation justified, had given offence to a victor.’ This spectacular evocation of the scene, this real choreography of the passage through the yoke, has been studied by Kissel 1997: 501–7. He compares it to other ritualized forms of the closure of military operations and the humiliation of the defeated, in the field but also back home, as in the case of the triumph itself.

190

Rites and Practices of Warfare in Italy

also by demanding from the consuls a formal peace treaty, guaranteed by hostages.17 By the following year, according to tradition, thus in 320 (but this tight chronology is eminently suspect here), the Romans apparently took their revenge by besieging Luceria, which was held by the Samnites and reducing them to famine by blockading the surrounding countryside of Daunia. The 7,000 Samnites in the garrison were eventually able to leave, safe and sound, but first had to go under the yoke, wearing nothing but a simple garment. It was even said that this ignominious treatment was also meted out to Pontius, the son of Herennius, the victor at the Caudine Forks.18 The complete reversal of the situation from one year to the next, and the similarity between the episodes of Caudium and of Luceria cannot of course be credited in the terms in which Livy reports them. But the important thing for our purposes is what was at stake, the symbolic charge represented by the act of sub jugum mittere, sending under the yoke. Its historicity may well be open to doubt in the precise context of the year 320, yet its importance and consequence in the general context of Italic-Roman relations are none the less undeniable.

17 Contemporary opinion, shared by the annalist Claudius Quadrigarius, regarded this deal as a treaty (foedus). Livy (8. 29. 11) prefers to think of it as a promise (sponsio). By doing so, he gives it to be understood that the Roman people was in no way committed to the pax Caudina and was free to start the war again whenever it wished, on condition that they returned to the enemy the guarantees of the pact. The Livian account even achieves the transformation of the beaten and dishonoured consul into a hero, since it is Sp. Postumius himself who asked to be handed over to the Samnites—who, what’s more, did not accept him. It has often been emphasized, not unreasonably, that all this was only an anticipation of an episode of 136 , the release of the consul Mancinus, after an infamous surrender to the Numantines (who, what’s more, did not accept him either). De Sanctis 1907: 312–13 (= 1960, 297–8); Briquel and Brizzi 2000: 270–3, who follow the view of Magdelain 1943. 18 Livy 9. 15. 6–8: Papirius Cursor replies to the Samnite envoys that: militem se cum singulis vestimentis sub iugum missurum. (‘He was going to send the soldiers under the yoke, wearing a single garment.’) The conditions are accepted: septem milia militum sub iugum missa praedaque ingens Luceriae capta, receptis omnibus signis armisque quae ad Caudium amissa erant . . . . Haud ferme alia mutatione subita rerum clarior victoria populi Romani est, si quidem etiam, quod quibusdam in annalibus invenio, Ponti Herenni filius, Samnitium imperator, ut expiaret consulum ignominiam, sub iugum cum ceteris est missus.’ (‘Seven thousand soldiers are sent under the yoke; the huge booty from Luceria is captured and all the standards lost at Caudium are recovered . . . . The victory of the Roman people shines more brilliantly than almost any other sudden reversal of fortune, if it is in fact true, as I find in the annals, that the Samnite general, son of Pontius Herennius, so as to expiate the consuls’ humiliation, was himself sent under the yoke together with the others.’)

Rites and Practices of Warfare in Italy

191

In 307 , the Samnites were defeated at Allifae. They capitulated in their camp, going under the yoke, while their allies were sold at auction.19 In 294 , there was another battle, at Luceria. After the Roman victory, gained with difficulty, 7,800 Samnites went naked under the yoke.20 But this episode introduces irregularities into what might be called the canonical pattern of sub jugum mittere. Usually, one of the adversaries was blockaded—in a stronghold, a camp, or a natural trap— and it was to get out of this that they consented to be sent under the yoke. The two sides made a pact to resolve the situation in this way. The defeated army was humiliated, but could then leave without being taken prisoner.21 In 294, at Luceria, on the contrary, it was a pitched battle. 4,800 Samnites were killed, 7,800 were captured and it was the captives who were made to go under the yoke. The losers’ free will was reduced to nothing, there was no pact between the two opposing sides. This was a total perversion of the rules of war and the senate was not fooled. The consul M. Atilius was refused the triumph, amongst other things, ‘for sending prisoners under the yoke without their having agreed by making a pact’ (quod captivos sine pactione sub jugum misisset).22 We may be tempted to think that such a lack of respect for the rules shows that the custom was losing its significance. Indeed, the sending under the yoke of the conquered Samnites in 294 is the last time such a practice is mentioned by Livy. But it is important for sure not to forget that Livy’s first decade breaks off at the following year. The fact remains that no such incident is recorded in the Periochae, nor in books XXI–XXV of the ab Urbe condita, nor in parallel sources. Livy 9. 42. 8: Postero die vixdum luce certa deditio fieri coepta et pacti qui Samnitium forent ut cum singulis vestimentis emitterentur, ii omnes sub iugum missi. Socii Samnitium nihil cautum; ad septem milia sub corona veniere. (‘Next day when it was scarcely light the surrender began to be made and also the conditions about the Samnites, that they should be sent out, wearing a single garment; they were all sent under the yoke. there was no care for the allies of the Samnites; about seven thousand were auctioned off.’) 20 Livy 10. 36. 14: captivorum numerus fuit septem milium octingentorum, qui omnes nudi sub iugum missi. (The number of captives was seven thousand eight hundred, all sent under the yoke naked.) 21 Rüpke (1990: 211–12) quite rightly refutes the current opinion which turns the missio sub iugum into a ‘ritual of cathartic purification’. He is exactly right to refuse to see the yoke as a ‘magic door’ and also refuses to accept its merging with the venditio sub hasta, the sale of prisoners of war ‘under the spear’, precisely because the defeated, when forced to pass under the yoke, are normally returned free, instead of being sold as slaves; cf. Kissel 1997: 505. On the venditio sub hasta in the late republican and imperial era, see the monograph of Garcia Morillo (2005: parts 41–2). 22 Livy 10. 36. 14 and 19: The consul sees himself refusing the triumph ob amissa tot milia militum (‘because so many thousand soldiers had been lost’). 19

192

Rites and Practices of Warfare in Italy

One isolated passage of Frontinus’ Stratagemata mentions Roman soldiers being sent under the yoke by Hannibal and consequently being forced by the consul Otacilius Crassus to camp outside the retrenchment.23 This questionable incident leads on to a little series of what our sources call ‘sending under the yoke’ inflicted on Roman armies by foreign peoples. But this expression is probably being used rather loosely here, to give the Romans some idea of what their troops suffered, using a well-known form of words. Moreover, we are sometimes told that these are merely ‘hearsay’ or ‘rumours’. In the winter of 110–109 , Jugurtha blockaded Aulus Albinius’ camp at Suthul and would not let his army leave Numidia until it had gone under the yoke.24 In 107 , the consul L. Cassius was defeated at Agen by the Helvetii. His army, said Caesar, went under the yoke.25 During the principate of Nero, in  61–2, the Parthians seized Armenia, and made the Roman legions go under the yoke. At least this is what Eutropius26 and Suetonius27 say, whereas Tacitus,28 who is more reliable, only speaks of rumor and a sort of stage set that might vaguely suggest the preparations for going under the yoke, but that is not the same thing. In the final analysis, then, going under the yoke appears to be a custom that is mainly, indeed exclusively Italian, Italic, and above all Roman. This is clearly asserted in a famous passage by Dionysius of Halicarnassus—‘it is a Roman custom . . .’:

23 Frontinus, Stratagemata 4. 1. 19: Otacilius Crassus consul eos, qui ab Hannibale sub iugum missi redierant, tendere extra vallum iussit, ut immuniti assuescerent periculis et adversus hostes audentiores fierent. (‘Otacilius Crassus the consul ordered those who had returned after being sent under the yoke by Hannibal to move outside the rampart, so that having no defences they would become accustomed to dangers and more daring in the face of the enemy.’) 24 Sallust, Jugurthine War 38. 9 and 49. 2. Cf. Brizzi 1990. 25 Caesar, Gallic Wars 1. 7. 4 and 49. 2. The Swiss painter Ch. Gleyre was freely inspired by this episode: cf. Paschoud 1995. 26 27 Eutropius 7. 14. Suetonius, Nero 39. 1. 28 Tacitus, Annals 15. 15: addidit rumor sub iugum missas legiones et alia ex rebus infaustis, quorum simulacrum ab Armeniis usurpatum est. Namque et munimenta ingressi sunt, antequam agmen Romanum excederet, et circumstetere vias, captiva olim mancipia aut iumenta agnoscentes abstrahentesque; raptae enim vestes, retenta arma, pavido milite et concedente, ne qua proelii causa exsisteret. (‘Rumour added that legions had been sent under the yoke and other tales, deriving from the unfortunate situation, one of which was that a statue had been seized by the Armenians. For they had penetrated the fortifications before the Roman column was able to leave and also lined the roads recognizing and snatching away previously captive slaves or beasts; clothes had also been plundered, weapons retained, while our soldiers were terrified and let possessions go, in case some cause for a battle should arise.’)

Rites and Practices of Warfare in Italy

193

In fact, it is a Roman custom, when they have defeated enemies who lay down their arms, to plant two straight stakes into the ground and to attach a third transversally at the top; then to lead the prisoners of war under these stakes and to let them go home free once they have passed under. They call this the yoke.29

This passage is one of those that—apparently—describe most clearly the device called the yoke and for this reason it should be compared to an entry by Festus (and to the passage in Livy, Book III, already quoted, that deals with the humiliation of the Aequi):30 Defeated enemies are said to be sent under the yoke, when stripped of all their arms, two spears are planted in the ground and a third one fastened to their tops, as a kind of door, and they are ordered to pass through underneath it.31

The difference is obvious: Dionysius of Halicarnassus speaks of pieces of wood, two vertical and one transversal, Festus of three lances, two stuck in the ground, the third tied horizontally to the top of the other two. The reason is clear: Dionysius is conflating this with another, well-known device: the tigillum Sororium, the sister’s beam, under which Horatius goes to purify himself after the murder of his sister, who was engaged to one of the Curiatii whom he killed. That these two devices, even if they are similar, were not to be confused was clear to the ancients. Livy, like Festus (or his source Verrius Flaccus, who probably took this from Livy), describes on two separate occasions, in different terms and without seeking to compare them, the two structures of beams and lances. Later Livy and Festus make it quite clear that Horatius was not sent under the yoke. He was ‘velut sub jugum missus’ (‘sent under the yoke, as it were’).32 This nuance is very important.

29

Dionysius of Halicarnassus, Roman Antiquities 3. 22. 7. Livy 3. 28. 11; cf. above n. 13. Festus 394L: sub iugum mitti dicuntur hostes victi, ereptis omnibus armis telisque, cum hastis defixis duabus in terra, tertiaque ad summum earum deligata, specie Iani, iubentur subeuntes transire. 32 Festus 380L: ‘It is called the “sister’s beam” for this reason: by an agreement between King Tullus Hostilius and Mettius Fufetius, leader of the men of Alba, there was a battle involving three Horatii and three Curiati, on the understanding that power would go to the victors. Our Horatius was the victor and on his way home his sister, meeting him and knowing of the death of her betrothed by her brother’s hand, avoided his kiss. For this, Horatius killed her; and even though his father acquitted him of any crime, he was accused of parricide in the court of the duumviri, and when condemned, appealed to the people. He won his appeal, but submitted to passing under a beam (his father had set up two beams with a third fixed above)—so it was just like being sent under the yoke. After altars at the 30 31

194

Rites and Practices of Warfare in Italy

So we should avoid confusing two practices which may have points of contact, but which are nonetheless profoundly different, one (taking place?) at the gates of the city and thus a ritual for returning to civilized space, the other in the field, in both the geographical and military sense of the word. We should also be careful not to carry over uncritically everything that has been said, by G. Dumézil33 amongst others, by the adherents of ‘rites de passage’ (since Frazer),34 or by the specialists in Roman topography35 about the ritual of the tigillum Sororium, a threshold on which the soldier laid down his warlike frenzy. That is something quite different. Distancing ourselves from this comparison with the very archaic tigillum Sororium also allows us not to give too much importance to the problem of origins, and not to get too obsessed about it. Everyone will agree that going under the yoke refers to an archaic concept of warfare and of society, a shame culture, even if this concept is less used by anthropologists today than used to be the case in 1960s.36 Going under the yoke is intended above all to bring humiliation upon those who underwent it37 and it is important to observe that Livy routinely calls this practice ignominia for the defeated enemy.38 But it is no less important to note that it remained in use until the beginning of the Hellenistic era. There are doubtless several reasons for its disappearance. War(fare) became a way of obtaining slaves, as we see in 307 . And, furthermore, imperialist war presupposes the lasting hegemony of one of

spot had been consecrated to Juno Sororia and Janus Curiatius, Horatius was freed from all guilt for the crime, the auspices giving approval. So that is why that beam is called “the sister’s beam”.’ 33

Dumézil 1942: 110–15; Dumézil 1985: 36–8. Versnel (1970: 137ff.) offers a considered critique of these theories, which have been re-proposed incessantly since The Golden Bough. He notes (p. 147) that ‘iugum is found nowhere to denote a gate-like structure’ (cf., all the same, Festus 394L: ‘specie Iani’) and quite rightly rejects the confusion between passing under the yoke and return by the gate (triumphal!) of the warrior (victor!) into the city. For him, what makes most sense in the ritual of having the defeated army pass below the yoke, is ‘the magic power of the hasta and the undressing and disarming of the enemy’. In any case, as mentioned above in n. 21, the passage under the yoke is not at all assimilable to a sub hasta venditio (sale under the spear) precisely because the defeated who suffered this humiliation were not prisoners of war and departed as free men—at least in principle. 35 Coarelli 1983: 111–18. 36 On its use in Roman history-writing, see esp. Flower 1996. 37 Rüpke (1990: 211–12) and, following him, Kissel (1997: 505) highlight ‘the aspect of humiliation’ (Demütigungsaspekt) as the essential significance of passage under the iugum ignominiosum. 38 Livy 3. 23. 5 (cit. n. 12); 3. 28. 9–11 (cit. n. 13); 4. 10. 4–5 (cit. n. 14); 9. 6. 1; 9. 15. 8 (cit. n. 18). 34

Rites and Practices of Warfare in Italy

195

the adversaries, which is incompatible with the act of releasing the defeated army, an operation that re-balances the opposing forces. Incontestably, sending under the yoke refers back to a traditional concept of victory where the actual possession of spoils, particularly weapons, and the symbolic humiliation of enemies were enough. The Romans and Samnites did not—yet—adopt a war of extermination (that came later, under Sulla). In this sense, sending the Romans under the yoke at the Caudine Forks does not seem to me to be a third way between vengeance and forgiveness as might be suggested by the rhetorical debate that the Samnite chief Caius Pontius and his father Herennius were supposed to have had in 321.39 And as is argued in a recent article,40 that incidentally contains several good analyses. Going under the yoke was not one amongst many possible solutions. It was normal practice, which was right at that time, because it was traditional and recognised as such by the warring adversaries. But this concept of war was soon condemned.

THE LEGIO LINTEATA I shall go much more quickly over this second discussion. I am returning to what I said in a conference held in Rome in 2004, the proceedings of which were published in 2007.41 In 293  in a supreme effort to conquer the Romans, the Samnites (Pentri) concentrated all their forces in Aquilonia. In the middle of their camp (in mediis castris) they set up a square enclosure 200 feet along each side, marked out and covered with linen cloth. An old priest, Ovius Paccius, offered sacrifices there according to an ancient rite. Then the elite of the warriors went in to swear the oath. Each recruit was led up to the altars ‘more as a victim than as a participant in the sacrifice’, magis ut victima quam ut sacri particeps. Those who refused to swear were killed on the spot, round the altars (obtruncati circa altaria), in such a manner that their bodies lay in the middle of the sacrificial slaughter, inter stragem victimarum.42

39

40 41 Livy 9. 3. 4–9. 4. 1. Urso 1997. De Cazanove 2008. Livy 10. 48: [4] Tum exercitus omnis Aquiloniam est indictus. Ad quadraginta milia militum, quod roboris in Samnio erat, convenerunt. [5] Ibi mediis fere castris locus est consaeptus cratibus pluteisque et linteis contectus, patens ducentas maxime pedes in omnes partier partes [6] Ibi ex libro vetere linteo lecto sacrificatum, sacerdote Ovio Paccio quodam, homine magno natu, qui se id sacrum petere adfirmabat ex vetusta Samnitium religione, qua quondam usi maiores eorum fuissent, cum adimendae Etruscis Capuae clandestinum 42

196

Rites and Practices of Warfare in Italy

The linen enclosure is usually considered as a model for temporary Italic places of worship, the archetype of the archetypal sacred space ex vetusta Samnitium religione.43 Frequent attempts have even been made to find in the monumental or monumentalized Samnite sanctuaries a

cepissent consilium. [7] Sacrificio perfecto, per viatorem imperator acciri iubebat nobilissimum quemque genere factisque; singuli introducebantur. [8] Erat cum alius apparatus sacri, qui perfundere religione animum posset tum in loco circa omni contecto area in medio, victimaeque circa caesae, et circumstantes centuriones strictis gladiis. [9] Admovebatur altaribus magis ut victima quam ut sacri particeps, adigebaturque iure iurando, quae visa auditaque in eo loco essent, non enuntiaturum; [10] dein iurare cogebant diro quodam carmine in execrationem capitis familiaeque et stirpis composito, nisi isset in proelium, quo imperatores duxissent, et si aut ipse ex acie fugisset, aut si quem fugientem vidisset, non extemplo occidisset. [11] Id primo quidam abnuentes iuraturos se obtruncati circa altaria sunt, iacentes deindeinter stragem victimarum document ceteris fuere, ne abnuerent. [12] Primoribus Samnitium ea detestation. obstrictis, decem nominati ab imperatore; eis dictum ut vir virum legerent, donec sedecim milium numerum confecissent. Ea legio linteata ab integumento consaepti, in quo sacrata nobilitas erat appellate; his arma insignia data et cristatae galeae, ut inter ceteros eminerent. (‘Then the whole army was summoned to Aquilonia. Assembled there were about 40,000 troops, the full strength. of Samnium. Almost in the middle of the camp, there was a place that was fenced round with hurdles and boards and covered over with linen; it stretched 200 feet equally in all directions. There, a certain priest of great age, Ovius Paccius, who declared that he had sought out the ritual from the ancient religio (record of worship) of the Samnites; a ritual that their ancestors had once used, at the time they had a secret plan to snatch Capua from the Etruscans. After the sacrifice, the general gave orders through a herald to summon all the men most noble for their birth and their deeds. These were brought in, one by one. Together with some other sacred apparatus, which might fill one’s heart with awe, there were also altars in the middle around in the whole enclosure; there were also slain victims about and, standing by, centurions with drawn swords. the soldier was brought to the altar, more like a victim than a participant in the worship and bound by an oath not to disclose whatever was seen and heard in that place. Then, they forced him to swear through a terrible chant imposing a curse on himself, his family and his lineage, if he failed to go into battle, where the commanders had led, or had deserted from the line of battle himself, or had seen a man deserting and not instantly killed him. Some at first refused to swear and were cut down by the altars, then lying amongst the carnage of the victims—a warning to the others not to refuse. Once the leading men among the Samnites had been coerced by the curse, ten were nominated by the commander; they were ordered that each man should choose another man until they had reached the figure of 16,000. The legion was named “Linteata” after the covering of the enclosure, where they had been sworn in. They were provided with brilliant armour and crested helmets to distinguish them from the others.’) The bibliography on this subject is very large. Among others, see Salmon 1967 and more recently Tagliamonte 1996, 183–5. In addition, on Livy’s point of view and his distortions, see Saulnier 1981. To the problem of identifying Aquilonia in Samnium (here set aside) various solutions have been suggested in the last thirty years: cf. de Cazanove 2008. On the assimilation (only rhetorical) between this ceremony and human sacrifice, see de Cazanove 2000, 253–4. 43

La Regina 1984, 234–5.

Rites and Practices of Warfare in Italy

197

version in stone of such enclosures, notably at Pietrabbondante44 and also at S. Giovanni in Galdo.45 Yet, if we go back to the material facts that Livy mentions in his description, we can make a comparison that takes us in a completely different direction: the enclosure (the locus consaeptus) is situated more or less in the middle of the Samnite camp (mediis fere castris)46—even though the linen enclosure is usually treated as if it were isolated, and not an integral part of the encampment; it is a square with sides of 200ft (patens ducentos maxime pedes in omnes pariter partes);47 it has walls made of hurdles and boards (cratibus pluteisque) and a cover (integumentum) of linen cloth (it is a locus . . . linteis contectus); we must emphasize that it is a covered place, which therefore cannot under any circumstances have served as a model for temporary places of worship open to the sky. So, it must have looked like a huge tent; inside is all the sacrificial equipment (apparatus sacri), amongst other things the altars (arae, altaria).48

Categorized in this way, the elements so described strongly suggest a surprising comparison: the linen tent in the middle of the Samnite camp matches point by point the central space in the Roman camp, Polybius’ famous digression (in Book 6) on the Roman camp, which has the advantage for us of not being too distant in time—less than a century and a half—before the Samnite Wars. The praetorium—the ‘general’s tent’, in Polybius’ words, is the geometrical centre around which the whole camp is built: ‘The standard is placed where the tent should be pitched, and a quadrilateral is measured out round the standard, so that all its sides are a hundred feet from the standard and its surface area is four plethrae’.49 Then the other tents and the spaces for moving around in are ranged around the praetorium, which therefore stands in the middle of the camp that Polybius, and later Flavius Josephus,50 both compare to a town. The praetorium, according to the figures given by Polybius, is therefore a square with sides measuring 200 feet. Indeed, the archaeologically attested praetoria are very often square or more or less square.51 One of 44 Coarelli 1996; cf. Sisani 2001. On the contrary. La Regina (1984, 235) only IV recognizes ‘a purely typological . . . merit’ in the comparison he sets up between the first phase of the sanctuary of Pietrabbondante (restored by him as a quadrangular enclosure of 200 Oscan feet = 55 m. a side) and Livy’s description. He does not go so far as to recognize a real coincidence between them. 45 46 Capini 1996; cf. Coarelli 1996, 8–9. Livy 10. 38. 5. 47 48 Ibid. (complete citation in n. 42). Ibid. 38. 8–9. 49 50 Polybius 6. 27. 1–2. Josephus, Jewish War 3 .83. 51 For a catalogue of the praetoria, together with their dimensions, one should always turn to the dissertation of Lorenz (1936: 12ff.).

198

Rites and Practices of Warfare in Italy

the most ancient ones, the praetorium in camp III at Renieblas near Numantia (which is thought to have been the consul Q. Fulvius Nobilior’s camp in 153 ),52 was considered by A. Schulten as having been built on a square plan, with sides of about 59 m = 200 Roman feet.53 It was on the highest point (altitude 1152) and in the middle of the huge camp that was itself of an irregular shape. Despite its great size, it had to be possible for the ‘general’s tent’, at least under the republic, to be entirely covered over on occasion. This emerges from a passage in Varro’s De lingua Latina, which compares the praetorium to an atrium testudinatum, i.e. one that lacked a compluvium (drain): ‘they used the term cavum aedium for the covered space which remained separate inside the house-walls, for use, in a general way, by everyone. If no part of it was left open to the sky, the whole structure was called a testudo (tortoise) because of its resemblance to a tortoise, as was the case with the praetorium in the camps’.54 The praetorium in the centre of the camp was ‘almost like a temple’, as Flavius Josephus says.55 Thus, when Livy evokes the scene in which he places the recruitment of the Samnite Legio Linteata, he is in fact describing something entirely similar to the central block in the legionary camp. What is very widely considered as a model for Italic places of worship(?), temporary or permanent, open to the sky or (very occasionally) covered—or at least as the only Oscan sanctuary sufficiently well known to literary sources—has an exact parallel that is purely Roman. This parallel, furthermore, is not properly speaking a sanctuary, but part of a military establishment, even if it can contain the furnishings of religion. E. T. Salmon, in his classic monograph Samnium and the Samnites, had already noted that Livy’s description may perhaps partly borrow from Polybius’ description of the Roman camp,56 but without further developing this brief point. Once we have recognized that the central enclosure in the Samnite encampment and the praetorium in the Roman camp are identical, we must explain this. Two types of solution can be offered.

52 Schulten’s dating (see the next n.) has been confirmed by discoveries of coinage: Hildebrandt 1979: 240–4; 248–51; 254; 265; 270–1. 53 Schulten 1912: esp. cols. 84–87, fig. 1 and pl. 2; Schulten 1929: 19ff. 54 Varro, De lingua Latina 5.161: cavum aedium dictum qui locus tectus intra parietes relinquebatur patulus, qui esset ad communem omnium usum. In hoc si locus nullus relictus erat, sub divo qui esset, dicebatur testudo ab testudinis similitudine, ut est in praetorio et castris. 55 56 Josephus, Jewish Wars 3. 82. Pp. 111 and 197 of the Italian translation.

Rites and Practices of Warfare in Italy

199

The simplest and certainly the most rational hypothesis: Livy (or his source) started from his own knowledge of the Roman encampment and then transposed it in order to redesign, to reinvent the surroundings in which to situate the recruitment of an Italic army. The opposite hypothesis, which we must at least mention, even if it is less neat and less likely than the other: if we choose to believe Livy’s description here, we must allow (the possibility) that the Samnites had a camp organized in accordance with the rules described by Polybius, from as early as 293 . If we want to pursue this idea all the way we may even be tempted to suggest that the characteristic layout of the Roman camp derived from an Italic model since, according to Frontinus’ account, the Romans did not have an organized camp until the war against Pyrrhus, but simply bivouacked with small scattered groups of tents like mapalia.57 And if this was the case, we could evidence the other Roman borrowings from the Samnites that I mentioned at the beginning.58 However, to complicate an already fairly vexed question, we should note king Pyrrhus’ admiration, according to Plutarch, for the perfect organization of the Roman camp, which he describes as ‘not barbarian’.59 The first hypothesis—the recruitment of the Legio Linteata staged by Livy in a ‘Roman-style’ camp—is, let us repeat, the most probable. In any case, what it is important to emphasize here is that the historian talks about the military camp, and, inside it, the general area that has, amongst other things, a religious aspect. A place of worship if you like, but a very unusual one, which cannot under any circumstances be taken as a model for Italic places of worship in general, although this is what is usually done.

57

Frontinus, Stratagemata 4.14: castra antiquitus Romani ceteraeque gentes passim per corpora cohortium velut mapalia costituere soliti errant, cum solos urbium muros nosset antiquitas. Pyrrhus Epirotarum rex primus totum exercitum sub eodem vallo constituere instituit: Romani deinde, victo eo in campis Arusinis circa urbem Malventum, castris eius potiti et ordinatione notata, paulatim ad hanc usque metationem quae nunc effecta est pervenerunt. (‘In the old days, the Romans and other peoples set up camps according to wherever the soldiers laid down, like mapalia, since antiquity only knew about city-walls. Pyrrhus, king of Epirus, was the first to set up his army within the same vallum (rampart). The Romans after that, when Pyrrhus had been defeated in the Arusine fields round the city of Malventum, having captured his camp and taken note of its organization, gradually worked their way to the present method of measurement.’) 58 Briquel 1986. 59 Plutarch, Pyrrhus 16. 17. The opposite view in Livy 35. 14. 8: Pyrrhum castra metari primum docuisse (‘. . . they were the first to have taught Pyrrhus to measure out the camp’).

200

Rites and Practices of Warfare in Italy B I B L I OG R A P H Y

Bénichou-Safar, H. (1993). ‘Le rite d’entrée sous le joug. Des stèles de Carthage à l’Ancien Testament’. RHR 210: 131–43. Berthier, A. and H. Tayeb (1970). ‘Une inscription à Saturne d’Aziz ben Tellis et la formule “sub iugum intrauit” ’. Bulletin d’archéologie algérienne 4: 301–12. Briquel, D. (1986). ‘La tradition sur l’emprunt d’armes samnites par Rome’, in A.-M. Adam and A. Rouveret, Guerres et sociétés en Italie aux Ve et IVe siècles avant J.-C. Paris, 1986, 65–89. Briquel, D. (2001). ‘La tombe Andriuolo de Paestum (IX, 31)’, in Briquel and Thuillier, 2001, 135–46. Briquel, D. and G. Brizzi (2000). Ch. VII ‘La marche vers le sud’, in Hinard, 2000, 245–92. Briquel, D. and J.-P. Thuillier, eds. (2001). Le censeur et les Samnites. Sur TiteLive, livre IX. Paris. Brizzi, G. (1990). ‘Giugurta, Calama e i Romani sub iugum’, in L’Africa romana. Atti del VII convegno di studio. Sassari, 855–70. Caiazza, D. (2004). Safinim. Studi in onore di Adriano La Regina per il premio ‘I Sanniti’, Piedimonte Matese. Capini, S. (1996). ‘Su alcuni luoghi di culto nel Sannio Pentro’, in Del Tutto and Palma, 1996, 63–8. Coarelli, F. (1983). Il Foro Romano. 1. Periodo arcaico. Rome. Coarelli, F. (1996). ‘Legio linteata. L’iniziazione militare nel Sannio’, in Del Tutto and Palma, 1996, 3–16. de Cazanove, O. (2000). ‘Sacrifier les bêtes, consacrer les hommes. Le printemps sacré italique’, in Verger, 2000, 253–76. de Cazanove, O. (2008). ‘Il recinto coperto del campo di Aquilonia: santuario sannita o praetorium romano?’, in Saturnia Tellus. Definizioni dello spazio consacrato in ambiente etrusco, italico, fenicio-punico, iberico e celtico. Atti del convegno internazionale di studi, Roma, 10–12 novembre 2004, 335–9. De Sanctis, G. (1907). Storia dei Romani II. La conquista del primato in Italia. Turin. (repr. 1960. Florence). Dumézil, G. (1942). Horace et les Curiaces. Paris. Dumézil, G. (1985). Heur et malheur du guerrier. 2nd edn. Paris. Flower, H. (1996). Ancestor Masks and Aristocratic Power in Roman Culture. Oxford. Fraccaro, P. (1975). Opuscula, 4, Della guerra presso i Romani. Pavia. Garcìa Morcillo, M. (2005). La ventas por subasta en el mundo romano: la esfera privada. Barcelona. Hildebrandt, H. J. (1979). ‘Die Römerlager von Numantia. Datierung anhand derNumantia Münzfunde’. Madrider Mitteilungen 20: 238–71. Horsfall, N. (1982). ‘The Caudine Forks: topography and illusion’. PBSR 50: 45–52. Humm, M. (2005). Appius Claudius Caecus. La république accomplie. BEFAR 322. Rome.

Rites and Practices of Warfare in Italy

201

Jones, H. (2004). Samnium. Settlement and Cultural Change. Proceedings of the 3rd E. Togo Salmon Conference on Roman Studies. Providence. Kissel, T. (1997). ‘Sub iugum mittere. Zur kollektiven Bestrafung unterworfener Kriegsgefangener im republikanischen Rom’. Antike Welt 28: 501–7. La Regina, A. (1984), [F. Corelli], Abruzzo Molise (Guide archeologiche Laterza), Bari, 1984, 234–5. Libourel, J.-M. (1973). ‘A battle of uncertain outcome in the second Samnite war’. AJPh. 94: 71–8. Lorenz, H. (1936). Untersuchung zum Prätorium. Katalog der Prätorien und Entwicklungsgeschichte ihrer Typen. Halle. Magdelain, A. (1943). Essai sur l’origine de la sponsio. Paris. Nock, A. D. (1926). ‘Intrare sub iugum’. CQ 20: 107–9. Paschoud, F. (1995). ‘Les Romains sont-ils passés sous le joug à Montreux? À propos d’un célèbre tableau de Charles Gleyre’. Mus. Helv. 52: 49–62. Rüpke, J. (1990). Domi militiae. Die religiöse Konstruktion des Krieges in Rom. Stuttgart. Salmon, E. T. (1967). Samnium and the Samnites. Cambridge (Ital. tr. Il Sannio e i Sanniti, Turin [1995]). Saulnier, Chr. (1981). ‘La coniuratio clandestina. Une interprétation livienne des Schulten Schulten, A. (1912). ‘Ausgrabungen in Numantia’. AA : 82–99 (Jahrbuch des Deutschen Archäologischen Instituts 27). Schulten, A. (1929). Die Lager bei Renieblas. Numantia IV. Munich. Sisani, S. (2001). ‘Aquilonia: una nuova ipotesi di identificazione’. Eutopia 1: 131–47. Soprintendenza archeologica di Roma (2000). Studi sull’Italia dei Sanniti. Milan; Rome. Bibliography, 303–35. Sordi, M. (1960). I rapporti romano-ceriti e l’origine della civitas sine suffragio. Rome. Sordi, M. (1965). ‘Sulla cronologia liviana del IV secolo’. Helikon 5: 3–44. Tagliamonte, G. (1996). I Sanniti. Caudini, Irpini, Pentri, Carricini, Frentani. Milan. Tagliamonte, G. (20062). I Sanniti. Milan. Urso, G. (1997). ‘Le force Caudine, media via tra vendetta e perdono,’ in Amnistia, perdono e vendetta nel mondo antico. Milan, 237–51. Versnel, H. S. (1970). Triumphus. An Inquiry into the Origin, Development and Meaning of the Roman Triumph. Leiden.

III Sacrifice

8 Festus and the Role of Women in Roman Religion Rebecca Flemming In his magisterial summary of ‘the religious roles of Roman women’,1 J. Scheid argues that these roles were essentially marginal to the religious life of the city.2 Roman women were characterized by a series of religious incapacities and disabilities which forced them to the edges of Roman religious practices: into subordinate ritual functions and a kind of sacred ‘passivity’, or into ‘suburban sanctuaries and the temples of foreign gods’, and even into ‘deviant’ religious activity and thought. It was not that they failed to participate in religious affairs, indeed some women, especially those in priestly offices, participated very fully; but that, in general, their participation itself played out their incapacity. It confirmed their subordination and constructed, through ritual, an image of what threatened male citizens at the boundaries of the Roman order, while the ritual itself helped manage that threat, maintain the order. At the centre of this picture is a woman’s ‘sacrificial incapacity’ (incapacité sacrificielle), her exclusion from the main elements of sacrificial ritual—the slaughter, butchering, and distribution of the meat—as well as various associated activities. Scheid claims to establish this principle at the outset, with the help of the ancient authors Festus and Plutarch, together with some of their modern interpreters, in particular O. de Cazanove; though M. Detienne’s (controversial) arguments about the exclusion of women from sacrifice in classical Greece also lurk in the

1 My thanks to Fay Glinister for giving me the opportunity to present my views on these matters (long held!) at the Festus Colloquium, and to all the participants for their helpful contributions. 2 Scheid 1992: 377–408 (Fr. original Scheid 1991: 406–37) cf. also Scheid 2001: 33–4.

Rebecca Flemming, Festus and the Role of Women in Roman Religion In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0009

206

Festus and the Role of Women in Roman Religion

background.3 This female incapacity then provides the basic framework for the rest of the article. So, those women who did perform a sacrificial role are seen as the exceptions that prove the rule, as either crossing into ‘male’ territory (like the Vestal Virgins, and, rather differently, the regina sacrorum and flaminicae), or as officiating in naturalized, ‘foreign’ cults (such as the cult of Ceres or Isis). The major matronal festivals, such as the Matralia, it is pointed out, generally did not involve blood sacrifice, and, again, exceptions such as the festivities surrounding the cult of Fortuna Muliebris, the Bona Dea, and, indeed, the Secular Games, should be understood as either encroaching on male roles, or marginalized by the site and modes of their performance, or foreign; and, of course, women’s conduct during the Bacchanalian scandal exemplifies deviant religious activity. The secondary role of women, below that of the paterfamilias, is also manifest in more everyday religious behaviour, completing this picture of female subordination and disability. Women’s sacrificial incapacity thus becomes the organizing principle around and through which all female religious activity at Rome can be arranged and understood. This picture of the restricted and marginal role of women in Roman religion has become widely accepted, though there is some hesitation on the detail. In their authoritative Religions of Rome, for example, M. Beard, J. North, and S. Price speak of the ‘limited role’ of women in Roman state cult, and emphasize that the paterfamilias ‘always’ took the lead in family religious action.4 Women passively attended, but had little opportunity to take an active part in, public religious events, and: much more fundamentally (though the evidence is not entirely clear), they may have been banned – in theory, at any rate – from carrying out animal sacrifice; and so prohibited from any officiating role in the central defining ritual of civic religious activity.5

There is, moreover, no reason to be surprised that such a strongly patriarchal society as Rome, and one that so closely entangled notions of religious and political power, should exhibit such clear inequalities in the ritual sphere. As Scheid points out, civic sacrifices were performed on behalf of, in the name of, the whole community, and to have a woman representing Rome in its relations with the gods in this way, would clearly contravene the general Roman principle that only men have this representational capacity.6 3 Cazanove 1987: 159–73; Detienne and Vernant 1989; for the counter-argument on Greek sacrifice, see Osborne 1993: 392–405. 4 5 6 Beard et al. 1998: 71. Beard et al. 1998: 297. Scheid 1992: 406.

Festus and the Role of Women in Roman Religion

207

There are, however, a number of problems with this picture of the role of women in Roman religion, especially in its strongest formulations. As Beard herself has pointed out in the ‘affectionate critique’ she has since offered of her classic article on ‘The sexual status of Vestal Virgins’, it is a mistake to treat gender categories as ‘givens’, and, in particular, as constructed entirely outside the operations of religion in any society, when in fact religion regularly acts as a ‘privileged space’ for the definition of gender roles.7 The project she undertook in her earlier essay, of identifying the ‘matronal’, ‘virginal’, and ‘male’ elements that were combined in the figure of the Roman Vestal (and, of course, revelling in rather than attempting to reconcile those contradictions and ambiguities) was, therefore, misleadingly narrow, and needs to be widened out to include an enquiry into what the Vestals themselves contributed to Roman notions of virginity, marriage, and motherhood, to relations between men and women more broadly. Scheid should take this criticism into account, especially since he relies on Beard’s ambiguity thesis (or at least his version of it) to help make the case for counting the Vestals’ sacrificial responsibilities among their ‘male’ attributes. But why should the female holders of priestly positions in Rome who acted in a sacrificial capacity be understood as crossing into ‘male’ territory, rather than as serving to expand the female realm? After all, it is only by these kinds of policing measures that the masculinity of sacrifice can be maintained, through, that is, an essentially circular form of argument. Similar objections have also been raised with respect to other parts of the marginality model—that pre-existing assumptions about women’s position within the Roman state and society have over-determined views of her religious roles.8 It is therefore assumed that those cults and rites in which women actively participated must, a priori, have been marginal, because of that very participation. There are also acute empirical difficulties with the more limited interpretations of female religious activity at Rome, especially with the case for female sacrificial incapacity so central to the arguments and overall conclusions of De Cazanove and Scheid. It is these, rather than the broader theoretical issues, that this paper will focus on, contending that the ancient evidence cannot support even uncertainty about Roman women’s participation in the main elements of sacrificial ritual. The Lexicon of Festus, in particular, not only explicitly contradicts any notions of general female exclusion from sacrifice, but also offers

7 8

Beard 1995: 166–77, at 169–70; the original article, Beard 1980: 12–27. Staples 1998: esp. 1–8; Schultz 2006.

208

Festus and the Role of Women in Roman Religion

numerous indications of women’s wider religious activities as an integral part of the Roman religious landscape. Though the main drive of this essay is refutational, therefore, it is none the less intended to contribute to the discussion in a more constructive way too, and to offer some suggestions of how the religious roles of Roman women might be more helpfully refigured.

FESTUS, PLUTARCH, AND FEMALE SACRIFICIAL ACTIVITIES Scheid opens his argument for female sacrificial incapacity in ancient Rome with a passage from Paul’s abridgement of Festus: Exesto, extra esto. Sic enim lictor in quibusdam sacris clamitabat: hostis, vinctus, mulier, virgo exesto; scilicet interesse prohibebatur. Exesto: be away! Thus the lictor used to shout out in certain religious rituals—‘Away with the foreigner, prisoner, woman, girl!’—obviously prohibiting them from being present.9

The excerpt is an intriguing one, particularly in its coupling of hostis and vinctus with mulier and virgo as if the two pairings were in some way equivalent; however, rather than supporting the case that ‘women . . . were forbidden to participate in sacrificial rituals’, it clearly suggests the exact opposite.10 The ‘quibusdam’, is surely pivotal here, so it is only in certain rituals that the lictor bans women, girls, enemies, and prisoners. In general they are present. The shout of ‘exesto’ is the exception not the rule: it requires a special effort to preclude women from participating in religious activities, and so also a special lexical explanation. That still leaves, however, the question of which sacra women found themselves excluded from, and their significance. For even if the general presence of women is thus established, the exceptions might none the less tell against their religious position. The list is, however, far from impressive. There is, of course, the male exclusivity of the cult of Mithras, but apart from that only the prohibition of female participation in the rites of Hercules performed at the Ara Maxima (in the Forum Boarium), and a ban on female attendance at a sacrifice made to Mars-Silvanus for

9 72.10P s.v. exesto; it may also be that a third pairing has been lost in the process of abridgement, for Paul’s epitome. 10 Scheid 1992: 379.

Festus and the Role of Women in Roman Religion

209

the well-being of a farm’s cattle herd, stand any scrutiny.11 Both of these instances are, moreover, exceptions in their own terms. Despite various claims to the contrary, in general women actively worshipped both Hercules and Silvanus, including participation in the sacrificial rituals involved, as C. Schultz and P. Dorcey have, respectively, shown.12 Indeed, the ancient source that provides the most detail about women’s exclusion from rites at the Ara Maxima serves to prove the general point. In Plutarch’s Quaestiones Romanae, a text that, like Festus’ Lexicon, seeks to explain selected aspects of Roman culture and religion, history, and tradition, one of the questions asked is (60): Διὰ τί, δυοῖν βωμῶν Ἡρακλέους ὄντων, οὐ μεταλαμβάνουσι γυναῖκες οὐδὲ γεύονται τῶν ἑπὶ τοῦ μείζονος θυομένων; Why, there being two altars of Hercules, do women neither participate in, nor receive taste of, the sacrifices offered on the Greater one?

It is then clear that women were completely excluded from the sacrificial events, from both the actual offering and the banquet that follows; but only at one altar. A contrast is drawn here between procedures at the two sites dedicated to Hercules, and the question only makes sense if women did honour the god, in the same manner, at the smaller altar. Nor is this the only altar to link women with Hercules: one of the inscriptions to which Schultz has recently drawn attention as evidence of female adherence to this deity records the restoration of an altar sacred to Hercules carried out (along with the building of a temple to the god) by one Publicia, daughter of Lucius Publicius and wife of Gnaeus Cornelius (who also contributed financially to the project).13 In the case of Silvanus also, study of the corpus of inscriptions dedicated to the god has helped to dispel previous assumptions about his male orientation. As Dorcey concludes, ‘nothing in the epigraphic record suggests that women acted any differently from men in cult worship’, and the group of female dedications includes again a (slave) woman who records her reconstruction of an Ostian shrine to Silvanus.14 Returning to the domain of civic religion, however, it is also worth drawing attention to various echoes of the Hercules pattern which are to be found in the cult of the Bona Dea. Her December rites certainly and

11 Ara Maxima: Propertius 4. 9; Macrobius, Saturnalia 1. 12. 28 and Plutarch, Roman Questions 60; sacrifice ‘pro bubus’: Cato, De agricultura 83. 12 Schultz 2000: 291–7; Dorcey 1989: 143–55. 13 CIL 6.30899 and see Schultz 2000: 294. 14 Dorcey 1989: 149; and see for discussion of men’s cult activities his book, Dorcey 1992. The inscription recording the slave-woman Theodora’s reconstruction is CIL 14.4327.

210

Festus and the Role of Women in Roman Religion

very famously excluded men, but inscriptions attest to male worship of this ‘women’s goddess’, just as much as to women’s involvement with the quintessentially ‘masculine’ Hercules and Silvanus.15 Both instances of exclusivity also have their special position emphasized further by the multiple aetiological speculations that surround them, some of which explicitly link the two rituals. So, the prohibition of women from the sacrifice at the Ara Maxima is figured by both Propertius and Macrobius as an intentional response to men’s prohibition from the festivities of the Bona Dea.16 Hercules, tired and thirsty after his epic battle with the (Italian based) monster Cacus, was, on account of his maleness, refused water by women celebrating the rites of the Bona Dea. Enraged, the hero forcibly took the water, and banned women from the cult he instituted to commemorate his famous victory. In recognition of its ‘foreign’ origin, however archaic and distinguished, the cult continued to be celebrated in the Greek style.17 It would seem, therefore, that the rituals at the Ara Maxima were marked out as exceptional in similar ways to those of the Bona Dea; their exceptionality being a function of sexual exclusivity per se, rather than depending on which sex was excluded. However, it is now time to turn to the other sources that Scheid adduces to support his view of female sacrificial incapacity, material concerned with the ‘cuisine of sacrifice’, which he thinks can be used to widen the scope of the lictor’s ‘exesto’ to establish a general rule. The second ancient witness called is, again, Plutarch, and another of his Quaestiones Romanae (85): Διὰ τί τὰς γυναῖκας οὔτ᾽ ἀλεῖν εἴων οὔτ᾽ ὀψοποιεῖν τὸ παλαιόν; Why, in ancient times, were wives not allowed to grind or cook?

As so often in this treatise Plutarch turns to the beginnings of Rome, and material he had collected for his parallel lives, to provide an answer. He suggests that this was part of the peace deal negotiated after the rape of the Sabine women and the ensuing conflict with the Sabine men. The obvious way to understand the passage, and one made explicit in Plutarch’s coverage of the same events in his life of Romulus, is that this is a move to mollify the Sabine men by lightening the domestic load of 15 On men’s attachment to the Bona Dea, see Brouwer 1989: 254–96. It also seems to me that the very slender evidence (Ovid, Ars amatoria 3. 637–8) about the May festivities involving her temple on the Aventine is less clear on its female exclusivity than most (including Brouwer) suppose. 16 Propertius 4. 9, Macrobius, Saturnalia 1. 12. 27–8; the connection is discussed more fully (indeed too fully) in Staples 1998: 13–36. For other aetiologies of the Bona Dea, see Brouwer 1989: 324–7. 17 Varro, De lingua Latina 6.15, Dionysius of Halicarnassus, Roman Antiquities 1. 40; Macrobius, Saturnalia 3. 12. 2.

Festus and the Role of Women in Roman Religion

211

their daughters.18 De Cazanove, however, argues that, in fact, this aspect of the treaty excluded Roman women from the sacrificial scene, since grinding grain and preparing meat are key elements of the sacrificial process, relating to the production of mola salsa (the mixture of emmer flour and salt which is sprinkled on all offerings to the gods), and the butchery of the sacrificial victim respectively.19 The attempted translation of these events from their explicitly domestic setting into the religious sphere, fails, however, on a number of fronts. Firstly it demands that, rather than negotiating household benefits for their daughters, the Sabine leaders are to be understood to have negotiated their systematic ritual exclusion. Moreover, it implies a structural contradiction between women’s religious and domestic activities that is very hard to believe. It is more usual to suggest the reverse; to align, for example, the Vestal Virgins’ responsibilities for the tending of the fire in the public hearth in the temple of Vesta, for cleaning the temple, and for the preparation of the mola salsa, precisely with (to quote Beard), ‘the domestic role of the early Roman Hausfrau’.20 Secondly, this move is not in any sense philologically compelled, or even encouraged. Alein is a general term for the grinding of grain, not a technical, ritual one, and, though in his answer Plutarch does use the verb mageirein, which is used of butchery in a sacrificial context as well as possessing a general culinary frame of reference, the word that appears in his question (opsopoiein) has no such religious connotations, usually referring to fancy cooking.21 Finally, there is the problem of ‘to palaion’, the fact that Plutarch is not talking about the Roman present, but about a distant, mythic, Roman past. Scheid suggests that, though the interdiction on women performing these domestic chores was short-lived, it is thanks to the survival of the sacrificial prohibition that this story was worth telling, that it served an aetiological function.22 It would, of course, then have to be a hidden aetiology, since that is not what Plutarch says he is explaining, nor has the need for any such explanation yet been established. As already noted, Plutarch does provide possible origins for the specific exclusion of women from sacrifices at the Ara Maxima, but, again, they would be rendered nonsensical if there had been a blanket ban in operation. If Plutarch’s inclusion of this particular question does need justification then it is more plausibly sought in his desire to re-use his Romulean material rather than anything much more profound. 18 20 21 22

19 Plutarch, Romulus 15. 4 and 19. 7. Cazanove 1987: esp. 162–8. Beard 1980: 13. The definition of opsopoieô in LSJ9 is ‘to dress food (esp. fish) nicely’ (1283). Scheid 1992: 380.

212

Festus and the Role of Women in Roman Religion WINE, WOMEN, AND THE GODS

The third and final plank in the argument for Roman women’s sacrificial disabilities concerns their troubled relationship with wine. Here the various stories about wine being forbidden to women in early Rome are mobilized to try to demonstrate a further way in which women were excluded from the sacrificial ritual. This mobilization is a complex one, and proceeds in two main stages. The first is the assertion that what was prohibited to women was not wine in general, but ‘pure’, ‘unadulterated’, wine in particular, this being what is signified by the term ‘temetum’ which appears in some of the accounts of the early ban; and that it was precisely this ‘pure’ wine which had to be given to the gods, nothing else would do. This symmetry is particularly emphasized by De Cazanove: what is forbidden for women is mandated for the gods, in respect not only to wine but also other aspects of sacrificial offerings.23 The second stage is to move from this interdiction on women drinking the specific kind of wine that was divinely ordained to claiming that they were, therefore, prevented from using this wine in any sacrificial procedures; they could not, as Virgil has Dido, pour the wine on the victim’s head before it was killed, nor could they pour libations of wine, and so on.24 The exception to this rule was, again, the cult of the Bona Dea. Unfortunately, neither part of the argument withstands scrutiny. To start, however, with what can be reasonably securely established (before moving on to the shakier parts of the reasoning), it does seem that, in a Roman religious context, only ‘pure’ wine could be offered to the gods. At least that looks like the best way to understand the rather syntactically challenging entry in the Lexicon of Festus on spurcum vinum (474. 31F): Spurcum vinum est, quod sacris adhiberi non licet, ut ait Labco Antistius lib. X commentari iuris pontifici, cui aqua admixta est defru[c]tumve, aut igne tactum est, mustumve antequam defervescat. Spurcum vinum is wine that is not allowed to be used in religious rituals, so says Labeo Antistius in Book 10 of his Commentary on Pontifical Law; wine to which water has been added, or defrutum, or wine that has been touched by fire, or mustum before it has finished fermenting.25

Taken in the context both of information about Roman wine-making and about Festus’ work as an epitomator of the larger Lexicon of Verrius Flaccus, it seems reasonable to think that this passage derives from a 23

24 Cazanove 1987. Scheid 1992: 405–6; Virgil, Aeneid 4. 60–1. My thanks to my colleagues Roland Mayer and Carlotta Dionisotti for help with this passage. 25

Festus and the Role of Women in Roman Religion

213

fuller listing of pontifically forbidden deviations from the basic wine production process. Religious law required that wine offered to the gods be based on good quality, pure mustum (freshly pressed grape juice), not mustum that had been cut with water, nor cut with, or indeed replaced by, defrutum (grape juice boiled down to increase its sugar content).26 It should be fermented in normal, stable, climatic conditions, not using heat artificially to assist the process, and the fermentation must have reached completion. That is, pontifical law prohibits deviations from the basic procedure by which a good quality wine would be produced, a wine which, in Columella’s phrase, ‘gives pleasure by its natural qualities’ (quod suapte natura placere), or in Pliny’s, ‘is a product of grape and soil’ (uva terraque constant).27 In other words, wines are judged suitable for ritual purposes in the same way that sacrificial victims are; they must be fine, unblemished specimens of their species.28 The quality and purity of the wine to be offered to the gods is then reasonably clear. What is much less clear is that the reverse was true for women: that they were forbidden exactly what was divinely mandated. Of all the sources that refer to the ancient ban on female wine consumption it is perhaps Aulus Gellius (Noctes Atticae 10. 23) who comes closest to providing support for this view, but not close enough. Gellius begins his discourse on this topic by repeating what others have written about the temperate lives led by women in early Rome and Latium; that they completely abstained from wine (vinum, ‘which was called temetum in the old language’, quod temetum prisca lingua appellatur), an abstention that was policed by the customary kissing amongst kinfolk. That is he begins by claiming a total prohibition of female wine drinking, a claim that is echoed in the other texts also, as is the point that temetum is simply the archaic word for ‘wine’.29 Gellius then offers a possible qualification to this blanket ban, or at least to its implementation. According to the writers he is following, archaic women did customarily drink a set of wine-related, or wine-based drinks—lorea (a thin, afterwine, made from a second pressing or dregs), passum (raisin-wine), murrina (myrrh-wine), and dulcia (sweetened wine drinks)—and there are other indications that passum continued to be the quintessential 26 As Pliny indicates (Natural History 14. 119), this ruled out Greek wines, which were customarily made from a mixture of mustum and water (most famously sea-water, as in the Coan wines). For a full account of Roman wine-making see Columella 12. 18–45. 27 Columella 12. 19. 2; Pliny, Natural History 14. 80. 28 There are also, according to Pliny, rules about the vine from which the grapes are taken to produce a wine that is ‘fas’, and also about their pressing (Natural History 14. 119). 29 See Pliny, Natural History 14. 89–91; Valerius Maximus 6. 3. 9, and Dionysius of Halicarnassus, Roman Antiquities 2. 25. 6–7; see also Festus 500.9F, s.v. temetum.

214

Festus and the Role of Women in Roman Religion

women’s drink. Of these, only the lorea directly corresponds to what Festus outlaws for divine offerings, though their general inappropriateness as sacrificial material can presumably be deduced from the fact that all involve deviating from the basic wine production process. The mismatch between spurcum vinum and Gellius’ female tipples is significant, however, as it establishes a category of wines (and related drinks) that were not acceptable to either the gods or women, and so rather confounds (or at least weakens) the argument for exclusive symmetry in this respect—that exactly what was prohibited to women was prescribed for the gods. Gellius also fails to make the status of the exceptions he mentions clear in the way that De Cazanove and those that follow his line require. They need something more than simple leniency in the application of an overall interdiction, something more than a pragmatic acceptance that women have got to drink something, and it might as well be thin, sweet, and spiced wines. They need this pattern to be formed by explicit exceptions being made to a general rule, exceptions based on consideration of something like the pontifical regulations on purity. Indeed, it is not just that Gellius fails to make this clear, he actually suggests an alternative interpretation, one that has provided the dominant way of understanding this material. The centrepiece of Gellius’ discussion is his citation of Cato, stating that the punishment for women drinking wine was ‘no less than’ (non minus quam) that for adultery; a point that is picked up by other ancient authors, such as Valerius Maximus (6. 3. 9) who asserts that excessive drinking impairs moral judgement and leads women into all sorts of vices. This then is all part of a wider discourse about female morality and wifely good conduct, about the ethical exemplarity of a perhaps mythical Roman past. It is doubtful whether there was ever an actual rule or ‘law’ about women and wine, but rather a deep-seated expectation, or custom, of female moderation in this respect; an expectation that was enforceable within the workings of the Roman family and state, and was, so the sources claim, actually enforced. The case most often quoted in this context is that of the wife of Egnatius Maetennus, whom he clubbed to death for drinking wine wholesale (e dolio); he was then acquitted of the murder by Romulus.30 Other stories involve a family who starved to death a matrona who stole the keys to the wine cellar; and a wife who was found guilty of drinking more wine than necessary for her health without her husband’s knowledge and fined a sum equivalent to her dowry (Pliny, NH 14.89). These instances are all, therefore, about excessive drinking,

30

Pliny, Natural History 14. 89; Valerius Maximus 6. 3. 9 (who names him Mecennius).

Festus and the Role of Women in Roman Religion

215

drinking that is out of the husband’s control; while the drinks that Aulus Gellius lists as customary for women can be seen as falling on the other side of expectations about moderation. Female indulgence in a bit of raisin wine now and then does not challenge male control of the wine cellar, of household resources, nor lead to adultery. It is, then, not a woman’s relations with the gods at stake here, but her conduct as a wife. The dominant, moralizing, interpretation of these passages has, however, been challenged by G. Piccaluga who argues that several of the customary female tipples in Aulus Gellius are actually more potent than ‘pure wine’, that they are effectively fortified wines, with a high alcohol content, and, therefore, more dangerous to women’s sobriety and chastity than the ordinary stuff.31 So, an alternative explanation of this prohibitive and permissive pattern is necessary, which is where the gods might enter the picture, and an exclusive symmetry with women come in to play. Piccaluga has a point, but not a very convincing one, for though passum and various sweetened wines, could be pretty strong, they could also be very weak, made with a lot of water, or with second pressings, for example.32 Moreover, women were hardly going to be drinking these wines neat, and presumably the more powerful the concoction the more it was watered. This, it should be stressed, was the general practice—only barbarians drank their wine unmixed—and, if as various ancient sources (and modern scholars) suggest, the Roman gods took their wine un-watered (at least on occasion) then that would serve to distinguish them from both Roman women and men.33 So, while Piccaluga may have cast further light on women’s drinking habits, she has not provided compelling reasons to dislodge the ancient discussions of the subject from their moralizing context. These texts seek to establish sets of expectations about, and ideals of, wifely conduct, to conjure up past standards of Roman austerity as exemplary tools; they do not serve to exclude women from sacrificial ritual, to make ‘pure’ wine into something shared only by men and gods. Women’s use of wine in a sacral (not to mention secular) context can also be demonstrated in a more positive manner, through a range of ancient evidence which simply assumes that women offered wine to the 31

Piccaluga 1964: 195–237. See e.g. Pliny, Natural History 14. 82 for poor quality raisin wines. 33 The view that unmixed wine (merum) was offered to the gods is not specifically supported, however, by the excerpt of Festus under discussion here, which prohibits only the introduction of water during the wine production process (as confirmed by Pliny), and the introduction of the term ‘undiluted’ into the English translation of Scheid’s article in this respect is unfortunate and misleading. It is suggested, however, by e.g. Ovid, Amores 3. 10. 47 and Plutarch, Roman Questions 57, both passages being discussed below. 32

216

Festus and the Role of Women in Roman Religion

gods like everybody else. Plutarch’s Quaestiones Romanae can once more be pressed into service. He asks, for example (57): Διὰ τί τῇ Ῥουμίνῃ θύουσαι γάλα κατασπένδουσι τῶν ἱερῶν, οἶνον δ᾽ οὐ προσφέρουσιν; Why do women sacrificing to Rumina pour libations of milk during the rites, but make no use of wine?34

This question, once again, only makes sense if this was an exception, if women generally did pour libations of wine to the gods; and it is answered without any reference to the sex of the sacrificiants. The response revolves entirely around the character of the goddess Rumina whose role as divine wet-nurse makes wine an inappropriate and milk a most appropriate offering in her particular case. It should also be noted that milk is offered to a number of other deities also, by both men and women, in what are generally considered to be a set of especially rustic and archaic rites.35 Festus, at least as excerpted by Paul, takes the religious relationship between women and wine a step further. He offers the following definition (455.14P): Simpulum vas parvulum non dissimile cyatho, quo vinum in sacrificiis libabatur; unde et mulieres rebus divinis deditae simpulatrices. A simpulum is a small vessel, not dissimilar to a cyathus, from which wine used to be poured out as an offering in sacrifices; from which, also, women devoted to divine matters are called simpulatrices.

For these women to have taken their name from this vessel, or indeed for this derivation to have entered the Latin etymological tradition, they must have surely been using it to pour libations in regular ritual settings. Nor, it should be stressed, is the simpulum a marginal or arcane religious artefact, quite the reverse. It appears in the literary sources as a symbol of traditional Roman austerity and piety, retaining its plain earthenware construction and religious centrality, while all around it objects were becoming over-elaborate and corrupt.36 As Varro explains (De lingua Latina 5. 124), while the sophisticated Greek cyathus replaced the simpulum as the dipping ladle of choice at banquets, its rough Roman 34

Cf. Plutarch, Romulus 4. The use of milk in this cult is also confirmed in Varro, De re rustica 2. 11. 5 and Nonius 167L. 35 Milk is, for example, offered to Pales in the Parilia (see e.g. Ovid, Fasti 4. 745–6; Tibullus 1. 1. 36), and to Silvanus (Horace, Epistles 2. 1. 144) and Priapus (Virgil, Eclogues 7. 3. 3). Pliny claims it is Romulean rites that preserve this ancient practice (Natural History 14. 88). 36 See e.g. Pliny, Natural History 35. 158 and Juvenal 6. 343.

Festus and the Role of Women in Roman Religion

217

simplicity persisted in a sacrificial context. The numismatic record similarly displays the simpulum as a symbol of religious authority, and indeed of priestly office. This item is a mainstay of Roman coinage, from that of M. Aemilius Lepidus in 61  to the emperor Decius in  250, and including almost everybody in between, appearing usually in conjunction with other priestly items and sacrificial implements—such as the lituus (priestly wand) and hat, or the secesipita (sacrificial knife) and axe.37 The naming of a group of women ‘dedicated to divine matters’ after such a key symbol of Roman religious tradition and authority, one indissolubly linked with wine, thus challenges many of the restrictions that have been placed on female religious activities at Rome. Before examining the figure of the simpulatrix (and her associates) in more detail, however, one last point of problematic intersection between women, wine, and public cult needs to be dealt with. This concerns the worship of Ceres at Rome, the changing pattern of which initially seems to fit Scheid’s model especially well; in the third century , a new ‘Greek’ version of her cult was introduced to Rome which, as H. Le Bonniec has suggested (in the most substantial treatment of the subject to date), included amongst its distinctive features both a prominent role for women and a prohibition on the use of wine.38 On closer scrutiny, however, the picture becomes more complicated, though enlightening none the less. In particular, developments surrounding Ceres serve to put the spotlight on the part of Scheid’s argument that depends on the ‘foreignness’ of various cults in which women officiated (including at sacrifices) to protect his rule of female sacrificial incapacity. ‘Foreign’ status—the fact that though formally adopted into the religious programme of the Roman state these cults remained distinct, retaining the mark of their non-Roman origins, and thus providing a contrast with ‘indigenous’ religious practices—preserves the exceptionality of their sacrificing women, their reversal of ‘normality’. But does it?

GREEKS AND WOMEN: THE CULT OF CERES AT ROME Paul’s abridgement of Festus is certainly typical in emphasizing the Greek (in the broadest sense of the word) origins of a particular set of female festivities for Ceres: 37 For examples of these early and later coinages see e.g. RRC 419 (M. Lepidus); HCC III p. 249 (Decius). 38 Le Bonniec 1958: Spaeth 1996.

218

Festus and the Role of Women in Roman Religion

Graeca Sacra festa Cereris ex Graecia translata, quae ob inventionem Proserpinae matronae colebant. Quae sacra, dum non essent matronae, quae facerent propter cladem Cannensem et frequentiam lugentium, institutum est, ne amplius centum diebus lugeretur. The Greek festive rites of Ceres, which matronae celebrated on account of the finding of Proserpina, were brought over from (Magna) Graecia. Since there were no matronae to celebrate these rites because of the slaughter at Cannae and the great number of mourners, it was decreed that mourning should not last more than a hundred days.39

It is also from this story about the special measures taken in relation to these matronal rites in 216 , a story variously recounted by Livy, Valerius Maximus, and Plutarch, that this Greek import is dated.40 This is the first appearance in the historical record of such sacra, and a passage from Arnobius (Adv. Nat. 2. 7. 3) indicates that in 216  not much time had elapsed since their introduction. Prior to this Ceres had (often together with Tellus) been involved in a range of rustic religious practices in Rome and its environs, as well as occupying an important place—together with Liber and Libera—in Roman state cult. A temple to this divine triad had, according to Dionysius of Halicarnassus, been vowed, built, and finally dedicated (following consultation of the Sibylline Books) in the first decade of the fifth century , and became associated with the festival of the Cerialia on April 19.41 These rituals (and perhaps others glimpsed in the sources too), provide the implicitly Roman point of contrast with the distinctively Graeca sacra of Paul and Festus. However, their Romanness may actually have been constructed through this contrast, and the matter of simple historical precedence, as much as anything else, since ‘Hellenizing’ elements are also discernible in the temple of Ceres, Liber, and Libera, and in aspects of its associated cult (as the Sibylline connection would also suggest).42 The other point of distinction is the exclusively female character of the new sacra, a point that is underlined by references in a number of literary sources and inscriptions commemorating female officiants—sacerdotes— who led the matronae in their sacred activities.43 This broke with the 39 86.7–11P; cf. 268.27F s.v. peregrina sacra. That ‘Graecia’ (literally, ‘lands lived in by Greeks’) should be understood in this Italian (and Sicilian) way, will become increasingly clear. 40 Livy 22. 56. 4–5 and 34. 6. 15; Valerius Maximus 1. 1. 15; Plutarch, Fabius Maximus 18. 1–2. The accounts vary both on the question of whether the rituals were in fact celebrated in 216, and on the number of days to which mourning was henceforth limited. 41 Dionysius of Halicarnassus 6. 17. 2–4 and 6. 94. 3; and see also Le Bonniec 1958: 213–378. 42 See e.g. Pliny, Natural History 35. 154 for the Greek elements of the temple. 43 See e.g. Cicero, Pro Balbo 55; Valerius Maximus 1. 1. 1; CIL 6. 2181 and 2182.

Festus and the Role of Women in Roman Religion

219

previously inclusive nature of Ceres’ main cult practices, conducted under the male auspices of the flamen Cerialis and plebeian aediles. It should be noted, however, that there is some evidence suggesting the presence of allfemale ‘mysteries’ of Ceres at an earlier (as well as later) date, and it is also worth emphasizing that a number of other ritual activities—mainly expiatory—involving Ceres, Proserpina, and groups of women appear on the Roman religious scene not long after the matronal sacra.44 The picture is a little more complicated, then, but still basically contrasting. What Paul’s excerpt makes no mention of, though, is any prohibition of wine in these new Greek rites, and this is perhaps the weakest part of Le Bonniec’s argument in this respect, as he tries to reconcile the conflicting and confused ancient evidence about the place of wine in the worship of Ceres by proposing that its interdiction was another particularity of the Graeca sacra which contrasted with practices in other areas of her cult. The ancient debate about the propriety of offering wine to Ceres is best illustrated by Servius who, in his commentary on the Georgics (1. 344), sharply criticizes those who have accused Virgil of contravening religion by suggesting in the first poem that the seasonal libations to Ceres should include wine. And this accusation is, indeed, levelled at Virgil, in almost precisely these terms, by one of Macrobius’ symposiasts in his Saturnalia (3. 11). Servius retorts, however, that ‘the pontifical books make no such prohibition’ (pontificales namque hoc non vetant libri), nor can a line from Plautus’ Aulularia, in which the absence of wine from the play’s wedding preparations produce protests that the event is turning into a ‘marriage of Ceres’ (Cereris . . . nuptias) be used to prove the point.45 The sacrum of Ceres, Servius asserts, is not the same as her marriage, which is properly called the ‘wedding of Orcus’ (Orci nuptiae) and solemnly celebrated by the pontifices in a ritual without wine. The clear implication here is that this ban was unique, in all other aspects of Ceres’ cult the wine flowed freely. Le Bonniec, however, wants to extend this interdiction beyond the rather enigmatic Orci nuptiae to encompass the more important matronal rites, while concurring with Servius (not to mention Virgil and many other Latin writers) about the general presence of wine in Ceres’ cult activities at Rome.46 He relies on two pieces of evidence to support this view.

44 On these mysteries see Wagenvoort 1980: 114–46; which revises the conclusions he reached in Wagenvoort 1956: 150–68. On the expiations, see the summary in MacBain 1982: 127–35. 45 Plautus, Aulularia 354, which is indeed cited by Macrobius’ symposiast (Saturnalia 3. 11. 2). 46 Le Bonniec 1958: 416–17 and 438–41.

220

Festus and the Role of Women in Roman Religion

The first comes from Dionysius of Halicarnassus. In his attempt to provide mythic Arcadian origins for all Roman institutions, the historian credits Evander and his heroic band of early settlers with the foundation of a temple to Demeter on what would become the Palatine, together with nêphalia (wineless or sober sacrifices, generally based on honey) performed by women according to the Greek custom, which, he claims, remained unchanged in his own (Augustan) times.47 There are several aspects of Dionysius’ testimony that fail to inspire historical confidence, even putting aside his Arcadian agenda. No temple of Demeter has ever been associated with the Palatine, for example, nor was the conjunction of women and nêphalia characteristic of Greek cults of the goddess. The most famous and wide-spread all-female ritual of Demeter was the Thesmophoria; but, though (at least in its Attic form) this incorporated a day of general fasting, of abstention from all food and drink, including wine (before the great feast that followed!), there is no suggestion that this latter was particularly prohibited, nor that wine offerings to the goddess were banned.48 Conversely, while no wine was to be poured on the altar of the Despoinae (the Mistresses—Demeter and Persephone) at Olympia, for example, it was both men and women who were so enjoined, and a similar inclusiveness characterizes other instances of this rule, whether in relation to Demeter or other deities.49 Despite all these problems, Le Bonniec assumes that Dionysius is inventing an Arcadian aetiology for an actually existing, and accurately described, Augustan practice, and identifies the contemporary ceremony as Ceres’ Graeca sacra festa, so adding sobriety to its other features.50 Even if the assumption of Augustan ritual realism is accepted, however, it is still not clear that the matronal sacra provide the best fit. H. Wagenvoort has suggested, for example, that it is in fact the Orci nuptiae which are referred to here—known to be wineless, though not exclusively female—or at least that this passage should be taken as supporting the long-term existence of a distinct set of female mysteries 47

Dionysius of Halicarnassus, Roman Antiquities 1. 3. 1; and see e.g. Plutarch, Quaestiones. Conviviales 4. 6. 2 for a discussion of the nature of nêphalia. 48 On the Thesmophoria see e.g. Burkert 1985: 242–6. The smuggling of the wineskin into the festival in Aristophanes’ play (Thesmophoriazousai 733–64), only goes to demonstrate the point. It is the general rule of fasting that is broken, and in such a way as to emphasize one of the female vices under discussion—drunkenness. It seems likely that a similar day was included in other Greek Thesmophoria, though definite references are lacking. 49 On Olympia (where wine is additionally not to be poured on the altar of the Nymphs and of all the Gods) see Pausanias 5. 15. 10; and a similar prohibition appears in the same work at e.g. 1. 26. 5 (referring to the Athenian altar of Zeus Hypatos). 50 Le Bonniec 1958: 416.

Festus and the Role of Women in Roman Religion

221

of Ceres which meet all of Dionysius’ criteria.51 Another possibility would be to cross-reference his statement with the only other specific appearance of nêphalia offered to Demeter at Rome in the ancient sources—that is in the first Sibylline Oracle preserved by Phlegon of Tralles.52 The text is a very troubled one, but it details a complex set of expiatory rites to be used in response to the birth of a hermaphrodite.53 Amongst other rituals (such as the presentation of a treasury and the sacrifice of oxen) performed by different groups of women to Demeter, Persephone, Pluto, Hera (and possibly Zeus), they seem to include wineless libations poured onto the fire by sacrificially knowledgeable old women.54 Though a group of youths make a late appearance in proceedings, it does seem that women, sobriety, and Greekness are conjoined here, but without the implicit regularity of Dionysius’ delineation. The safest conclusion to draw from this plethora of possibilities and from the problems inherent in Dionysius’ testimony is simply that it cannot be used to support the view that wine was prohibited from the Graeca sacra of Ceres referred to by Paul and others. Moreover, the two clear cases of rituals without wine seem to stand firmly on either side of the key distinctions that Le Bonniec and others understand to be at work here. The Orci nuptiae look to belong to the earlier ‘Roman’ (or at least more naturalized) and inclusive (or, more precisely, male-led) layers of Ceres’ cult, while the expiatory rituals of the Sibylline Oracle are later, ‘Greek’ introductions, and female-dominated. Nor has an investigation into the alleged Greek roots of Dionysius’ nêphalia helped the cause. It is useful, however, in clarifying the meaning of sober offerings and their place in the repertory of classical sacrifice more generally. Orcus’ Wedding and the Sibylline procedures may not be located similarly in relation to women’s participation, but they do share a certain chthonic orientation, which can also be discerned in a number of other Greek cult practices in which wineless libations feature. The chthonic theme is not, it should be stressed, a universal one in such cases. Nêphalia fit into a far more complex pattern than that; a complexity that is underlined by the way in which they are often combined with other sacrificial 51

Wagenvoort 1980: esp. 136–41. Phlegon, Miracula 10; my thanks to John North for bringing the relevance of this text to my attention. 53 The best edition to date is that of Diels 1890. The dating of this oracle is controversial: Phlegon places it in 125 , but Diels wants to backdate it to the first two androgyne prodigies referred to in the historical sources (207 and 200 ). The matter is also helpfully discussed in MacBain 1982: 127–35, where 133  is preferred. 54 Phlegon, Miracula 10 (Diels, lines 22–3): Τρὶς τόσα, νήφαλα πάντα, πυρὸς μαλεροῖο τιθέντων / Ὅσασαι ἐπισταμένως θυσίαν γραῖαι προτίθενται; cf. e.g. Cicero, Verrine 2. 4. 99. 52

222

Festus and the Role of Women in Roman Religion

elements and events, not just various blood sacrifices, but also libations of wine. So, for example, in the epigraphically preserved sacrificial calendar from the Attic deme of Erchia, nêphalia appear several times, linked in particular to holocausts and to Zeus Meilichios.55 The rites for this deity are particularly complicated in their specification of the wine and wineless offerings that are to accompany the different stages of the sacrifice—wine was not to be used until the victim had been killed, dismembered, and was being partially offered to the gods, partially consumed by the worshippers, when wine became an important part of proceedings.56 It is erroneous, therefore, not only to suggest that the requirement of sober offerings is sexually determined, but also to think in terms of a blanket ban, a general interdiction on wine, in this context. Nêphalia seem more of a positive choice, selected to give special emphasis to a ritual action or sequence, and often operating within a broader overall package that may indeed include wine. So, sex may not be the issue here; but given that, as Paul states, the Graeca sacra of Ceres were celebrated, ‘on account of the finding of Proserpina’, thus establishing a chthonic connection of some sort, it is entirely possible that nêphalia were offered during these festivities, which would help Dionysius but not Le Bonniec, who wants a complete prohibition on wine in these proceedings. However, it is now time to admit Le Bonniec’s second witness, Ovid, who provides one of the few contemporary references to the actual contents of these sacra, though, as is to be expected, it is very far from a straightforward description of events. In his Amores, the poet complains about the effects of the annual rite of Ceres on his sex-life, picking up on the aspect of these festivities that seem to have received the most attention: the sexual abstinence required of its female participants.57 As the ritual demands, his girlfriend sleeps alone, and Ovid laments the separation, arguing that the festival should celebrate the goddess’ generosity, fecundity, and capacity for love, not her misery. It should mark her joy at finding her daughter, not the

55 LSCG 18. Nêphalia appear in lines B. 16–20, Γ. 20–35, Δ. 20–3, E. 12–15 (holocausts); A. 40–3 (Zeus Meilichios); and Γ. 48–53, Δ. 41–6 and E. 60–4. 56 The procedure is helpfully discussed by Jameson 1965: 158–72, esp. 162–5. He also draws parallels with a complex sacrifice for Zeus Polieus on Cos. 57 Ovid, Amores 3. 10; and see also his Metamorphoses 10. 431–6, which, though ostensibly located in Cyprus, certainly has some Roman roots. Festus also makes reference to the ‘castus Cereris’ as a time during which private mourning is remitted (144.3F minuitur populo luctus; see also Moreau 2007: 69–86), which is most likely a reference to the Graeca sacra (and there is an inscription very dubiously restored [C]ereres ca[stu]: CIL 12 331). Ceres and chastity are also connected more widely.

Festus and the Role of Women in Roman Religion

223

loneliness of her loss. His entreaties culminate in the proclamation that (Ovid, Amores 3. 10. 47–8): Festa dies Veneremque vocat cantusque merumque Haec decet ad dominos munera ferre deos. A festive day calls for sex, singing and unmixed wine; / these gifts are fitting offerings to the gods, our masters.

For Le Bonniec, the meaning of this text is clear: Ovid reproaches the goddess for proscribing wine just as she proscribes love, forbidding wine from being both consumed and offered by those celebrating her rites.58 There is, however, little to suggest that this couplet should be taken in anything other than a general sense, as a statement of the poet’s idea of a good time, of the principle that festivals dedicated to the divinities should be fun, that fun consisting in the proverbial wine, women, and song. And it should be noted that Le Bonniec is, given his interpretative approach, presumably committed to Ceres’ Graeca sacra being marked also by silence, or at least a ban on singing. The implausibility of this point perhaps explains his failure to mention it. Ovid, like Dionysius, is not really up to the task Le Bonniec sets him. It remains possible either that, as in the Thesmophoria, elements of abstention from food and drink as well as sex were involved in the Graeca sacra at Rome, or that nêphalia were implicated at some point (or indeed both). But neither of these should be mistaken for a complete prohibition on wine, or linked to the exclusively female nature of the rites. This latter aspect now requires some more focused attention, particularly (to return to Scheid’s arguments) the question of how the issues of female religious authority or representivity, and Greekness, interrelate. As already mentioned, despite considerable uncertainty about what actually went on in the rituals of Ceres’ Graeca sacra, it is clear that they were all female, both officiated over and participated in by women alone.59 The fullest account of these female officials—sacerdotes—is provided by Cicero in his defence of L. Cornelius Balbus. ‘Our forefathers’ (nostri maiores), Cicero explains, ‘introduced these priestesses 58

Le Bonniec 1958: 416. Though how seriously the usual specification of the participants as matronae—the wives or widows of citizens—should be taken is more debatable. As Spaeth points out (1996: 108–10), Valerius Maximus’ formulations (1. 1. 15) certainly suggest that young, unmarried, women also participated; as they seem to have done also in the cult of Demeter at Catena in Sicily (Cicero, Verrine 2. 4. 99). More information about festivals to Demeter and Persephone (Thesmophoria) in Magna Graecia would, of course, help considerably with interpreting the Roman activities, but is unfortunately not forthcoming (Strabo 5. 4. 4, for example, is hopelessly vague). 59

224

Festus and the Role of Women in Roman Religion

from Greece along with the rites themselves, to ensure their proper and pious performance in their new home’ (Pro Balbo 55): Sed cum illam, quae Graecum illud sacrum monstraret et faceret, ex Graecia deligerent, tamen sacra pro civibus civem facere voluerunt, ut deos immortales scientia peregrina et externa, mente domestica et civili precaretur. has sacerdotes video fere aut Neapolitanas aut Velienses fuisse, foederatarum sine dubio civitatum. But, though they selected the woman who was to teach and conduct this Greek rite from Graecia, they still wanted her to conduct the rites for the citizens as a citizen, so that she might supplicate the immortal gods with foreign and outside knowledge, but a domestic and civic mind. I understand that these priestesses were, in general, either Neapolitans or Velians, certainly women of federated cities.

Cicero then cites the case of one such—Calliphana—a Velian who was made a citizen by the senate and people of Rome in 93 .60 Both aspects—that the women were drawn from Magna Graecia and performed their ritual function ‘on behalf of the Roman people’, as Roman citizens—are confirmed elsewhere, by other literary sources and epigraphy. Two funerary inscriptions from Rome commemorate Casponia and Favonia, both public priestesses of Ceres (sacerdotes Cereris publicae populi Romani), and both the daughters of Roman citizens, with the former also a Sicilian.61 Plenty more public priestesses of Ceres can be found in the epigraphy of various Italian cities south of Rome, indicating the pool from which Rome could draw.62 It is also clear, again despite the uncertainty of detail, that these women were sacrificing on behalf of the Roman people. What kind of sacrifices were offered— whether indeed blood sacrifice was included or not—remains debatable; but both the patterns of the Greek Thesmophoria and Roman references (not to mention general ritual practice) indicate the sacrificial, and probably bloody, content of these ceremonies.63 So here, undoubtedly, are women exercising religious authority, interceding with the gods on behalf of the whole community of Rome, representing ‘the citizens as a citizen’, but also marked out as Greek, in possession of ‘outside knowledge’. The question is, however, the way in which the relationship 60

The same story is told, somewhat confusedly, by Valerius Maximus (1. 1. 1). CIL 6. 2181 and 2182. 62 CIL 9. 4200 (Amiternum?); 10. 812 and 1074 (Pompeii); 10. 4793 and 4794 (Teanum Sidicinum). 63 Livy 34. 6. 15 refers to Ceres’ sacrificium rather than sacra, for example, and see also Plutarch, Fabius Maximus 18. 1–2 in which sacrifices and a procession (also mentioned e.g. by Ovid, Metamorphoses 10. 431–6) are specified. 61

Festus and the Role of Women in Roman Religion

225

between two of these elements—the foreign and the female—is to be understood in respect to their authority and sacrificial activity. For Cicero, certainly, sex is not an issue. The priestesses come simply as part of the ritual package. As Festus outlines in his definition of peregrina sacra, these are cults, such as that of the Magna Mater brought from Phrygia, Ceres from Graecia, and Aesculapius from Epidaurus, which have been settled in Rome either through evocatio, in the context of war, or on account of certain religious considerations in peacetime, and ‘which are celebrated according to the customs of the peoples from whom they were received’ (quae coluntur eorum more a quibus sunt accepta).64 This means that they come with their own officials, of whatever sex or gender, so that the mos of the providing people may be properly followed. In the case of Ceres, the problem, as Cicero sees it, is the integration of that package into the Roman religious landscape given that the senate wanted to locate it in the public sphere, to make this Greek cult into a state cult in the fullest sense. Its officiants needed to be citizens, and so these they became. It is then on this basis—on the basis of the constitutional procedure used to bring the cult over and place it at Rome (properly carried out) and the citizenship of its officiants (properly granted or inherited)—that these women are entitled to communicate with the gods on behalf of the whole Roman community. They have been aligned, therefore, with the other women in a similar position at Rome— primarily the Vestals, but also the regina sacrorum and the flaminicae— and distinguished from the priesthoods (male and female) of other imported cults, such as those of Aesculapius and the Magna Mater, who do not attain this kind of status. They do indeed tread the boundary between Roman and non-Roman, a boundary which they effectively straddle, becoming Roman but retaining their Greek knowledge; but the boundary between men and women, either in a religious or any other sense, is not in the frame at all, not even implicitly. The contrast with the fate of the cult of the Magna Mater, fellow peregrina sacra after all, is particularly instructive in this respect. It demonstrates how, if the relationship between gender and religious authority or practice in the Graeca sacra of Ceres had been considered a contradictory or transgressive one, the Roman state would have taken action, would not have just let the matter lie. For here it seems that the Phrygian mos was problematic, including (or perhaps especially) as it related to the Roman gender system, so Roman citizens were kept

64 268.27F s.v. peregrina sacra; see also Scheid 1995: 15–31, for more discussion of the peregrina sacra within the broader context of ‘foreign’ cults.

226

Festus and the Role of Women in Roman Religion

separate from the Phrygian rites of the goddess.65 These were left to a Phrygian priest and priestess together with their non-Roman followers (the Galli), while Roman citizens participated only in the Megalensian Games, held in honour of the Magna Mater, but presided over by the praetors. Nor can it be argued that Cicero’s focus is simply a product of the particularities of the case: that he is hardly going to raise problems of women’s religious role in his defence of Balbus’ Roman citizenship.66 For surely, if Ceres’ priestesses did present these kinds of difficulties, if they did strike at the heart of a fundamental tenet of Roman religious practice—female sacrificial incapacity—protected only by their Greekness, then it seems unlikely that Cicero would have recruited them to Balbus’ cause in the first place. His precise formulations and the emphasis they tacitly (but tantalizingly) put on the rights of women as citizens, the content and importance implicitly given to female citizenship, must be treated with caution; but Cicero’s willingness (indeed keenness) to align Calliphana and Balbus, his assumption that their respective femaleness and maleness is irrelevant, should be taken seriously. It would seem, therefore, that the sacerdos Cereris addressed the gods on exactly the same basis as any other public religious official, her authority came from her own, properly arrived at, position in the Roman constitution; but she communicated in, as it were, a different language, in the Greek mode rather than the Roman one. So, in the end, Ceres’ new, Greek, cult speaks rather positively of women’s religious roles at Rome, at least from the third century  onwards, and tells against notions of prohibition, incapacity, and marginality. It is time now to collect together some of these more positive signs—the women who have been found to play a more active part in Roman cult and ritual—to bring this discussion to a close.

SOME (MORE CONSTRUCTIVE) CONCLUSIONS Before reaching the conclusions proper, there is one more piece of ancient evidence concerning Roman women and sacrifice that needs to be discussed. This is a sentence from Varro’s De lingua Latina (another

65

Dionysius of Halicarnassus, Roman Antiquities 2. 19. 3–5. Balbus’ citizenship, granted to him by Pompey in his native Spain, and ratified by the lex Gellia Cornelia in 72 , was legally challenged by his political enemies in 56, without success. 66

Festus and the Role of Women in Roman Religion

227

work that relates closely to Festus’ Lexicon), which appears in a sequence dealing with the terminology of clothing: Sic rica ab ritu, quod Romano ritu sacrificium feminae cum faciunt, capita velant. Thus rica (a small veil or mantlet) from ‘ritus’, because in the Roman rite, when women sacrifice they cover their heads.67

At first glance this etymology, however mistaken, seems to seal the case: Roman women did sacrifice Roman style, with their heads covered like their men-folk, but with a special garment not just a fold of the toga.68 It might be possible, however, to interpret this statement more narrowly: to understand feminae, not as all women, or women in general, but as only those women who did sacrifice, according to the Roman rite, and who covered their heads in this way. This more restrictive reading would seem, certainly, to be supported by the descriptions of the rica contained in the remains of Festus’ Lexicon. Rather confusingly this item of clothing appears twice in the surviving text. First in the plural, and together with riculae, both names for small ricinia (or recinia), ‘that is a mantlet made for use on the head’ (ut palliola ad usum capitis facta).69 The entry continues with a more specific definition borrowed from Granius: the rica is a female head-binding worn by the flaminica in place of the vitta (woollen bands used to bind the hair).70 The second occurrence of the rica, this time singular and alone, provides a still more elaborate and specific description (even as abbreviated by Paul): Rica est vestimentum quadratum, fimbriatum, purpureum, quo flaminicae pro palliolo utebantur. Alii dicunt, quod ex lana fiat sucida alba, quod conficiunt virgines ingenuae, patrimae, matrimae, cives, et inficiatur caeruleo colore. The rica is a four-cornered garment, fringed and purple, which flaminicae are accustomed to use in place of a mantlet. Others say that it comes from fresh, white wool, that freeborn citizen girls, with both father and mother living, manufacture it, and it is then dyed with azure dye.71

67 Varro, De lingua Latina 5. 130; and on relations between Festus and Varro see Glinister 2007: 11–32; Llommé 2007: 33–47. 68 This passage is so interpreted, for example, by Staples 1998: 186 n. 93. 69 342. 27F s.v. ricae. The recinium is defined just above (342. 20F), as a four-cornered, toga-like garment of considerable antiquity; and see also the further discussion in Sebesta 1994: 46–53, esp. 50. 70 342. 28F ricae; the vitta is also discussed by Sebesta 1994: 49. 71 369. 1P s.v. rica, cf. the Festan fragments at 368. 3F s.v. rica. The flaminicae themselves, of course, were subject to similar qualifications.

228

Festus and the Role of Women in Roman Religion

Varro’s sacrificing feminae would, therefore, seem to be the flaminicae, already known to sacrifice, Roman style. The issue is somewhat confused by the fact that Festus, mostly via Paul, is a key informant about another item of clothing characteristic of the flaminica: the flammeum, a bright yellow garment also described (like the rica) as a ‘veil’, especially when covering the head of its other wearer—the Roman bride.72 And yet another item of female sacrificial headgear is variously mentioned by other authors. Or, perhaps items, as the description of the arculum provided by Servius (ad Aeneid 4. 137)— that it is a switch from a pomegranate tree, bent round, with the ends tied together with white woollen bindings, to form something like a crown, and worn on the head by the regina (sacrorum) ‘in certain sacrifices’ (in sacrificiis certis), while the flaminica Dialis has to use it ‘in all sacrifices’ (omni sacrificatione)—seems to be divided between two terms by Paul, and is only lightly alluded to by Aulus Gellius. The arculum, according to Paul, is a circlet placed round the head in order, conveniently, to hold upright the vessels which are carried on that part of the body ‘in public rituals’ (ad publicas sacras) (though by whom is unclear); while the ‘inarculum’ is a pomegranate switch bent round and worn on the head of the sacrificing regina sacrorum.73 Gellius simply refers briefly, in his listing of the various ritual prohibitions and compulsions that bound the flamen and flaminica Dialis, to the latter wearing a twig from a fruitful tree in her rica (NA 10.15.28). If the line of Boels is followed, and all these switches and twigs are the same, and the flammeum is considered an essentially non-sacrificial garment; then it would seem that the flaminica Dialis always wore both veil and crown when sacrificing (presumably carrying vessels on her head), while the regina sacrorum sometimes wore the crown (and presumably covered her head with some item of clothing also, though the rica is never specifically named in conjunction with her).74 It is then also worth noting that the Vestal Virgins were marked by different headgear again; and, ‘when they sacrifice’ (cum sacrificant) wear the suffibulum—white with a purple border, four-cornered and not long—on their heads, fastened with a fibula.75

72 On the flammeum, including the Festus/Paul references, see La Follette 1994: 54–64. Paul specifies the flaminica Dialis, but in such a way as to raise doubts as to whether Festus had. 73 15. 6P s.v. arculum and 101. 5P s.v. inarculum respectively. 74 Boels 1973: 77–100, esp. 82–6. 75 474. 3F s.v. suffibulum; see also La Follette 1994: 57–60.

Festus and the Role of Women in Roman Religion

229

So, either Varro’s etymology is supposed to derive from the sacrificial activities of only some of the women who sacrifice, or, in an attempt to return to a more generalized interpretation of Varro, rica could be used as the more inclusive category—encompassing all kinds of female sacrificial headgear. Certainly that would be the easiest way to read the antiquarian’s formulation—that all sacrificial head coverings worn by women can be called ‘rica’ from ‘ritus’—which would again extend women’s ritual possibilities. Festus (and Paul) would, however, tend to refocus matters on established figures of female religious authority— especially the flaminicae—though also highlighting their sacrificial function. So, while Varro has again added to the richness of Roman religious activity for women, has opened up, indeed, a whole sacrificial typology according to what is worn on the head—he does not himself, it must be admitted, deliver the knock-out blow to the notion of female sacrificial incapacity. Fortunately, there is no need. This notion is already down and out, repeatedly contradicted by the sources, including Festus, whose Lexicon has helped broaden, in a range of ways, the scope of female religious activity at Rome. Despite its problematic preservation, this text supports a more positive understanding of women’s ritual roles, including in sacrifice, and it is worth revisiting, briefly, some of the female religious figures who have emerged from under the shadow of the major players— such as the Vestals, flaminicae, and sacerdotes Cereris—by way of real conclusion. That is, most particularly, the simpulatrices—‘women devoted to divine matters’—named, so Paul claimed, from the simpulum, that ongoing symbol of Roman piety and priestly function. For these women appear (or at least lead) elsewhere in the Lexicon too, in a context which aligns them with other groups of ritually active, and qualified, women. Another female sacerdos mentioned by Festus is the piatrix, ‘who was accustomed to perform expiations’ (quae expiare erat solita), and was also called, by some, simulatrix, by others, saga or expiatrix.76 Whether or not the first alternative should be amended to ‘simpulatrix’ (as the earlier editor Müller did, but not Lindsay), the location of these women in the realm of expiation is a resonant one.77 A considerable 76 232. 33F s.v. piatrix; other sacred women mentioned in Festus’ Lexicon but not discussed here include the enigmatic (but definitely sacrificing) Salian Virgins (439. 18F saliae); the flaminia (the sacerdotula who assists the flaminica Dialis: 82. 23P flaminia); and the damiatrix (priestess of the Bona Dea: 60. 1P damium). 77 The word simulatrix is applied to Circe by Statius (Thebaid 5. 551), bringing with it notions of both transformation and deceit, and making it, in my view, less likely than simpulatrix, but (given the negative connotations that can be placed on saga) not impossible.

230

Festus and the Role of Women in Roman Religion

number of expiatory rituals involving women have, after all, been encountered in the course of this discussion (and there are more in the Roman repertory). These rituals would generally have been conducted under the auspices of the decemviri—the keepers of the Sibylline Books—though other priestly colleges may also have been involved.78 However, as well as the various female groupings called into ritual action, as such—as a group of twenty-seven virgins, for example—the Sibyl, at least, specifies procedures to be performed by female religious experts: ‘sacrificially knowledgeable old women’ to be precise.79 Which would tie in most closely with the saga, also defined separately by Paul as ‘a woman experienced in sacred matters’ (mulier perita sacrorum).80 For, though these old women did pour libations, given their Greek connections and the wineless nature of the offerings, they are unlikely to have done so from a simpulum; quite apart from the problems of etymology and spelling. A number of looseends and uncertainties remain, therefore, but it is clear that there were groups of female religious experts—sacerdotes even—associated with rituals of expiation, and who serve, like the female worshippers of Hercules and Silvanus, like the women pouring libations of milk to Rumina and all the rest, to expand women’s role in Roman religion well beyond its presently accepted boundaries. This expansion certainly has limits, though, since so much has been omitted from this discussion—both in terms of female festivals and in terms of developments outside the scope of Festus’ antiquarian interests—they have not yet been reached. None the less, this essay does not form part of an argument for religious equality at Rome; rather it has attempted to clear the space for a better understanding of women’s subordination in the religious sphere, both quantitatively and qualitatively. A view of the religious roles of the Roman woman that is dominated by her imagined sacrificial incapacity—that takes her alleged exclusion from the main elements of sacrificial ritual as its starting point, as its main interpretative matrix—must now be replaced, perhaps not by its inverse, by a view that is dominated by female sacrificial capacity and inclusion, but certainly by a view that incorporates these incontrovertible features of the Roman religious landscape. And that does actually indicate a surprisingly egalitarian substratum to Roman 78 The haruspices were, of course, responsible for actually getting rid of the monstrum, and the curule aediles and pontiffs, as well as the decemviri (and the senate), all appear, for example, in Livy’s account of the androgyne prodigy of 207  (27. 37. 4–15). 79 Phlegon, Miracula 10. 22–3 (Diels) (cf. n. 53, above). 80 427.3P saga; cf. the Festan fragments at 426. 14–18F.

Festus and the Role of Women in Roman Religion

231

religion. The underlying assumption in Roman cosmology does seem to be that everybody could address the gods, everybody could sacrifice, that the relation with the divine, at a basic level, was a universal; but this primordial, fundamental, evenness, was overlaid by the formation of human societies, particular peoples (like the Romans), who organized themselves and their collective relations with the gods in particular, interconnected, ways. Despite all the differentiation that occurred, however, the order that was brought to Rome’s communal relations with the divine, a principle of inclusion was maintained, along with a presumption of religious capacity. Exclusion and incapacity had to be actively imposed—the lictor had to shout ‘exesto’ at certain rituals; and Cato (De agricultura 143.1), to prohibit unauthorized religious activity by the vilica of his estate, had to incapacitate her—they did not come naturally. Moreover, though the communal religious authority exercised by women was restricted in that it did not extend into the more strictly political and military realms, but stayed close to the dedicatedly sacred rhythms of the city, that authority was still exercised on the same basis as that of men. That common basis was constitutionality—that the proper procedures had been followed in the acquisition of that authority, and that they continued to be followed in the performance of that role. It is not that these women have crossed into the male domain, but that the Roman constitution was as capable of placing women in positions of religious power as men, as capable of making women religious representatives of the community as men (and that they are essentially sacrificing representatives has been underlined by almost every reference to priestesses in Festus). So, also, what is crucial in this respect about the sacerdos Cereris is not that she is foreign, but that she is a Roman citizen. Her foreignness, like the femaleness of the Vestals, may well be linked to the detailed content of her role and function; but her authority and representational qualities are grounded elsewhere, in the very heart of the Roman system itself.

A FT E R W O R D This essay come out at a time when the scholarly tide was turning on questions of women’s involvement and authority in Roman religion (at least in the anglophone world). Importantly, Celia Schultz had just published Women’s Religious Activity in the Roman Republic (2006)—too hot off the press to be anything more than noted in my discussion—and the first of Emily Hemelrijk’s articles on priestesses of the imperial cult in the Latin West had appeared in L’Antiquité Classique 74 (2005), with more to follow. Hemelrijk also (and independently)

232

Festus and the Role of Women in Roman Religion

came to the same conclusions as argued for here about women and sacrifice in her contribution to a volume on ritual dynamics and religious change in the Roman Empire (2009), while Valérie Huet found visual evidence to support a range of female sacrificial roles in her survey of Roman sacrificial reliefs containing a female presence (2008). Meghan DiLuzio has since (2017) devoted a whole monograph to the role of priestesses in the civic cult of Republican Rome, emphasising their centrality to key public rituals and institutions. The whole business of ancient Roman (and Greek) sacrifice itself has also come under renewed scholarly scrutiny (e.g. Faraone and Naiden (eds.) 2012; Schultz 2016), and developments on this side of the discussion could add further depth and nuance to the arguments advanced in this essay.

B I B L I OG R A P H Y Beard, M. (1980). ‘The Sexual Status of Vestal Virgins’, JRS 70, 12–27. Beard, M. (1995). ‘Re-reading (Vestal) virginity’, in R. Hawley and B. Levick (eds.), Women in Antiquity: New Assessments. London, 166–77, at 169–70. Beard, M., J. A. North, and S. Price (1998). Religions of Rome 1. Cambridge. Boels, N. (1973). ‘Le statut religieux de la “Flaminica Dialis’ ”. RÉL 51: 77–100, esp. 82–6. Brouwer, H. (1989). Bona Dea: The Sources and a Description of the Cult. Leiden—New York. Burkert, W. (1985). Greek Religion: Archaic and Classical (trans. Raffan). Oxford. Cazanove, O. de (1987). ‘Exesto: L’incapacité sacrificielle des femmes à Rome (à propos de Plutarque Quaest. Rom. 85)’. Phoenix 41: 159–73. Detienne, M., J.-P. Vernant, et al. (1989). The Cuisine of Sacrifice (trans. P. Wissing). Chicago. DiLuzio, M. J. (2017). A Place at the Altar: Priestesses in Republican Rome. Princeton. Dorcey, P. (1989). ‘The role of women in the cult of Silvanus’. Numen 36: 143–55. Dorcey, P. (1992). The Cult of Silvanus: A Study in Roman Folk Religion. Leiden. Faraone, C. A. and F. S. Naiden, eds. (2012). Greek and Roman Animal Sacrifice. Cambridge. Glinister, F. (2007). ‘Constructing the past’, in Glinister and Woods, 2007, 11–32. Glinister, F. and C. Woods, eds. (2007). Verrius, Festus & Paul: Lexicography, Scholarship, & Society. BICS Supplement 93. London. Hemelrijk, E. A. (2005). ‘Priestesses of the imperial cult in the Latin West: titles and function’. L’Antiquité Classique 74: 137–70. Hemelrijk, E. A. (2007). ‘Local empresses: priestesses of the imperial cult in the cities of the Latin West’. Phoenix 61: 318–49. Hemelrijk, E. A. (2009) ‘Women and sacrifice in the Roman Empire’, in O. Hekster, Sebastian Schmidt-Hofner, and Christian Witschel (eds.), Ritual Dynamics and Religious Change in the Roman Empire. Leiden, 253–267.

Festus and the Role of Women in Roman Religion

233

Huet, V. (2008). ‘Des femmes au sacrifice: quelques images romaines’, in V. Mehl and P. Brûlé, 2008, Rennes. Jameson, M. (1965). ‘Notes on the sacrificial calendar from Erchia’. BCH 89: 158–72, esp. 162–5. La Follette, L. (1994). ‘The costume of the Roman bride’, in Sebasta and Bonfante, 1994, 54–64. Le Bonniec, H. (1958). Le culte de Cérès à Rome: des origines à la fin de la république. Paris. Lhommé, M.-K. (2007). ‘Varron et Verrius’, in Glinister and Woods, 2007, 33–42. MacBain, B. (1982). Prodigy and Expiation: A Study in Religion and Politics in Republican Rome. Brussels. Moreau, P. (2007). ‘Le lexique de Festus témoin de la naissance d’une science de la parenté à Rome’, in Glinister and Woods, 2007, 69–86. Osborne, R. (1993). ‘Women and sacrifice in classical Greece’. CQ 43: 392–405. Piccaluga, G. (1964). ‘Bona Dea: due contributi all’interpretazione del suo culto’. SMSR 35: 195–237. Scheid, J. (1992). ‘The religious roles of Roman women’, in P. Schmitt Pantel (ed.), A History of Women in the West I. From Ancient Goddesses to Christian Saints, trans. A. Goldhammer. Cambridge, MA, 377–408. (Fr. original 1991 ‘D’indispensables “étrangères”: les roles religieux des femmes à Rome’, in P. Schmitt Pantel (ed.), Histoire des femmes en occident I. Paris, 406–37.) Scheid, J. (1995). ‘Graeco ritu: a typically Roman way of honoring the gods’. HSCPh 97: 15–31. Scheid, J. (2001). Religion et piété à Rome. Paris. Schultz, C. E. (2000). ‘Modern prejudice and ancient praxis: female worship of Hercules at Rome’. ZPE 133: 291–7. Schultz, C. E. (2006). Women’s Religious Activity in the Roman Republic. Chapel Hill, NC. Sebesta, J. L. (1994). ‘Symbolism in the costume of the Roman woman’, in J. L. Sebesta and L. Bonfante (eds.), The World of Roman Costume. Madison, WI. Spaeth, B. S. (1996). The Roman Goddess Ceres. Austin, TX. Staples, A. (1998). From Good Goddess to Vestal Virgins: Sex and Category in Roman Religion. London. Wagenvoort, H. (1956). ‘Initia Cereris’, in his Studies in Roman Literature, Culture and Society. Leiden, 150–68. Wagenvoort, H. (1980). ‘The goddess Ceres and her Roman mysteries’, in his Pietas: Selected Studies in Roman Religion. Leiden, 114–46.

9 Interpreting Sacrificial Ritual in Roman Poetry Disciplines and Their Models Denis Feeney

The interpretation of sacrificial ritual in Roman poetry is a more pressing and rewarding issue than it might have seemed even twenty years ago, when many would have regarded both Roman ritual and Roman literature as equally formalist and arid. We may now be more prepared to entertain the possibility that Roman poetry and Roman ritual are both capable of doing important cultural work, and to accept that the interaction between the two, in the form of poetic engagement with ritual, might likewise be doing important cultural work. It remains, however, very difficult to analyse this interaction between what we call literature and what we call ritual, just as it remains very difficult to analyse any case of interaction between what we call text and what we call context.

DISCIPLINES AND MODELS

The Need for Models It will be helpful to begin by being as explicit as we can about our models, of ritual, and of literature. I take it that we are always using models of one kind or another, whether we acknowledge it consciously or not. More importantly, we always need models of one kind or another because the mass of data will defeat us otherwise. The lack of explicit models means that we just flounder in the sea of evidence—to prove the point, you have only to read the old Pauly-Wissowa entry under ‘Opfer (Sinn)’. And to Denis Feeney, Interpreting Sacrificial Ritual in Roman Poetry: Disciplines and Their Models In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0010

Interpreting Sacrificial Ritual in Roman Poetry

235

see that genuine advances have been made since then thanks to the self-conscious importing of models into Classics from other disciplines, principally anthropology, you have only to read Andreas Bendlin’s entries in Der Neue Pauly under ‘Opfer: Theorien’ and ‘Ausblick’. The challenge is to try and clarify what is at stake in the choice of models, and especially what is at stake in the interchange of models from one discipline to another. We are often told that the boundaries between disciplines are falling away, and that history, anthropology, literary criticism, and political science are coalescing. However welcome and exciting such developments may be, there is a risk that we will end up in a position analogous to that adopted by people who deny that the distinctions between genres are relevant to the study of Ovid’s Fasti or Metamorphoses. As a number of recent studies have taught us, the creative transgression of boundaries does not annul the categories, but redefines them.1 I take the problem of sacrifice as a test case partly by way of recantatio for not having talked about sacrifice as an issue in Feeney 1998; I gestured towards the problem in the chapter on ‘Ritual’ (‘To moderns, sacrifice is a vital aspect of ritual’, 119), and then went on to say nothing specifically about it. Mainly, however, sacrifice appeals as a test case because the role of sacrifice in literature, specifically in Virgil’s Georgics, has recently occasioned a debate that is highly illuminating for the current enquiry. Habinek 1990 and Thomas 1991, followed in particular by Morgan 1999, have turned a searchlight onto the problem of the sacrificial dimension to the bugonia at the climax of the Georgics. I advance no new reading of Georgics 4, and make no claim to solve any of the issues of interpretation. I choose this starting point because the debate illuminates with particular clarity what is at stake in the confrontation between disciplines and their models. I shall then take up the lead provided by Fantham 1992, and follow the theme of sacrifice from Virgil’s Georgics into Ovid’s Fasti, in order to provide another test case of the interaction between ritual and literature.

Models of Sacrifice Before turning directly to Virgil and Ovid I should give an account of the models and working hypotheses I am using in the case of sacrifice and of literature, although I remain aware that the motivations for an

1

Conte 1986: 100–29; Hinds 1987 and 2000; Barchiesi 2001a.

236

Interpreting Sacrificial Ritual in Roman Poetry

individual’s preferences and practices in this regard must, at some level, remain opaque to him or her. Some first principles, then, so far as I have access to them, brusquely presented.2 The meaning of sacrifice is not a question of origin. In the debate over this question in Hamerton-Kelly’s Violent Origins between Walter Burkert, René Girard, and Jonathan Z. Smith, it is Smith who clearly emerges triumphant. The meaning of ritual is not to be found in the survival of some prehistoric trace, whether it be neolithic hunting guilt (Burkert) or a Remus/Abel human scapegoat sacrifice (Girard); the meaning of ritual is not, as Smith puts it, ‘somehow grounded in “brute fact” ’, but instead in what he calls ‘the work and imagination and intellection of culture’.3 It is always the current work of ritual that matters, not where it might once have come from. This may appear to be a hard perspective for students of the ancient world to work with, since the antiquarian religious work of the ancients is so overwhelmingly aetiological. The methodology of the ancients, however, gives no ground for modern foundationalist theories of explanation by historical origin, since, as we shall see in this paper, ancient aetiological methods are so often intent on ‘muddying the waters of the source’ and making the origin of sacrifice a problem.4 For all his scepticism about origins, Smith does offer, more or less as a jeu d’esprit, an aetiological myth for the origin of animal sacrifice which is far more historically plausible than Burkert’s or Girard’s, namely, the selective culling of domesticated animals in breeding. The Roman literary evidence certainly fits Smith’s myth, as we shall see, linking sacrifice always with the world of the agriculturalist and his domesticated animals, not with hunting wild animals. It is salutary to read the work of Jared Diamond, and to learn how bizarre domestication is, how recent it is as part of our species’ history, and how few animal species have ever successfully undergone it.5 We may think of the wild animal as the numinous and uncanny, but from an evolutionary point of view the really weird freaks are all around us, in the shape of the domesticated animals. Still, Smith affects not to care if his origin myth is true or not, because for him the meaning and work of ritual are contemporary and ongoing, however apparently fossilized the forms. According to him, and to Catherine Bell, whose work develops his in many respects, ritual is not

2 For a fuller discussion and documentation of a number of these issues, see Feeney 1998. 3 Smith 1987: 198. 4 To borrow the phrase used of Ovid by Barchiesi 1997: 218. 5 Diamond 1997: 157–75, showing that only five species are really significant in the history of domestication (sheep, goat, cow, pig, and horse).

Interpreting Sacrificial Ritual in Roman Poetry

237

precultural, nor is it foundational. This anti-foundational way of looking at ritual is rather at odds with the traditional assumptions of structuralism or symbolic anthropology, as represented in Classics particularly by such figures as Vernant, Vidal-Naquet, and, in his rather different way, Burkert.6 Now, the impact of structuralism and of symbolic anthropology on the study of ancient religion has been extremely valuable, and will certainly leave its traces in any imaginable future synthesis, but its main drawback is the way that it posits an overarching holistic and unifying thoughtworld for any given society, a mentalité. Such an approach almost inevitably ends up seeing ritual as an expression of this overarching mentalité, and especially as underpinning it in a foundational sense. But such a supposition is very dubious, and Maurice Bloch in particular has exposed its weaknesses, above all its tendency to obscure the fact that ritual is only one of many mentalités or knowledge-systems in any society, and by no means the foundational knowledge-system; ritual is, or can be, extremely self-contained, so that it cannot readily be ‘read off ’ as a metaphor for other knowledge-systems or power-structures in the society.7 Wilkins has recently explored this question in connection with the language of the Iguvine Tablets: ‘ritual language inhabits a specialised domain even within the subject culture and within the whole context of the practice and evolution of the social uses of language. Ritual language . . . can be seen to have its own domain, and within that domain, its own rules’.8 According to these approaches, there is no one mentalité that fits a whole society, whether that mentalité is identified with ritual or anything else.9 Although Bloch has his eye on anthropology and does not explicitly take account of New Historicism or Cultural Poetics, his criticisms could clearly be extended by analogy to take in these other varieties of anthropologically derived holism; I shall return to these questions at the end of the paper. A corollary to this scepticism about one great overarching system is that one must expect to find a multiplicity of interpretations of ritual activity.10 The Roman attitude to sacrifice is not a recoverable entity; indeed, sacrifice at Rome is described by Richard Gordon as being ‘to a degree a vacant sign’.11 At this point we must also remind ourselves that

6

An overview of the French school in Buxton 1981. Bloch 1989: esp. ch. 1. 8 Wilkins 1994: 164; my thanks to Ann Kuttner for this reference. 9 See also Lloyd 1990 for a trenchant criticism of the mentalité mentality. 10 Feeney 1998: 127–9. 11 Gordon 1990: 206; cf. Der Neue Pauly, 8. 1250: ‘Da die röm. Rel. ein offenes, fließendes system war, bleibt die Suche nach einer “Bed.” wohl fruchtlos’. 7

238

Interpreting Sacrificial Ritual in Roman Poetry

ritual is not a discrete category in ancient thought, and nor is sacrifice exactly a discrete subcategory of it.12 In Rome there is no Platonic form or idea of sacrifice ‘out there’, which is then represented or captured more or less imperfectly by an artist. When we conduct a quasianthropological search for the meaning of ritual or of sacrifice in ancient texts, our object of enquiry is very much a modern construct, for ancient authors have extremely little in the way of explicit theorizing about sacrifice: ‘Although modern scholars may construct an explanation of Roman sacrifice by putting into modern words themes and associations which were almost entirely implicit and unspoken for the actors, the system itself produced no theological account of the meaning and purpose of sacrifice’.13 In fact, as I tried to show earlier in the case of divinity, and as I shall try to argue here in the case of sacrifice, at Rome, just as in Greece, it was primarily what we call literature that did the job of exploring what Gordon calls the ‹meaning and purpose› of divinity or sacrifice.

Models of Literature The engagement with sacrifice in literary texts adds more layers of complication to this already complicated picture. No literary text offers us a representation, in the strict sense, of anything, let alone sacrifice. In making this claim I am of course employing a model from literary criticism or hermeneutics, or, rather, signalling a shared concern from a number of different literary critical or hermeneutic models, whether the Contean generic approach, deconstruction, or even the old New Criticism. The idea that literary texts represent or reflect reality is having an odd comeback, but I think we have to take very seriously the objections to this idea which are posed by such literary-critical or hermeneutic models. At the most basic level, any text or genre has its own priorities, traditions, methodologies. Further, Roman literature is very self-conscious about its own distinctive way of engaging with ritual.14 Roman authors know perfectly well that ritual in their texts is not a facsimile of ritual in other contexts, just as they know that anything in their texts is not a facsimile of anything in other contexts. The apparently real and concrete and grounded nature of sacrificial ritual is so strongly present to us that we can fall into making assumptions about the transparency of literature’s 12 Feeney 1998: 117–18. Arguably, ritual is not a discrete category of inquiry at any time or place: such is the main argument of Bell 1997. 13 14 Gordon 1990: 206. Feeney 1998: 32–8; Barchiesi 2000 and 2002.

Interpreting Sacrificial Ritual in Roman Poetry

239

engagement with sacrifice that would arouse scepticism or derision if we entertained them in the case of, for example, love elegy’s engagement with biography or carnality.15 Again, I return to these issues of representation and textuality at the end of the paper.

VIRGIL’S GEORGICS

Walter Burkert in the Georgics Many of the issues I have been discussing so far are visible, or just beneath the surface, in the starkly differing papers on the bugonia in the Georgics by Thomas Habinek and then, in response, by Richard Thomas. It is clear that Thomas’s fundamental objection to Habinek’s method is that he sees Habinek as importing from Greek studies a structuralist anthropological model whose modern themes and associations, according to Thomas, may conceivably have something to do with the Greek world but have nothing to do with the Roman world.16 In some respects Thomas’s criticisms are cogent, especially when he objects to Habinek’s use of the standard Greek sacrificial model to dictate a necessarily ameliorative interpretation of the resurrection of the bee community: as Habinek puts it: ‘Social interaction and human culture come to be seen in a positive light, and, with them, the institution of sacrifice that makes their existence possible’.17 Thomas is right to say that this is an overly procrustean imposition of a particular model, in which the model is driving the interpretation, and he makes some telling points in detail, but his fundamental methodological reservation about Habinek’s methodology is ill-founded. In trying to locate the sacral or quasi-sacral passages of the Georgics explicitly within some larger interpretative context, Habinek may be on the wrong train but he is on the right track.18 The friction between the sacrificial patterns inside and outside the poem demands interpretation. To Thomas, however, the very use of an extra-literary sacrificial model is illicit, as becomes clear in a series of rhetorical questions towards the end of his article, in the course of which he quotes Habinek’s characterization of sacrifice: ‘Can we ever say of “the Romans” (or even “the Greeks” for that matter) that 15

See e.g. Wyke 2002 for a discussion of the related issues in elegy. 17 Thomas 1991: 216f. Habinek 1990: 216. 18 A line I stole, with full apparel, from Professor Joseph Farrell—whom it is a pleasure to thank for his characteristically generous and helpful correspondence on these problems. 16

240

Interpreting Sacrificial Ritual in Roman Poetry

for them “sacrifice” is a means of establishing the relationship between human and divine, of defining the order of society and the universe, and of restoring that order when it has been disrupted’ (p. 212)—even if we add footnotes referring to Burkert’s theories on Greek religion? Would not some Romans find such a proposition as ridiculous and trite as we do? Would not some be as horrified and repulsed at witnessing the slaughter of oxen as we would? Or would they feel that they had thereby affirmed correct relations with the gods—whoever they were?’19 There were indeed various views on the merits of animal sacrifice both in the Roman and Greek worlds, and we shall be seeing some horror and revulsion expressed by Virgil and especially by Ovid later in the paper. Nonetheless, a good deal of Roman state cult is underpinned precisely by some such view of ritual and sacrifice as maintaining order between the state and its gods, the pax deorum, and restoring that order when it has been disrupted.20 Habinek’s Burkertian formulae are too vague to serve as determinative guides for the exegesis of an immensely complicated literary text, as Thomas quite rightly points out, but the model itself may have something to offer a literary reading, so long as it is not regarded as homogeneous and unitary, or prescriptive in terms of the literary readings it can enable or disable, but instead as an initial set of intellectual or imaginative possibilities. The ritual and sacrificial underpinning of the pax deorum is the view presupposed by many Roman observers of Roman state cult, and it is the view presupposed by Virgil in the Georgics—as may be seen in particular when Cyrene tells Aristaeus to supplicate the nymphs and seek pacem, so that they will in return grant pardon in response to his prayers, and cease their anger.21 Virgil does not take over such a view casually or by default because that is how his society as a whole just naturally saw things, but for his particular purposes: he takes this selective point of view as his starting point not in order to replicate it, but in order to give power to his own departures. Before investigating these departures of Virgil, we need to consider another important methodological point highlighted by Thomas’s criticisms of Habinek. As we have seen, Thomas objects in principle to the application of a Burkertian Greek sacrificial model to a Latin literary text, largely on the grounds that the model is not framed in terms that would

19

20 Thomas 1991: 217. Rüpke 2001. 4. 534–6: namque dabunt ueniam uotis, irasque remittent. On the linking of pax and uenia, see Wissowa 1912, 390–1, and on the remission of divine ira in response to human uota when the pax deorum is breached, see Rüpke 2001: 21. It is interesting that Mynors 1990: ad loc. comments on the traditional language of pax and uenia, while Thomas (1988) does not. 21

Interpreting Sacrificial Ritual in Roman Poetry

241

have been accessible to the original participants: Greeks did not think in these terms about sacrifice, and nor did Romans. This is what Gordon describes, in the words already quoted,22 as ‘constructing an explanation of Roman sacrifice by putting into modern words themes and associations which were almost entirely implicit and unspoken for the actors’. The difference, of course, is that Gordon sees this hermeneutic conundrum as inevitable, whereas Thomas sees it as illicit and anachronistic. But Gordon is right. Any historical or anthropological project is going to need models or frames of analysis that are incongruent with the experience of the participants. We cannot be them, and we must process the data into some kind of shape for it to make any sense to us.23 The challenge for the historian or anthropologist is to be aware of this inevitable incongruity or disparity between the observer’s and participant’s experience, and so to avoid two opposite errors: one is to project the model onto the participants, and claim that they really knew this structure, though maybe only subconsciously—Gordon is in fact rather close to this position; the other is to say that the facts speak for themselves and do not need ordering in a structure for an outsider to get a grasp on them. For the purposes of analysis, the participants’ perspective is regularly unsatisfying. Dirk Obbink puts the point very well in his discussion of ancient and modern theories of sacrifice: ‘I do not want to suggest that ancient theories in the matter have necessarily any greater chance than modern theories of being right. They are often demonstrably wrong: paradoxically, their very proximity in time and cultural context to the phenomena in question puts them at a distinct heuristic disadvantage’.24

Patterns of Sacrifice in the Georgics As a result of his hostility to what he sees as a New Historicist imposition of non-literary models, Thomas virtually ends up implying that sacrifice is not important or interesting to Virgil. He points to the catastrophic failure of the one real sacrifice narrated in the poem, during the Noric plague in Book 3 (486–93), as if to suggest that the quest for a meaning to sacrifice in the Georgics is pointless.25 It is certainly true that the

22

See p. 4. The first chapter of Kennedy 1993 is indispensable on this topic. 24 Obbink 1993: 80. For further discussion of this interpretative paradox, see Feeney 1995: 311f. 25 Thomas 1991: 215–16. 23

242

Interpreting Sacrificial Ritual in Roman Poetry

catastrophic failure of the recognizably Roman and ritually correct performance of sacrifice in Book 3 highlights the absence of regular and successful sacrificial practice elsewhere in the poem. There is indeed very little reference to normative Roman sacrificial practice in the Georgics. In Book 1 Virgil glances at the felix hostia of the Cerealia, but does not describe its sacrifice—and his offering of honey, milk, and wine together has no Roman parallel (1. 343–50); at 2. 192–4 he evokes a sacrifice complete with wine and ‘steaming entrails’ (fumantia exta, 194); at 2. 380 he gives an aetiology for the sacrifice of the goat to Bacchus (not actually part of Roman cult at all), and follows it up with an evocation of the sacrifice of the goat (2. 393–6); at 2. 536–8 he alludes to the impious feasting on plough-oxen that marks the end of the Golden Age, in a manner that is ultimately inextricable from a sacrificial reference, however deliberately inexplicit it remains;26 and at 2. 146–8 and 3. 23 he alludes to, without narrating, the slaughter of oxen at the Roman triumph.27 We never, then, actually see a regular Roman sacrifice in the Georgics; if modern students of sacrifice are frustrated by this state of affairs, they should reflect that Virgil is not interested in documentary realism or helpful proleptic collaboration. His interest in sacrifice runs deeper, in fact, than one might gather from the list in the previous paragraph. Monica Gale has made the best and most sustained case for Virgil’s ability to use sacrifice as a systematic way of thinking about human beings’ relationship with animals, and, by extension, with the natural world as a whole.28 She traces a developing process of disenchantment or reorientation through the poem, with the institution being apparently taken for granted in the first book, and then gradually denaturalized, as empathy for the sacrificial victim increases and the freakish nature of human interaction with the rest of the natural world is systematically unlayered. In his wish to argue against Habinek’s use of external models: Thomas comes very close to saying that we may only read with an eye to sacrificial connotation if the text enacts a sacrifice with punctilious correctness. As I suggested above, however, we should expect that literature will not represent—re-present—patterns of action from other spheres. That is not where the techne of the poet resides, as Aristotle taught us. We should not be surprised by the fact that Virgil’s most sustained engagements with sacrificial patterns come at two highly anomalous moments—the Noric plague and the bugonia. 26 27 28

So, rightly, Dyson 1996: 278f. and Gale 2000: 107 n. 161. Cf. 3. 160, where sacrifice is one of the reasons for rearing oxen. Gale 2000: esp. 101–12.

Interpreting Sacrificial Ritual in Roman Poetry

243

The final book of the poem ends, before the sphragis, with the ritual action performed by Aristaeus to placate the nymphs and the shades of Orpheus and Eurydice, a ritual action which results in the completely unexpected emergence of the bees, nine days later, from the rotting carcasses of the slaughtered oxen (4. 534–58).29 In many respects Harrison is basically right to say that ‘the Aristaeus epyllion is above all an illustration in mythological form of orthodox Roman procedure in a plague context’.30 But it is a strange fantasy, with a strangely refracted relation to the supposedly actual Egyptian practice of bugonia, let alone to contemporary Roman practice.31 Virgil tells us what Egyptian bugonia is like earlier in Book 4, and it is not at all like what Aristaeus does.32 At this point I must mention the Stanford connection. Susan Stephens and Alessandro Barchiesi are both independently working on bugonia and sacrifice in Callimachus’ Egypt and Virgil’s Italy, and they have both been extremely generous in helping me see the larger parameters for the ritual Virgil narrates at the end of his poem. It is clear from their work that the end of the Georgics is part of a Virgilian and Augustan debate with a Callimachean and Ptolemaic debate over cultural norms and centres of gravity. Callimachus had appropriated Greek norms to an Egyptian context (so Stephens and see commentary, pp. 157ff.); the Aristaeus epyllion in particular combats Callimachus’ appropriation of Greek norms to Egypt by appropriating them back to Hellas, and to Italy (so Barchiesi); the Egyptian practice of bugonia is barbarized and distanced and made groundless as a base for Greco-Roman sacrificial practice (so Stephens and Barchiesi combined). These are very rich projects, and when they are published they will make a big difference to our readings of Callimachus and Virgil. For my present purposes, what is most fruitful about these projects is how much light they shed on the long-standing problem of the bizarre nature of the aetiology at this climactic moment of the poem. The very 29 Mynors 1990: 321, following Servius ad 4. 553, rightly stresses that the resurrection of the bees from the carcasses is not foreseen by Cyrene. 30 Harrison 1979: 52 n. 6. 31 The ritual of leaving sacrificial victims unburnt and unconsumed is actually not as unheard-of as is often claimed: see Mynors 1990: 321 for the animalis hostia or animale sacrificium, in which only the victim’s life (anima) was offered to the deity (his reference to Macrobius’ Saturnalia should be 3. 5. 1–5). Latte (1960: 379) is no doubt correct to say that this category is a piece of antiquarian casuistry with no consequence for cult practice, but the presence of the category in the tradition offers Virgil enough purchase. 32 Thomas 1988: on 538–58. Stephens Hinds (apud Myers 1994: 155 n. 86) points out how the fantastic nature of Virgil’s bugonia is picked up in the wonderful joke of Ovid’s Pythagoras, who refers to bugonia as cognita res usu, when libro would be nearer the mark (Metamorphoses 15. 365).

244

Interpreting Sacrificial Ritual in Roman Poetry

nature of aetiology is called into question here, even at the level of the bad casting as protos heuretes of Aristaeus, who has to be told everything.33 What exactly is being explained by its origin in the fourth Georgic? Bugonia, strictly speaking, but that is really a blind. Harrison and Habinek are right to suggest that the propitiatory practice of Aristaeus adumbrates contemporary Roman practice, but the links between that past moment and current practice are tenuous, to say the least. Virgil keeps going back and back to ‘explain’ the present, but there is no ultimate grounding for his explanation. Sacrifice is inextricably enmeshed in contemporary society, but it cannot be given a deep foundation in some determining earlier, extra-cultural moment.34 How such a reading meshes with the contemporary self-representation of the Princeps is an open question. Each reader must still interpret the consequences of the sacrificial possibilities at the climax of the poem being treated in this grotesque and self-consciously fantastic way at a time when the princeps himself is already well embarked on his career-long practice of making sacrifice and its representation central to the new Rome and the role of its new leader.35 Barchiesi well brings out how crucial the power of origins was to the image of the Princeps as sacrificer: ‘One of the reasons for the great importance and diffusion of this visual representation of the sacrifice is that this rite is repeated in time and guarantees its own origins’.36 Thomas, in other words, could have made his own ‘pessimistic’ reading of the end of the Georgics more convincing by taking the sacrificial models more seriously. But then so could Habinek have made his own ‘optimistic’ reading more convincing by looking more carefully at what sacrifice might be represented as doing in Caesar’s Rome rather than in archaic Greek texts. Virgil’s text needs to be historically contextualized, but not in a way that implies that there is a reading of sacrifice ‘itself ’ out there in the world, one unifying interpretation, that will effect closure: Elsner’s paper on the Ara Pacis, a powerful exposition of the polysemic resonances of sacrifice in Virgil’s society, makes it plain how hard it is to enlist the institution as a closural device.37 Our reading of sacrifice, in other words, is part of a loop that leads into our reading of the poem and back out again. Don Fowler has 33 Putnam 1979: 314 n. 61: ‘He seems throughout the episode to have little or no selfreliance or insight of his own. He does what he is told, not what he himself determines’. My thanks to Mira Seo for pointing this out to me. 34 My debt to Smith 1987 (in Hamerton-Kelly) is obvious, and I must also acknowledge my debt to a conversation on this point with David Leitao at the Stanford conference. 35 36 Gordon 1990. Barchiesi 1997: 219. 37 Elsner 1991. The general point about the illusory power of extra-textual referents to establish closure is eloquently made by Fowler 2000: 173–4, 192.

Interpreting Sacrificial Ritual in Roman Poetry

245

acutely shown that what is at stake between Habinek and Thomas is a mutually self-reinforcing attitude towards closure inside and outside the text: the issue of whether sacrifice is more open or more closed as an institution folds back into the issue of whether Georgics 4 is read as a text with more or less closure.38 The same issue is fundamental to Llewelyn Morgan’s reading of the bugonia.39 He argues, rather as Habinek does, that the ritual at the end of the poem is redemptive, optimistically looking forward to a regeneration, and that it is a model for a redemptive social regeneration out of the impious sacrifices of the civil wars, under the new reign of Caesar Octavian. He acknowledges the transgressive and anomalous dimensions to the sacrificial actions at the end of the poem, but he sees them as part of a virtuous circle—the more impious and bloody the sacrifice, the more powerful and creative the redemption. Morgan uses his reading of sacrifice to elucidate vividly some of the poem’s main sources of power, showing in particular how the Virgilian fascination with the creativity of violence and the violence of creativity is illuminated by the generation of life from sacrificial death. Yet, as in the argument of Habinek, there is an instinct to close down the open problems of the text by referring them to a supposedly stable external referent.40 Morgan, again rather like Habinek, sees the institution of sacrifice as grounded in paradox, but as still bearing a unified (if paradoxical) meaning.41 Another difficulty with his overall argument is that it cannot pass the Karl Popper test: it is unfalsifiable. The more impiety and horror an opponent adduces against the argument for the redemptive power of sacrifice, the more powerful the argument becomes—the more grotesque and appalling the sacrifice, the more paradoxically powerful the redemption. Some Roman readers may well have read the bugonia in this way, but I cannot believe that the poem makes such a reading inevitable. Morgan’s argument also bestows a moral justification on a pattern that Virgil may be representing as merely inevitable. Is all this killing, in sacrifice or in civil war, genuinely redemptive and constructive, or is it only a pattern of action that Romans are locked into?42 If there is no way 38

39 Fowler 2000: 286–7. Morgan 1999. Feldherr (2002: 70–1) makes a very powerful case against a similar appeal to sacrifice in Morgan 1998; I have found Feldherr’s whole argument very helpful. 41 Morgan 1999: esp. 113–16. 42 Putnam 2000:159 well points out that ‘as in the case of Remus’ death, the negative energy associated with bugonia begets not some idealizing higher ethical scheme but another set of bees and presumably a renewal of their inherently martial identities’. Cf. Perkell 1989: 76, on how for the ancients ‘bougonia apparently signified an exchange of death for life rather than rebirth or resurrection’. 40

246

Interpreting Sacrificial Ritual in Roman Poetry

in to sacrifice, no validating ground of origin, then it appears that there is no way out either, no way of getting off the treadmill. A main source of the power of the Romulus and Remus myth, newly significant in the civil wars, is its circularity, most memorably evoked in Horace’s Epode 7.43 The killings of the civil war are not necessarily the prologue to a definitive settlement, but may be only a replaying of a prototypical pattern of events. The institution of sacrifice would corroborate the claustrophobic power of this approach, since one of the keys to sacrifice is its repetitiveness: the same thing happens again and again, at the same time, in the same place.44 If we are looking for redemption, circularity may be counterproductive. Sacrifice emerges as Gordon’s ‘vacant sign’ indeed. As Catherine Bell puts it, ‘the strategies of ritual may well generate the sense of a basic and compelling conflict or opposition in light of which other contrasts are orchestrated’.45 In Virgil’s case, the most important of these contrasts would be between order and entropy, the life of the individual and of society, and between the life of the Golden and the Iron Age. Is sacrifice a normative way of keeping the world going round, a sacred act, or is it itself a symptom, a trace of humanity’s denatured state, a sign of impiety, the definitive mark of the civilized imperial power?

OVID’S FASTI

The Two Faces of Ceres In following the theme of sacrifice from the Georgics to the Fasti, my starting point is Elaine Fantham’s important article on Ovid and the Georgics, in which she makes an entirely convincing case for Ovid responding thoughtfully and systematically to the Georgics throughout the Fasti in his treatment of sacrifice and the life of agriculture.46 Fantham begins by examining Ovid’s two principal passages about the goddess Ceres, and gives a detailed account of the evident contradictions between them.

43 Compare the hideous repetitious force of iterum and bis in the description of the civil war battlefields at the end of Georgics 1 (490–1). 44 My thanks to Nicholas Purcell for stressing to me the importance of the repetitiveness of ritual. 45 46 Bell 1992: 37. Cf. Gale 2000: 107–9.

Interpreting Sacrificial Ritual in Roman Poetry

247

The first passage offers an image of Ceres as the emblem of the Iron Age, as the Agonalia of 9 January give Ovid the opportunity for his first and programmatic account of animal sacrifice (1. 335–456). It is important for Ovid to have animal sacrifice early on, not just because he is a Hesiodic poet, but because he is a good neoteric, and he follows the neoteric pattern isolated by J. E. G. Zetzel, in an important article: Ovid likes to start where Virgil ends.47 Before navigation and commerce there was simple non-animal sacrifice (337–8), but the goddess of agriculture, once we move into the Iron Age of ploughing and sailing, was the first to demand animal sacrifice in the form of a sow (349); and Ovid goes on to describe how the gods demand and receive animal sacrifice of every kind, cutting a swathe through the animal kingdom. He tells us of the goat (354) and the innocent ox and sheep (362)—and with the ox he introduces his own version of the Aristaeus story (362–80), thereby capitalizing on the sleight of hand by which Virgil was able to associate animal sacrifice within the aetiological penumbra of the bugonia.48 The killing of the ploughing ox is the climax of this section (383f.), and the killing of the ploughing ox, regularly, but not invariably, associated with its sacrifice, is the definitive mark of the end of the Golden Age and of man’s estrangement from nature in Aratus, the end of Georgics 2, and in Pythagoras’ speech in Metamorphoses 15.49 After the ox there follow the horse (385), deer (389), dog (390), and ass (391). Later, in Book 4 (681–712), we learn that Ceres also likes burning foxes, and at the end of that same book (941f.) we are told that a dog is sacrificed on April 25th for the purely contingent reason that he shares a name with the dog star. In the Book 1 programmatic passage, after the aetiology of ass-sacrifice, with the story of Priapus and Lotis (391–440), Ovid returns to the catalogue of victims, listing all the birds that are killed (441–56), because they are too communicative, revealing to human diviners what the gods are thinking. The cruel impiety of life in the post-Golden Age appears to be radical, with greedy, ruthless, and competitive gods enforcing upon humans the requirement of treating animals, even their workmates, as helpless and 47

Zetzel 1983. Gale (2000: 111) remarks that Ovid’s move offers a perceptive commentary on Virgil. Lefèvre (1976, 46) and Porte (1985: 45, 444–5) see that Ovid’s story of Aristaeus is taken over from Virgil and is not itself an aition for ox-sacrifice, but do not observe that Ovid is cashing in a Virgilian trick in this feint. 49 Aratus Phaenomena 132; Virgil Georgics 2. 536–8; Ovid Metamorphoses 15. 120–1; Gale 2000: 107f. Note how, as soon as the old man Hyrieus in Fasti 5 recognizes the disguised Jupiter, the presiding deity of the Iron Age, he sacrifices his plough-ox to him (cultorem pauperis agni, 515). 48

248

Interpreting Sacrificial Ritual in Roman Poetry

terrified agents in the game of communication played out before the altars. The gods make us treat animals as enemies: Ovid derives the name for sacrificial victim, hostia, from conquered enemies;50 and he does not mean just the human enemies whose defeat is marked by sacrifice to the gods. And at the very beginning of this long programmatic section we are told that one of the etymologies for Agonalia, the first sacrifice, is agonia, the agony, metus of the sacrificial victims as they see the knives in the waterbowls (1. 327f.).51 This stark picture of current life is corroborated by a glimpse of the Arcadian Golden Age, before the birth of Jupiter (2. 289–98); here there is no ploughing, no imperial domination of the land through agriculture, no usus made of other animals, the horse or the sheep. To cap this programmatic section on animal sacrifice, animal sacrifice closes the first book, with pregnant sows being sacrificed to Ceres and Tellus (1. 671f.) and oxen at the altar of Pax (720); the so-called Italia relief of the Ara Pacis is therefore presumably identified by Ovid with Ceres, or Tellus, or both (1. 709–22).52 The Ara Pacis, with its bucrania over scenes of animal and vegetable fertility and abundance, is one of the monuments in Ovid’s Rome which shows the highest degree of selfconsciousness about how much killing has to go on in order to maintain the cycles, as the important article of Elsner 1991 demonstrates.53 The next Ceres passage comes in Book 4 (393–620). When Ovid discusses the sacrifices appropriate to the Cerealia, we see a completely different emphasis, as Fantham shows in detail. Especially, here we are enjoined to perform bloodless sacrifice, of spelt, salt, and incense, and to shun the ox as a sacrificial victim, so as to avoid the impiety of killing our fellow-worker (4. 413–16): A boue succincti cultros remouete ministri: bos aret; ignauam sacrificate suem. apta iugo ceruix non est ferienda securi: uiuat et in dura saepe laboret humo. Attendants, with your tucked-up clothing, remove your sacrificial knives from the ox: let the ox plough; sacrifice the slothful sow. The neck fitted for the yoke should not be struck by the ax; let the ox live and often toil in the hard soil.

50

Ovid 1. 336: hostibus a domitis. Cf. Fantham 1992: 47: I am indebted throughout to her analysis. See Galinsky 1996: 148–9 for the various identifications of the Italia relief. 53 It is important to realize, as Ann Kuttner points out to me, that the bucrania on the Ara Pacis are a continuation of a sacrificial and sculptural programme common in the postClassical Greek world, and in Italy as well: Nilsson 1955: 1. 88. 51 52

Interpreting Sacrificial Ritual in Roman Poetry

249

These lines are only the most dramatic case of a systematic contrast with the atmosphere of the Agonalia in Book 1.54 In Book 4 we do not see the bloodthirsty Ceres of Book 1; she is prima Ceres in 4. 401 as the first deity to boost human diet up the food chain from acorns to corn, not the prima Ceres of 1. 349, the first deity to demand animal sacrifice as revenge on the animal kingdom. Further, Ovid hides the iron of the Iron Age (405f.), apparently mitigating Ceres’ reign. Finally, the just quoted prayer that the plough ox should be spared the sacrificial axe is a hoped-for Golden Age survival into the present age, rather reminiscent of the strange chronological disturbances one so regularly encounters in the Georgics.

The One Face of Ceres In terms of ritual and aetiology, however, the whole passage in Book 4 is not, in the end, as divergent as it may seem from the devastating view taken of animal sacrifice in the Agonalia section of Book 1. Fantham argues that the first passage is so bitter and so different from the second one that the difference has to be explained biographically as the result of Ovid’s disillusionment with his society and its religion in exile. If the inconsistencies are not so radical, we may be less inclined to fall back on a biographical explanation. To begin with, Ceres is not disassociated from sacrifice even in Book 4. Just as in Book 1, Ceres here continues to receive her proper sacrifice of the sow (4. 414). And, as we know from the end of Book 1, the peace she delights in is guaranteed by the sacrifices of the white ox at the Ara Pacis of the pacific leader (4. 407f.). Above all, it is very important to keep reading from the end of this ritual prescription section into the immediately following myth of the rape of Proserpina, especially since Ovid himself says that it is very apposite to do so: ‘The [common-]place itself [in my poem / in the calendar] demands that I make public the rape of the virgin.’55 We are not disappointed. When Ceres misses Proserpina and goes searching for her, she is immediately compared to the mother cow in Lucretius who has lost her calf to the operators of the sacrifice mill (Lucretius 2. 352–9): nam saepe ante deum uitulus delubra decora turicremas propter mactatus concidit aras, sanguinis expirans calidum de pectore flumen; 54

Again, see Fantham 1992 for a full analysis and discussion. 4. 417: exigit ipse locus raptus ut uirginis edam. On the manifold wit of this line, see Barchiesi 1997: 75f. 55

250

Interpreting Sacrificial Ritual in Roman Poetry at mater uiridis saltus orbata peragrans quaerit humi pedibus uestigia pressa bisulcis, omnia conuisens oculis loca si queat usquam conspicere amissum fetum, completque querelis frondiferum nemus adsistens . . . .

For often before the fine shrines of the gods a calf falls, sacrificed beside the incense-burning altars, breathing out a warm stream of blood from its chest; but the mother, bereaved, wanders through the green glades and looks on the ground for the tracks made by the cloven hoofs, looking over all the places to see if she may anywhere catch a glimpse of her lost offspring, and stops and fills the leafy grove with her laments . . . .

And here is Ceres (Fast. 4. 459–62): ut uitulo mugit sua mater ab ubere rapto et quaerit fetus per nemus omne suos: sic dea nec retinet gemitus et concita cursu fertur et a campis incipit, Henna, tuis. As a mother moos when her calf has been snatched from her udder and looks for her offspring through every grove, so the goddess does not hold back her groans and is carried along at a run, and starts, Henna, from your plains.

Although in the didactic section before the myth Ceres is not associated with ox-sacrifice, in the myth she is associated by simile with one of the animals who will make the Iron Age work, by working, and by providing fodder for sacrifice. Strictly, Ovid keeps up even here the erasure of explicit mention of sacrifice, since he does not explicitly say, as does his model Lucretius, that the calf has been taken for the purposes of sacrifice; yet the pressure to read this dimension in to Ovid’s simile and to forge a link with the programmatic Agonalia passage is irresistible, since the sacrificial motivation for the taking of the calf is so powerful a part of the Lucretian model, which itself is designed to cast a pall over the institution of animal sacrifice.56 In the Georgics Virgil had already used significant diction from the Lucretian simile to colour his presentation of the devastation visited on the nightingale/Philomela when the durus arator steals her young, to reinforce his theme of the random cruelty that human intervention can inflict on the animal world (4. 511–15).57 Only twenty lines after the simile of Ceres and the mother cow, Ovid

56

Gale 2000: 105 on how in Lucretius the simile works to establish animal sacrifice as wantonly cruel and pointless. 57 Gale 2000: 135f., on the debt of Virgil’s amissos fetus and questibus implet (4. 512, 515) to Lucretius’ amissum fetum, completque querelis (2. 358).

Interpreting Sacrificial Ritual in Roman Poetry

251

acknowledges Virgil’s use of the Lucretian simile by himself alluding to Virgil’s allusion, comparing Ceres’ laments to the lament of Philomela for Itys, using language that recalls both Lucretius’ cow simile and Virgil’s adaptation of it for his own philomela simile.58 The compulsion to see the goddess Ceres enmeshed in the dynamics of sacrifice is reinforced even before the Itys simile, immediately after the Lucretian cow simile, when we see a hint at the enmity of Ceres towards her typical sacrificial offering, the pig; here we are told that she would have tracked down her daughter’s path there and then if pigs had not disturbed the tracks (463–6). In the Agonalia passage Ovid informs us that Ceres started the pattern of animal sacrifice by taking revenge on the pig for rooting up the new crops (1. 349–52). Ceres has got more than one reason for not liking pigs. If these glances at the issue of sacrifice at the beginning of Ceres’ search for Proserpina incline one to look for a prototypical Ceres of the Iron Age in Book 4 as well as in Book 1, then there is further confirmation later in the story, when Ceres, after much wandering in search of her daughter, comes to Eleusis, and is interrupted by Triptolemus’ mother as she is giving the boy immortality (549–56). Ceres informs the mother that the gift of agriculture will be a recompense for Triptolemus’ mortality (559 f.): iste quidem mortalis erit: sed primus arabit et seret et culta praemia tollet humo. He will be mortal; yet he will be the first to plough and sow and take up rewards from the cultivated earth.

This strong marking of Triptolemus as the first agriculturalist is at odds with other touches in the telling of the Ceres myth in Fasti 4, especially the presence of someone already cultivating the fields in Sicily when Ceres initially goes searching for her daughter (arua colentem, 487), and the reference at the end of the story, when Ceres is reconciled, to the way the fields gave a huge harvest after their period of being neglected and uncultivated.59 Still, especially in comparison with the version of the Ceres story in Metamorphoses 5, where there is no suggestion whatever of Ceres’ bereavement being a rupture between life before and after agriculture, we must be struck by the very different emphasis in the Fasti on the aetiological dimension of the Ceres myth. The gift of 58

Cf. Ovid’s querelis / implet, ut amissum (4. 481f.); see Fantham 1998: 47 on the links between this Ovidian passage and the Virgilian one, and on 481 f. for the links with the Lucretian simile. 59 Fasti 4. 617: largaque prouenit cessatis messis in aruis. See Fantham 1998, on 4. 559f.

252

Interpreting Sacrificial Ritual in Roman Poetry

agriculture is linked with mortality, both of humans and of the animals that humans live and work with. In the Ceres episode in Book 4, then, the indirect mythic explanation supplements and corrects the more overt didacticism of the exegesis section, and the overall impression is less at variance with the Ceres we see in Book 1 than we might initially think. In Book 1 the shocking nature of sacrifice is overt, and fully stressed, as Ovid concentrates all his efforts on denaturalising his audience’s familiarity with the institution, rather as Virgil activates a latent sense of disgust at the sacrificial evisceration of animals the humans care for.60 In Book 4 Ovid affects to ignore this perspective and to give another, more ameliorative view of the patron goddess of modern life, exempt from the nexus of killing, but the sacrificial imperative behind the life of civilization keeps breaking through. It breaks through in the form of the myth, with the Lucretian sacrificial simile for Ceres’ bereavement, with the reminder of her hatred of pigs, and with the treatment of the Triptolemus story as an aetiology of agriculture. It also breaks through more explicitly immediately after the end of the section on the Cerealia, when Ovid returns to the association of Ceres with animal sacrifice. Two short interludes totalling eight lines follow the Cerealia before Ovid gives us the Fordicidia of April 15 (4. 629–72), when a pregnant cow is sacrificed, ultimately a rite that started when Ceres failed (645). The sacrifice is to Tellus, so regularly linked with Ceres (634, 665). Two lines after the Fordicidia we encounter the final day of the Cerealia, April 19 (679–712), where Ovid tells us of how Ceres is honoured by the burning of foxes. Using the more open-ended Georgics as his point of departure, Ovid accentuates one of Virgil’s range of possibilities.61 In the Georgics, animal sacrifice is open to multiple interpretation, but Ovid concentrates, directly or obliquely, on one powerful Virgilian possibility: he represents sacrifice as a token of the loss of the Golden Age, as the life of agriculture involves humans in endlessly dominating the land and the animals that share it with us, and endlessly placating uncertain deities by giving them many varieties of that animal life. Human life is denaturalized, and sacrifice must be endlessly repeated in order to stave off the ever-present threat of having to pay the full consequences of that denaturalization.

60 Gale 2000: 105f. on Georgics 2. 194–6; cf. Ovid’s own evocation of this disgust in Fasti 4. 936: turpiaque obscenae (uidimus) exta canis. Ann Kuttner made me aware of the importance of the potential impact of this dimension of sacrifice. 61 Gale 2000: 108: ‘Ovid can be seen as making more explicit the tensions which I have been tracing in Virgil’.

Interpreting Sacrificial Ritual in Roman Poetry

253

The Specificities of Generic Preference The analysis of the two Ceres passages would obviously only be a beginning for a thorough-going study of sacrifice in the Fasti. One would need to follow up John Scheid’s fascinating study of the dialogues between the different kinds of sacrificial offerings in the cult of the Fratres Arvales (animals, plants, incense): the standard models are obsessed with animal sacrifice, and in English it sounds more than a little ridiculous to speak of ‘sacrificing’ cakes or vegetables, but Roman cult, and the Fasti, have a high degree of interest in non-animal sacrifice as well. One would also need to follow up Andrew Feldherr’s eyeopening discussion of how Ovid in the Metamorphoses treats the reader/citizen’s identification with the sacrificial victim, and see how his important findings work in the context of the Fasti.62 Nor have I touched on one of the most important aspects of sacrifice for Ovid, namely, as the arena for communication between humans and gods.63 Ovid casts his net very wide, but we need to remind ourselves how selectively this poem, or any other poem, treats or can treat the full range of the possible meanings of sacrifice. Richard Thomas, for example, makes much of the fact that we never see in the Georgics the full Burkertian sacrificial model of sacrifice followed by feasting. Yet different poetic and iconographic traditions vary greatly as to which elements of the full ritual they will represent, and each tradition is itself susceptible to evolution: the plastic arts, for example, show a dramatic change in their selection of the key moment of sacrifice in precisely the Augustan period, choosing the instant before the actual killing in preference to the procession.64 As far as sacrificial feasting itself is concerned, Gordon points out that representations of feasting are non-existent in public sacrificial sculpture of the Imperial period.65 Poetry, likewise, has its own variable preferences and emphases. Whatever they may do in the Aeneid and the Metamorphoses, where the sacrificial feasting of epic is quite common, Virgil in the Georgics and Ovid in the Fasti practically never link animal sacrifice and feasting explicitly. In the mention of goat sacrifice in Georgics 2 we have a description of the animal standing at the altar (et ductus cornu stabit 62

63 Feldherr 1997. Again, Rüpke 2001, for the general issues. Kuttner 1995: 131–5. See van Straten 1995 for Greek evidence. 65 Gordon 1990, 204: we may in fact have one depiction of feasting, in an image of the Vestal Virgins: Beard et al. 1998: 2. 150. Ann Kuttner draws my attention to the painting of a sacrificial banquet commissioned by Ti. Sempronius Gracchus to commemorate the victory of his slave army at Beneventum in 214 : on this painting, see now Koortbojian 2002. 64

254

Interpreting Sacrificial Ritual in Roman Poetry

sacer hircus ad aram, 395), and then of its exta being roasted on spits of hazel (pinguiaque in ueribus torrebimus exta colurnis, 396); but the actual killing is not mentioned, and the entire ritual is evoked in the future tense. As for Ovid in the Fasti, he does have one set-piece description of the full length of a sacrificial ritual, right down to the feast after sacrificial killing, when he describes the festival of the Terminalia in Fasti 2 (643–58).66 I take the point here to be related to the important fact that, according to the best ancient sources, originally this festival did not have blood sacrifice at all.67 Ancient authorities have it that Numa set up the institution of the Terminalia, like most of his other cults, as bloodless sacrifices.68 It is just possible that the Numa of the Metamorphoses returns from his lessons with Pythagoras to teach the Romans sacrificos ritus (15. 483) that are bloodless, even though I prefer the more ironic reading whereby Numa hears the learned speech of Pythagoras but does not believe it.69 In the Fasti, however, there is no doubt that Numa is a man of blood, who regularly performs blood-sacrifice, and is never seen performing bloodless sacrifice, even if he once averts human sacrifice by substituting an onion (3. 339f.).70 By contradicting a dominant tradition in this way, Ovid reinforces his theme that Roman civilization was normatively bound into Iron Age patterns of behaviour from the start. Or else, if he is suggesting that contemporary practice in the Terminalia is totally different from the original rites ordained by Numa, then we have another example of the disruptions in the continuity of this rite so finely elucidated by Barchiesi.71 After all, the entire point of Terminus is now moot both in place and time. The whole of the globe is now without bounds under Roman rule, as Ovid tells us (2. 683–4), so that Terminus no longer marks boundaries in space. And in terms of time, the terminal function of the Terminalia is a dead letter under the reformed Julian calendar. In the Republican calendar the Terminalia marked a cut-off point in February, after which the

66

Cf. Miller 1991: 120 for the unusual fullness of this description. Note the exceptionally full, practically Burkertian, description of sacrifice offered by Pythagoras in his didactic denunciation of the institution in Metamorphoses 15 (127–39), and the comical disruption of the sacrificial pattern in Fasti 2, when Romulus and Remus are called away to fight robbers while they are waiting for their sacrificial meat to cook, and Remus comes back first to finish the barbecue without Romulus (359–76). 67 My thanks to Martin Sirois for drawing this to my attention. 68 Plutarch, Roman Questions 267 C; Numa 8. 8; 16. 1. 69 Ovid, Metamorphoses 15. 73 f.: ora docta . . . sed non et credita. See Hardie 1997: 185 n. 14 for the first possibility; Barchiesi 2001b: 65–8 for the ironic reading. 70 For Numa’s blood sacrifice in the Fasti, see 3. 300, 4. 652, 671; cf. Barchiesi 2001b: 66. 71 Barchiesi 1997: 215–18.

Interpreting Sacrificial Ritual in Roman Poetry

255

intercalary month was inserted; in the Julian calendar, the Terminalia no longer terminate anything.72 A final example of sacrificial feasting in the Fasti is, once again, highly anomalous, and that is the Ourion myth in Book 5, where Jupiter, Poseidon, and Mercury feast with a poor old man, and finally all stand together and urinate on the hide of the plough-ox that has just been sacrificed to them and that they have just consumed (495–536). Note that Ovid, as a poet of Fasti, who is only allowed to say quae licet et fas est, finds urinating gods unsayable and leaves out the precise details.73 And even the emphasis on the feasting in this tale is to be explained as a result of its nature as a Callimachean Molorchus or Hecale story, where hospitality is vital.

Resisting Holism A weak way of dealing with the problems I have been sketching would be to say that different genres (epic, elegy, didactic, sculpture, and the different genres of ritual in their own right) have distinctive ways of doing things that refract their objects—in our case, sacrifice—in different ways. But this approach would immediately lay itself open to the criticism that it leaves unexamined the idea that there is an object out there to be refracted in the first place, when in fact mediation and representation and encoding are operative all the way down, on both sides of what we represent as the fence between ‘life’ and ‘literature’.74 This approach would also lay itself open to another criticism, namely, that it leaves unexamined a much larger historicizing assumption, the assumption that when literature engages with ritual or anything else it is participating in an identifiable larger system of meaning in which the terms are always set in advance by conditions which are more primary or authentic or real. Here I return to the problems I noted at the beginning of this paper. As I remarked there, historicizing approaches—and this is the more true, the more they are informed by structuralism and symbolic anthropology—almost inevitably posit a holistic mentalité, a global 72

No coincidence, then, that the next festival Ovid treats after the Terminalia is the Regifugia, which likewise used to mark the end of something, the monarchy, but has now also become a dead letter. Ovid marks the point at the end of the Regifugia, where the last line of the episode, dies regnis illa suprema fuit, is immediately followed by fallimur (2. 852–3): on the importance of reading straight on from line 852 to fallimur in line 853, see Reeve 1995: 507. 73 74 5. 532: pudor est ulteriora loqui. So, Conte 1994: 105–28.

256

Interpreting Sacrificial Ritual in Roman Poetry

system of meaning in which literature participates. At their most extreme, such historicizing approaches will have it that literature expresses the circumstances of its social production. Even when the issue of the cause and effect relationship is finessed, as it regularly is in New Historicism or Cultural Poetics, with their metaphors of ‘circulation’ or ‘negotiation’, we are still left with a model which posits a totalizing synchronic structure. This is why the governing trope of New Historicism is synecdoche. The power of synecdoche comes through very powerfully in Gallagher and Greenblatt’s introductory essay to Practicing New Historicism. They give, in effect, a charter for synecdoche, with such telling phrases as these: ‘If every trace of a culture is part of a massive text . . .’; ‘if an entire culture is regarded as a text . . .’.75 Later, when discussing the impact of Clifford Geertz, in an overt acknowledgment of the power of symbolic anthropology, they show how Geertz works out from a fragment to reveal the ramifications throughout the social system of thought: ‘Part of Geertz’s power was his ability to suggest that the multilayered cultural meanings by which he was fascinated were present in the fragments themselves’.76 The part stands for the whole, which is always somehow there, and primary. Our use of the terms ‘text’ and ‘context’ can pitch us into similar holistic traps. The language of ‘text’ and ‘context’ can help guard against mere formalism and aestheticism, but it can also keep us thinking of texts as parasitic upon something quintessentially more substantial and really there, and recoverable in that substantiality and reality, while likewise keeping alive the illusion that there is a recoverable cause and effect relationship between the context and the text it is often seen as producing.77 I am not denying that there is a cause and effect relationship between texts and the conditions of their social production, only that this relationship is recoverable. As David Perkins has shown in his profoundly unsettling book, any act of contextualizing is inevitably partial and arbitrary.78 Still, we cannot read without contextualizing, and the two poems I have been discussing here must be read by us, as by their original audiences, in a series of contexts before they can be interpreted. In their very different ways, the Georgics and the Fasti give Roman readers tools for thinking about patterns of action that otherwise for the most part they may well have taken for granted. If a Roman did want to speculate on the nature and meaning of sacrifice, these texts would have been indispensable. Romans would have encountered in these poems ways of

75 77

76 Gallagher and Greenblatt 2000: 14f. Ibid., 26. 78 Cf. Fowler 2000: 129. Perkins 1992: 121–52.

Interpreting Sacrificial Ritual in Roman Poetry

257

thinking that would not map directly onto their usual experience of ritual; but I hope I have made a case for suggesting that this is the usual state of affairs for the investigation of ritual. Different experiences and different analytical frames are bound to be incommensurable, to some degree. And we, too, have lessons to learn from the challenge of reading these poems through the spectacles of a ritualist. If we come to the poems with no model of sacrifice in our minds at all, we will find it very difficult to see the religious or cultural work they are doing. But if we come to these poems with a full-blown model of sacrifice in our minds, determined to see it exemplified, and convinced that the relationship between the literature and the ‘real’ category of ritual must be one of synecdoche, we will be disappointed, or, more probably, we will do violence to the poems’ specific strategies. We need to acknowledge not only that we cannot read without some kind of contextualizing model, but also that the imposition of such a model from another discipline can only be a preliminary heuristic step, for direct imposition of the model will fail to do justice to the way any given text may be working. It is not only literary critics who will be badly served if we jettison the category of the literary.

B I B L I OG R A P H Y Barchiesi, A. (1997). The Poet and the Prince: Ovid and Augustan Discourse. Berkeley—Los Angeles. Barchiesi, A. (2000). ‘Rituals in ink: Horace on the Greek lyric tradition’, in M. Depew and D. Obbink (eds.), Matrices of Genre: Authors, Canons, and Society. Cambridge, MA, 167–82. Barchiesi, A. (2001a). ‘The crossing’, in S. J. Harrison (ed.), Texts, Ideas, and the Classics: Scholarship, Theory, and Classical Literature. Oxford, 142–63. Barchiesi, A. (2001b). Speaking Volumes: Narrative and Intertext in Ovid and Other Latin Poets. Trans. M. Fox and S. Marchesi. London. Barchiesi, A. (2002). ‘The uniqueness of the Carmen Saeculare and its tradition’, in T. Woodman and D. Feeney (eds.), Traditions and Contexts in the Poetry of Horace. Cambridge, 107–23. Beard, M., J. North, and S. Price (1998). Religions of Rome. Cambridge. Bell, C. (1992). Ritual Theory, Ritual Practice. Oxford. Bell, C. (1997). Ritual: Perspectives and Dimensions. Oxford. Bloch, M. (1989). Ritual, History and Power: Selected Papers in Anthropology. London. Buxton, R. G. A. (1981). ‘Introduction’, in R. L. Gordon (ed.), Myth, Religion and Society: Structuralist Essays by M. Detienne, L. Gernet, J.-P. Vernant and P. Vidal-Naquet. Cambridge, ix–xvii.

258

Interpreting Sacrificial Ritual in Roman Poetry

Conte, G. B. (1986). The Rhetoric of Imitation: Genre and Poetic Memory in Virgil and Other Latin Poets. C. Segal (ed.). Ithaca. Conte, G. B. (1994). Genres and Readers: Lucretius, Love Elegy, Pliny’s Encyclopedia. Trans. . W. Most. Baltimore. Diamond, J. (1997). Guns, Germs and Steel: A Short History of Everybody for the Last 13,000 Years. London. Dyson, J. T. (1996). ‘Caesi iuuenci and pietas impia in Virgil’. CJ 91: 277–86. Elsner, J. (1991). ‘Cult and sculpture: sacrifice in the Ara Pacis Augustae’. JRS 81: 50–61. Fantham, E. (1992). ‘Ceres, Liber and Flora: Georgic and anti-georgic elements in Ovid’s Fasti’. PCPh 38: 39–56. Fantham, E. (1998). Ovid: Fasti Book IV. Cambridge. Feeney, D. (1995). ‘Criticism ancient and modern’, in D. Innes, H. Hine, and C. Pelling (eds.), Ethics and Rhetoric: Classical Essays for Donald Russell on his Seventy-Fifth Birthday. Oxford, 301–12. Feeney, D. 1998. Literature and Religion at Rome: Cultures, Contexts and Beliefs. Cambridge. Feldherr, A. (1997). ‘Metamorphosis and sacrifice in Ovid’s Theban narrative’. MD 38: 25–55. Feldherr, A. (2002). ‘Stepping out of the ring: repetition and sacrifice in the boxing match in Aeneid 5’, in D. S. Levene and D. P. Nelis (eds.), Clio and the Poets: Augustan Poetry and the Traditions of Ancient Historiography. Leiden, 61–79. Fowler, D. (2000). Roman Constructions: Readings in Postmodern Latin. Oxford. Gale, M. (2000). Virgil on the Nature of Things: The Georgics, Lucretius and the Didactic Tradition. Cambridge. Galinsky, K. (1996). Augustan Culture: An Interpretive Introduction. Princeton. Gallagher, C. S. and S. Greenblatt (2000). Practicing New Historicism. Chicago. Gordon, R. L. (1990). ‘The veil of power: emperors, sacrificers and benefactors’, in M. Beard and J. North (eds.), Pagan Priests: Religion and Power in the Ancient World. Ithaca, 199–231. Habinek, T. (1990). ‘Sacrifice, society, and Vergil’s Ox-born Bees’, in M. Griffith and Mastronarde (eds.), Cabinet of the Muses: Essays on Classical and Comparative Literature in Honor of Thomas G. Rosenmeyer. Atlanta, 209–23. Hamerton-Kelly, R. G., ed. (1987). Violent Origins: Walter Burkert, René Girard, and Jonathan Z. Smith on Ritual Killing and Cultural Formation. Stanford. Hardie, P. (1997). ‘Questions of authority: the invention of tradition in Ovid Metamorphoses 15’, in T. Habinek and A. Schiesaro (eds.), The Roman Cultural Revolution. Princeton, 182–98. Harrison, E. L. (1979). ‘The Noric plague in Vergil’s Third Georgics’. Papers of the Liverpool Latin Seminars 2: 1–65. Hinds, S. (1987). The Metamorphosis of Persephone: Ovid and the Self-Conscious Muse. Cambridge. Hinds, S. (2000). ‘Essential epic: genre and gender from Macer to Statius’, in M. Depew and D. Obbink (eds.), Matrices of Genre: Authors, Canons, and Society. Cambridge, MA, 221–44.

Interpreting Sacrificial Ritual in Roman Poetry

259

Kennedy, D. (1993). The Arts of Love: Five Studies in the Discourse of Roman Love Elegy. Cambridge. Koortbojian, M. (2002). ‘A painted exemplum at Rome’s Temple of Liberty’. JRS 92: 33–48. Kuttner, A. L. (1995). Dynasty and Empire in the Age of Augustus: The Case of the Boscoreale Cups. Berkeley. Latte, K. (1960). Römische Religionsgeschichte. Munich. Lefèvre E. (1976). ‘Die Lehre von der Entstehung der Tieropfer in Ovids Fasten 1, 335–456’. RhMus 119: 39–64. Lloyd, G. E. R. (1990). Demystifying Mentalities. Cambridge. Miller, J. F. (1991). Ovid’s Elegiac Festivals: Studies in the Fasti. Frankfurt/Main. Morgan, Ll. (1998). ‘Assimilation and civil war: Hercules and Cacus’, in H.-P. Stahl (ed.), Vergil’s Aeneid: Augustan Epic and Political Context. London, 175–97. Morgan, Ll. (1999). Patterns of Redemption in Virgil’s Georgics. Cambridge. Myers, K. S. (1994). Ovid’s Causes: Cosmogony and Aetiology in the Metamorphoses. Ann Arbor. Mynors, R. A. B. (1990). Virgil: Georgics. Oxford. Nilsson, M. P. (1955). Geschichte der griechischen Religion. Munich. Obbink, D. (1993). ‘Dionysus poured out: ancient and modern theories of sacrifice and cultural formation’, in T. H. Carpenter and C. A. Faraone (eds.), in Masks of Dionysus. Ithaca—London, 65–86. Perkell, C. (1989). The Poet’s Truth: A Study of the Poet in Virgil’s Georgics. Berkeley. Perkins, D. (1992). Is Literary History Possible? Baltimore. Porte, D. (1985). L’Étiologie Religieuse dans les Fastes d’Ovide. Paris. Putnam. M. C. J. (1979). Virgil’s Poem of the Earth: Studies in the Georgics. Princeton. Putnam. M. C. J. (2000). ‘Review of Morgan 1999’. Vergilius 46: 155–62. Reeve, M. (1995). ‘Conclusion’, in O. Percere and M. Reeve (eds.), Formative Stages of Classical Traditions: Latin Texts from Antiquity to the Renaissance. Spoleto, 497–511. Rüpke J. (2001). ‘Antike Religionen als Kommunikationssysteme’, in K. Brodersen (ed.), Gebet und Fluch, Zeichen und Traum: Aspekte religiöser Kommunikation in der Antike. Münster, 13–30. Scheid, J. (1990). Romulus et ses Frères: Le collège des frères arvales, modèle du culte public dans la Rome des empereurs. Rome. Smith, J. Z. (1987). ‘In Violent Origins ed. Hamerton-Kelly, 198. Thomas, Richard F. 1991. ‘The Sacrifice at the end of the Georgics, Aristaeus, and Vergilian closure’. Classical Philology 86: 211–18. Wilkins, J. B. (1994). ‘The Iguvine Tablets: problems in the interpretation of ritual text’, in C. Malone and S. Stoddart (eds.), Territory, Time and State: The Archaeological Development of the Gubbio Basin. Cambridge, 152–72. Wissowa, G. (19122). Religion und Kultus der Römer. Munich. Wyke, M. (2002). The Roman Mistress: Ancient and Modern Representations. Oxford. Zetzel, J. E. G. (1983). ‘Catullus, Ennius and the poetics of allusion’. Illinois Classical Studies 8: 251–66.

IV Rome and Italy

10 The Ritual of Centuriation Frontinus, Hyginus Gromaticus, and Centuriation Daniel J. Gargola

At the opening of their treatises on boundary paths or limites, Frontinus and Hyginus Gromaticus, two Roman writers on surveying techniques, present descriptions of a rite, and they regularly refer to this rite or to its consequences in the remainder of their works. In this essay, we will seek to answer how surveying could be regarded as a rite; how the passages in question relate to the larger texts in which they are included; what they reveal about the relationship, if any, between Roman surveying practices and Roman public cult. In the process, such an investigation may also shed some light on the class of Roman texts that modern scholars often deem ‘antiquarian’.

FRONTINUS, HYGINUS GROMATICUS, AND CENTURIATION Frontinus’ and Hyginus Gromaticus’ texts form part of a collection, the Corpus Agrimensorum Romanorum, that addresses many aspects of surveying.1 Works in the Corpus, which vary greatly in date, address a range of concerns in public and private law: measuring and marking land, resolving disputes between private individuals over boundaries or 1 For the texts, see Blume, Lachmann, and Rudorff (eds.) 1848–52; Thulin 1913; Behrends, Clavel Lévêque, Conso, et al. 1998; Clavel Lévêque, Conso, Gonzales, et al. 1996; Campbell 2000. In this essay, works will be cited according to Thulin’s edition (Th) if they are included in it, and by Campbell’s text (C) if not.

Daniel J. Gargola, The Ritual of Centuriation: Frontinus, Hyginus Gromaticus, and Centuriation In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0011

264

The Ritual of Centuriation

the sizes of fields, and setting the proper form for boundaries. One purports to be a prophecy of the Etruscan divinity Vegoia warning against tampering with boundary markers. Frontinus’ and Hyginus Gromaticus’ works specifically addressed the establishment of limites. Both wrote when the techniques they were describing were passing, or had passed, from use, and both were clearly dependent on earlier sources. Frontinus may be the Flavian governor of Britain who also wrote the Strategemata and De aquis urbis Romae; the only evidence is the name.2 Hyginus, usually called Gromaticus to distinguish him from another Hyginus in the Corpus, is usually placed between the beginning of the third quarter of the first century  and the early third century, but his style and the rare literary allusion provide the only evidence.3 Frontinus and Hyginus Gromaticus begin with descriptions of the origins of centuriation. Frontinus’ opening lines (pp. 10. 20–11.8 Th) put the establishment of limites firmly in the sphere of divination: Limitum prima origo, sicut Varro descripsit, a[d] disciplina[m] Etrusca[m]; quod aruspices orbem terrarum in duas partes diuiserunt, dextram apellauerunt, septentrioni subiacere, sinistram quae a meridiano terra esse occasum, quod eo sol et luna spectaret, sicut quidam {carpiunt architecti delubra in occidente recte spectare scripserunt. aruspices altera[m] linea[m] a septentrione ad meridianum diuiserunt terram, a me[ri]dia[no] ultra antica, citra postica nominauerunt. The first origin of limites, as Varro has described, is in the Etruscan disciplina, because the haruspices divided the world into two parts: they called ‘right’ that part which lay to the north and ‘left’ that part of the world which was to the south, looking west because the sun and the moon looked in that direction; similarly some {architects have written that temples should rightly look to the west. The haruspices divided the earth with another line from north to south, and, from the middle, they called the far side of the line antica, the near side postica.

Frontinus (p. 11. 9–14 Th) goes on to claim that the Roman maiores adapted this practice for their own ends: Ab hoc fundamento maiores nostri in agrorum mensura uidentur constituisse rationem. primum duo limites duxerunt; unum ab oriente in occasum, quem uocauerunt decimanum; alterum a meridiano ad septentrionem,

2

On Frontinus’ identity, see Campbell 2000: xxvii–xxxi; Behrends et al. 1998: viii–ix. For Hyginus Gromaticus’ date, see Campbell 2000: xxxv–xxxvii; Clavel Lévêque et al. 1996: xii–xiii. 3

The Ritual of Centuriation

265

quem cardinem appellauerunt. decimanus autem diuidebat agrum dextra et sinistra, cardo citra et ultra. From this basis our ancestors evidently established their system of measuring land. First they drew two limites, one from east to west, which they called the decumanus, the other from south to north, which they called the kardo. So, the decumanus divided the land by a ‘right’ and a ‘left,’ the kardo by a ‘this side’ and a ‘beyond.’

Hyginus Gromaticus begins with a proclamation of centuriation’s virtues, praising its ease of use and its beautiful appearance in the field and on maps, and then describes the origins of centuriation in very similar terms to Frontinus’ (pp. 131.8–132.12 Th): Constituti enim limites non sine mundi ratione, quoniam decumani secundum solis decursum diriguntur, kardines a poli axe. unde primum haec ratio mensurae constituta ab Etruscorum haruspicum [uel auctorum habet, quorum artificium] disciplina; quod illi orbem terrarum in duas partes secundum solis cursum diuiserunt, dextram appellauerunt quae septentrioni subiacebat, sinistram quae ad meridianum terrae esset, occasum, quod eo sol et luna spectaret; alteram lineam duxerunt a meridiano in septentrionem, et a media ultra antica[m] citra postica[m] nominauerunt . . . . Ab hoc exemplo antiqui mensuras agrorum normalibus longitudinibus incluserunt. primum duos limites constituerunt: unum, qui ab oriente in occidentem dirigeret. hunc appellauerunt duo[de]cimanum ideo, quod terram in duas partes diuidat et ab eo omnis ager nominetur. alterum a meridiano ad septentrionem; quem kardinem nominauerunt a mundi kardine[m]. For limites are established on a principle not unassociated with cosmology, since the decumani are directed according to the course of the sun, the kardines by the axis of the world. Whence, in the first place, this system of measuring was established by the learning of the Etruscan haruspices. For they divided the world into two parts according to the course of the sun, and called ‘right’ the area lying toward the north, and ‘left’ the area lying toward the southern part of the earth, west, because the sun and moon look there. They drew another line from south to north, and named from the middle what was beyond the line antica (‘in front’), and what was on the near side, postica (‘behind’) . . . . On this pattern the ancients enclosed measured areas of land with right-angled lengths. First, they established two limites: one to be directed from east to west they have called the duodecimanus because it divides the earth into two parts and all the land is named after it; the other to be directed from south to north they have called the kardo, after the hinge of the world.

Both authors were describing a form of surveying known to modern scholars as centuriation, a technique developed and deployed in the

266

The Ritual of Centuriation

colonies and viritane assignments that accompanied Roman expansion in Italy.4 The process of surveying was highly formulaic and the surveyor’s instruments were simple. Chief among them were the surveyor’s cross or stella and the more complex groma;5 with them, surveyors laid out straight lines and formed right angles. To these were added measuring rods, perticae or decempeda, which enabled surveyors to measure the standardized distances that were essential to their craft. The Latin groma, it should be noted, derives from the Greek gnomon through an Etruscan intermediary. Initially, field systems varied in form from project to project, but soon basic patterns appeared. Surveyors first established a series of parallel boundary paths or limites, collectively known as decumani, that were separated from each other by a distance that was standardized for each project. Then they established another set of lines (not paths) or rigores, defined only by boundary markers, that crossed the decumani at right angles. The result was a series of rectangular modules, which would in turn be subdivided, although in a less regular fashion, to form individual allotments. These modules were called scamna when their long sides ran parallel to the decumani and strigae when their short sides did so. Traces of scamnation and strigation have been detected at early colonies such as Cales (334 ), Alba Fucens (303), and Cosa (273). Centuriation was more elaborate. When centuriating a field, surveyors first laid out an especially broad road, the decumanus maximus, which in theory ran from east to west. Then they defined a north-south road, almost as wide, that ran perpendicular to the first, the kardo maximus. Using these axes as a base, surveyors next laid out a series of parallel kardines and decumani by measuring set distances, usually twenty actus (c.708 m), to each side of the kardo maximus and the decumanus maximus. The result was a regular grid of paths and roads, separated by a standard distance and intersecting everywhere at right angles, that created a number of square modules or centuriae, each of two hundred iugera. Settlers’ plots later were created within these centuriae. The crossing axes of the centuriation grid possessed a special terminology that, as we will later see, connected these lines with some of the central ritual practices of the Roman state. Ideally, kardo maximus and decumanus maximus were oriented toward the cardinal directions. From the perspective of an observer looking down the decumanus maximus, surveyors made a four-fold division of the field which allowed the 4 For the development of the techniques, see Hinrichs 1974; Gabba 1985: 265–84; Chouquer and Favory 1991; Schubert 1996. 5 See Lewis 2001.

The Ritual of Centuriation

267

identification of individual centuriae and the plots within them. Thus, from the observer’s perspective, any one decumanus was either to the ‘right’ (dextra) or to the ‘left’ (sinistra) of the decumanus maximus, while, from the same perspective, an individual kardo was either ‘on this side’ or ‘behind’ (kitra or postica) or ‘beyond’ or ‘in front of ’ (ultra or antica) the kardo maximus. By counting the number of roads intervening between any given centuria and the main axes, the surveyors were able to identify the location of any square: thus an inscribed Gracchan boundary marker (ILS 24) proclaimed its location as the intersection of the eleventh limes this side (kitra) of the kardo maximus with the first limes to the left (sinistra) of the decumanus maximus. Frontinus’ and Hyginus Gromaticus’ versions of the origins of centuriation refer to the organizing axes and not to the centuriae that they contained; the shift from scamnation/strigation to centuriation centers in part around the use of these axes. The date for the appearance of centuriation can be set only within broad limits. Scholars often assume that the division of a field into centuriae and its organization by orthogonal axes formed a single system from its origins, but this need not have been the case. The earliest known centuriae appear between the mid-third and the early second centuries . Unfortunately, the physical traces of centuriation networks do not permit the easy detection of a kardo maximus and a decumanus maximus. The axes do not antedate the appearance of centuriae, since fields organized by scamna or by strigae apparently had no roads perpendicular to the decumani that passed completely through the network of limites and rigores. The Gracchan boundary marker of 132  (ILS 24), which records its position with respect to a kardo maximus and a decumanus maximus, provides the earliest evidence and a terminus ante quem. Centuriation, it should be emphasized, was a Roman technique, not an Etruscan one. While the Etruscans did use similar instruments to create the field systems in their settlements in the Po valley (as did the Greeks in their colonies in southern Italy), they did not survey in as regular a manner as the Romans did, nor did they organize their fields by orthogonal axes, just the aspect of centuriation that Frontinus and Hyginus Gromaticus linked with Etruscan practices. The failure of the Etruscans to use centuriation, however, does undermine the two authors’ claims.6 Both state explicitly that the Romans developed centuriation ‘on the basis of ’ or ‘on the example of ’ the haruspices’ techniques of divination.

6 Schubert 1996: 7 takes the Etruscans’ failure to use centuriation as a reason to dismiss Frontinus’ and Hyginus Gromaticus’ claims as antiquarianism.

268

The Ritual of Centuriation

What is the significance of centuriation’s defining forms? Preparing land for distribution by laying out limites clearly served—and probably served very well—to provide a relatively simple means of forming a large number of plots of equal size; centuriation, then, was a utilitarian operation. Yet Frontinus and Hyginus Gromaticus, as we have seen, associate it with religious techniques of some prestige. Some scholars, acting on the, often unexpressed, assumption that religion and technology form different spheres, have denied the existence of any connection between the two areas of activity, beyond a few superficial resemblances.7 Accepting the possibility of a link, one faces new problems and complications. Some scholars, operating on the once-dominant model that viewed Roman religion as a living force only in the earliest periods of Roman history, sought to place the origins of centuriation in the same early period, but firmer dating for the beginnings of centuriation has made such a chronological scheme untenable. More recent studies suggest that Roman religion long retained the force to shape actions, beliefs, and perceptions, making possible the formation of links between augury and surveying at a much later date.8 Still, the existence of a tie remains uncertain. Indeed, the very length of the debate has been part of the problem, since it has spread over periods in which modern conceptions of Roman religion and of the history of centuriation have changed markedly. Over the years, scholars have put forward several secular explanations for centuriation and its forms. One especially influential modern school associates centuriation and scamnation/strigation with lands under different legal regimes. Several Greek and Roman authors report that Romulus, when founding Rome in much the same manner that a colonial commissioner founded a colony, gave to each citizen of the new city two iugera of land, the so-called heredium, and some modern scholars have suggested that the modules of centuriation were called centuriae because they contained one hundred plots of this size.9 Because

7 Thus Hinrichs (1974: 78–84) rejects any link between augury and centuriation and states that he will seek a pragmatic source for the techniques; Weinstock (1946: 101–29, esp. 129) separates ‘theological speculation’ from ‘gromatical practice’; Castagnoli (1971: 78–81) suggests that the technical writers imposed cosmic theories on standard surveying practices long after their development. 8 For a more recent view of Roman religion, see Scheid 1985; Beard et al. 1998. 9 For the heredium and its treatment by scholars, see Gabba 1978: 253–8; for the (possible) relationship between the heredium and centuriae, see Gabba 1985: 265–7. Frontinus (pp. 13.13–14.10 Th) and Siculus Flaccus (pp. 117. 26–118. 3 Th) both claim that a centuria was formed from one hundred two-iugera plots, but they do not explicitly connect these plots with Romulus or with Rome’s foundation.

The Ritual of Centuriation

269

of this possible link between centuriae and the foundation myths of Rome, some scholars suggest that centuriae defined lands that would remain Roman in citizen colonies and in viritane assignments, while scamna or strigae marked tracts that would cease to be Roman when granted to settlers in Latin colonies.10 The examination of actual field systems, however, have made it clear that the different forms of surveying marked chronological stages in the development of Roman practices, and that centuriation probably was used in some third-century Latin colonies.11 Other scholars have seen in centuriation an attempt to organize rationally the patterns of use in the countryside in a manner similar to Greek orthogonal town-planning, or they have viewed it as an essentially military activity, for the Romans certainly used the same instruments and similar techniques when establishing the army’s camps.12 But second-century camps, as Polybius (6. 41–2) reveals, were not organized around orthogonal axes—only one of the defining routes went completely through the camp—and it is just this attribute of centuriation for which Frontinus and Hyginus Gromaticus claim religious associations. Processes, however, can have pragmatic functions and ideological or symbolic ones at the same time, and a too rigid division of Roman practices into largely modern categories of thought can do violence to the ways the Romans themselves viewed the matter. As we will see, the resemblances between the orthogonal axes of centuriation and certain techniques of divination are too close to be dismissed easily.

CENTURIATION AND AUGURY Although Frontinus and Hyginus Gromaticus placed the roots of centuriation in certain Etruscan practices, other scholars have seen close connections between centuriation and Roman augury. The early history of these Roman and Etruscan techniques is largely lost to us—and it may well have been obscure to Roman authors. In the area of our primary focus—the orthogonal axes and the vocabulary associated with them—Roman and allegedly Etruscan procedures may have been very similar. By the second century , if not earlier, the Romans certainly 10

See, for example, Rudorff 1992: 2. 227–464, esp. 296–7, 419. Note that Frontinus (p. 1. 6–16 Th) identifies land that has been divided and allocated by centuriae, scamna, and strigae with colonies in general, while he distinguishes scamna and strigae from centuriae by identifying the first two as the ancient custom (more antiquo). 12 For centuriation as a military activity, see Hinrichs 1974: 78–84. 11

270

The Ritual of Centuriation

had assimilated some Etruscan rites, while attributing some Roman customs to the Etruscans.13 At the very least, the practices that are the objects of our concern, in the ways that the Romans used them, may well have come to form a single system (see below). The Roman priestly college of the augurs occupied a central position in the republican order, where they were the guardians and masters of techniques intended to identify legitimate public actors and actions and to secure the consent of the gods for their deeds. To do this, the augurs possessed a science that included various techniques of divination and the rituals and rules needed to prepare and regulate templa, spaces needed for a range of public and sacred acts.14 The possible links between surveying and augury cluster around these templa and the way in which they organized space. Templa varied markedly in form and in function. Varro provides the most complete overview, but it is unclear how complete his list of three categories is and where forms of templa known from other sources fit among them.15 Varro’s templum in caelo was the visual field, defined against the sky, into which the auspicant looked to search for divine signs such as lightning. Other templa, probably Varro’s templum in terra, were defined against the land by an auspicant looking down over the surrounding countryside from a high place to search for signs by observing the flight of birds. Enclosures established on the ground to serve as places from which observers would create these templa may have shared the category of templum in terra with the visual fields that were created from them; the late republican templum at Bantia, a clearly defined rectangular space containing stone markers to aid in the construction of an auspicant’s field of vision and to assist in the interpretation of the signs observed there, is the best-known example.16 Another form of templum, which may not have been included in Varro’s list, is a clearly defined rectangular enclosure on the ground—a locus inauguratus—that served as a place where public officials would perform important functions, and also as the location for altars, temples, and shrines.

13 See Dumézil 1970: 661: ‘The Romans speaking of their own origins . . . were inclined to stamp their practices with the respectable Etruscan label, which gave them the prestige of antiquity and a kind of intellectual warranty’. 14 See Catalano 1960: 211–320; Linderski 1986: 2146–312. 15 See Varro De lingua Latina 7. 6: templum tribus modis dicitur: ab natura, ab auspicando, a similitudine; natura in caelo, ab auspiciis in terra, a similitudine sub terra. (‘templum’ is spoken of in three ways: from nature, from auspication, from resemblance; from nature in the heavens, from the auspicia on the earth, from resemblance beneath the earth’. On templa in general, see Catalano 1960: 248–319; Catalano 1978: 440–553, esp. 467–79; Linderski 1986: 2256–96. 16 See Torelli 1966: 293–315; Torelli 1969: 9–48.

The Ritual of Centuriation

271

Resemblances between templa and centuriation networks become clearer upon examining the ways augurs and auspicants created their templa and organized them internally. According to Varro, the templum in caelo was divided into four quarters with the eastern one called ‘left’, the western quarter called ‘right’, the southern section identified as antica, and the northern, postica.17 In Cicero’s depiction of the legendary augur Attus Navius’ performance of the augurium stativum, a rite used to determine where something was located or ought to be located, the augur stood in the middle of the vineyard facing south and divided it into four quarters.18 In Dionysius of Halicarnassus’ account (Roman Antiquities 4. 59. 2–61. 2) of a Roman embassy sent to consult the Etruscan expert Olenus Calenus about a prodigy discovered on the Capitol, the ambassadors were warned that the expert would seek to appropriate the sign for his city by creating a replica of the templum at Rome in which it was found: ‘. . . he will circumscribe with his staff some piece of ground, and then he will say to you: “this is the Tarpeian hill, this is the part that faces the east, this the part that faces the west, this point is north and the opposite is south”.’19 In the context of augury, it should be noted, the word regiones has as its primary meaning the lines that both subdivided a space and separated it from the external world; secondarily, the word denotes the parts of a space that these dividing lines created. Our most detailed example, and the one that probably best preserves the details and vocabulary of the rite, is also the most obscure: the creation of a templum in terra from the auguraculum on the arx at Rome. Here, Livy’s rendition of the inauguration of Numa Pompilius, the second king of 17 See Varro, De lingua Latina 7. 7: eius templi partes quattuor dicuntur, sinistra ab oriente, dextra ab occasu, antica ad meridiem, postica ad septemtrionem. 18 Cicero, De Divinatione 1. 31: ‘quid? multis annis post Romulum Prisco regnante Tarquinio quis veterum scriptorium non loquitur, quae sit ab Atto Navio per lituum regionum facta discriptio? qui cum propter paupertatem sues puer pasceret, una ex iis amissa vovisse dicitur, si recuperasset uvam se deo daturum quae maxima esset in vinea; itaque sue inventa ad meridiem spectans in vinea media dicitur constitisse, cumque in quattuor partis vineam divisisset trisque partes aves abdixissent, quarta parte, quae erat reliqua, in regiones distributa, mirabili magnitudine uvam, ut scriptum uidemus, inuenit’. (‘Many years after Romulus, when Tarquinius Priscus was the king, which of the ancient writers fails to mention that division of the regions imposed by Attus Navius with his lituus (augur’s staff)? As a boy, he was pasturing his pigs (that’s how poor he was) when one got lost and he vowed that if he recovered the pig, he would give to the god the biggest bunch of grapes in the vineyard. so when the pig was found, he is said to have stood in the middle of the vineyard, looking to the south, and when he had divided the vineyard into four parts and the birds had rejected three of the four, he divided the remaining one into four sections and found (so we read) a bunch of miraculous size.) 19 For an examination of Dionysius’ understanding of augural practices, see Vaahtera 2001: 94–143.

272

The Ritual of Centuriation

Rome, is the central text. According to Livy, the future king took his seat facing south, while the augur sat next to him facing east. Then, looking out over the city and the countryside, the augur spoke a prayer and marked with his lituus a line from east to west, calling ‘right’ the part to the south and ‘left’ the part to the north, and fixing in his mind a landmark opposite to him and as far away as the eye could reach.20 Varro (De lingua Latina 7. 8–9) has preserved the verbal formula used to define this templum, and it fits with Livy’s description.21 The augur first clearly defined the outer limits of his templum—according to Varro, using trees to set his fines— and then went on to set the regiones that subdivided the field. Livy seems to describe only a single division of the templum into a left and a right, but the pomerium itself crossed the templum, dividing it into an upper and a lower register.22 The augural templum at Bantia, which should replicate the divisions of the templum created from it for the auspices, clearly was divided into four quarters by east/west and north/south lines. These orthogonal axes were one of the defining features of a templum.23 Some scholars view templa as bounded spaces separated from the external world for some public or augural end. This characterization requires some qualification. Many, and probably most, templa were clearly and firmly bounded by the actions of the person who created them, but some templa may not have been defined in this way: Varro characterized the templum in caelo as ab natura (‘by nature’), which apparently means that the augur himself constructed no boundaries for it, while Cicero depicted Attus Navius only as setting the internal divisions of his templum—the vineyard itself seems to have provided the boundaries, not the augur.24 The orthogonal axes also were found in certain kinds of locus inauguratus: a stella, the sign of two crossing lines, was put on the entrances to all inaugurated places, with the words antica and postica inscribed on it, while the internal divisions survived into the temples that sometimes occupied templa.25 20 Livy 1. 18. 6–10, esp. 7–8: augur ad laeuam eius capite velato sedem cepit, dextra manu baculum sine nodo aduncum tenens, quem lituum appellarunt. inde ubi prospectu in urbem agrumque capto deos precatus regiones ab oriente ad occasum determinauit, dextras ad meridiem partes, laeuas ad septentrionem esse dixit; signum contra, quoad longissime conspectum oculi ferebant animo finiuit. 21 Norden (1939: 3–106) clarified the obscurities of this formula. 22 23 Thus Linderski 1986: 2279. Thus Catalano 1960: 289. 24 For these points, see Linderski 1986: 2266, 2287 n. 561. 25 For the stella, see Festus 476L: stellam significare ait Ateius Capito laetum et prosperum, auctoritatem secutus P. Servili auguris [stellam], quae ex lamella aerea adsimilis stellae locis inauguratis infigatur; Dolabella p. 224. 1–3 C: quare per aedes publicas in ingressus antiqui fecerunt crucem, ANTICA, et POSTICA? quia aruspices secundum aruspicium in duabus partibus orbem terrarum diuiserunt; una parte ab oriente in occidentem, alia

The Ritual of Centuriation

273

It is just this aspect of the augurs’ science that has the clearest links with centuriation. Surveyors too organized their fields by two orthogonal axes, the kardo maximus and the decumanus maximus, oriented according to the cardinal directions, and they named the divisions that these lines created in a similar fashion. Indeed, the late republican jurist and augur, Servius Sulpicius Rufus, is reported to have called those lines in limitation networks away from the rising sun postica.26 One should note, however, that our sources do not provide the augural names for the lines that created the internal divisions of a templum—kardo and decumanus, so prominent in surveying, are not known to have been augural terms— but Frontinus (p. 12. 2–3 Th) and Hyginus Gromaticus (p. 132. 11–12 Th) both note that the word kardo, the primary meaning of which is ‘hinge’, denotes the north-axis around which the heavens rotate, making the word a plausible candidate for the role. Much of the attention of those scholars who deny any significant links between centuriation and augury has been focused on a number of apparent discrepancies between surveying techniques and augural practices. On closer examination, however, these problems disappear. Centuriae and centuriated land were not templa. In a templum created for auspication, for example, the directions possess different values, so that, a meridiano in septentrionem. For the parts of a temple, see Festus (Paulus) 244–5 L: posticum ostium dicitur in posteriore parte aedium. ceterum antiqui etiam vicinum habitantem ad posteriorem partem aedium sic appellarunt. denique et quae ante nos sunt antica, et quae post nos sunt postica dicuntur, et dexteram anticam, sinistram posticam dicimus. sic etiam ea caeli pars, quae sole inlustratur ad meridiem, antica nominatur, quae ad septentrionem, postica; rursumque dividuntur in duas partes, orientem atque occidentem. (‘For the stella, see Festus 476L: Ateius Capito, on the authority of the augur P. Servilius, says that that stella means “fortunate and favourable”, just like the star, affixed to inaugurated places and made from a small bronze plate. Dolabella p. 224. 1–3 C: which is why the ancients put a cross at the entrances to public buildings, marked “Antica” (in front) and “Postica” (behind), because the haruspices according to their tradition, divided the world into two parts, one from east to west, the other south to north. For the parts of a temple, see Festus (Paulus) 244–5 L: The door in the rear part of buildings is called the ostium posticum; on the other hand, the ancients also called a neighbour posticum, who lived in the back part of a building. So, both what is in front of us (ante) is called antica and what behind us (post) postica: also, we call the right antica and the left postica. Likewise, that part of the sky lit by the sun from the south is called antica, the part from the north, postica; and those too are divided into two parts, east and west.’) Linderski (1986, 2289 n. 568) suggests that the augur P. Servilius is to be identified with the P. Servilius who was consul in 48 . Prosdocimi (1991: 37–43) argues that there was no similarity between the surveyors’ stella and the stella placed on inaugurated places, since the former had the shape of a cross, while the latter, he suggests, was rectangular, with the corners forming points that were star-like. But see Dolabella above. 26 Festus p. 262 L: posticam lineam in agris dividendis Ser. Sulpicius appellavit ab exori.

274

The Ritual of Centuriation

in the interpretation of signs, left is considered more favorable than right. Surveyors, on the other hand, did not value some quadrants of centuriated land over others—all arable centuriae possessed equal value. And a templum that was also a locus inauguratus received certain augural rituals of inauguration, none of which seem to have been performed around centuriae. Because of this, some scholars have argued that there could be no link between centuriation and the augurs’ science, since their operations did not produce the same results. Yet this should not come as a surprise. The ritual designation of the regiones that set the external boundaries and the internal divisions of a templum was only part of a larger process.27 Auspicants turned fields of view into templa by defining gestures, but also by reciting the rules that would govern its use and set its purpose, a formulaic legum dictio. And when inaugurating a locus inauguratus, the augur performed the additional rites of a liberatio and an effatio. Templa, then, were spaces with analogous internal organizations that were given their special character by a lex or by the performance of further rites. Centuriated spaces, which possessed a similar organization, would have possessed no special character unless it too was specified, and brought into being, by the surveyors or the officials under whose direction they worked. Thus, the relationship between centuriation and augury—if, indeed, there was such a relationship—is a matter of the analogous organization of spaces, and not of identity. If so, the closest analogy may well have been with the templa created in the augurium stativum, which seem to have had no hierarchical valuation of their constituent quarters, and which possibly involved no formal definition of the outer limits. Then there is the matter of orientation and the way that centuriation varies, or may have varied, from supposed norms. In this, it is important to keep in mind that liturgical orientations may differ in individual cases from actual orientations. The templum constructed from the Roman arx seemingly was oriented toward the south-southeast rather than toward the east. According to Vitruvius (4. 5. 1–2), architects could give temples an actual orientation that differed from the preferred westward one if street plans or the nature of the site required it. And Frontinus (p. 14. 14–21 Th) and Hyginus Gromaticus (p. 135. 1–14 Th) note that actual centuriation grids sometimes vary from the norms they had just set out when the lie of the land or the location of major roads would recommend

27 Catalano 1978: 468–70; Catalano (1960: 305–6) stresses that centuriae were not templa, that the creation of orthogonal axes was, of itself, not sufficient to create a templum, and that some spaces organized by orthogonal axes were, in fact, never templa.

The Ritual of Centuriation

275

it. Orientations established through formal procedures, in other words, may often have been primarily a matter of rite and of convention. Within these conventions, however, the augurs preferred certain orientations, which seemingly differ from those of Frontinus’ and Hyginus Gromaticus’ Etruscans. The augurs favored eastward and southward orientations. Thus, the parts of Varro’s templum in caelo are named from the perspective of a southward-facing observer, as are the segments of Attus Navius’ templum for the augurium stativum. But the templa created from the Roman arx and from the auguraculum at Bantia both have an eastward orientation. As an additional complication, Jupiter, for the purpose of the auspices, is usually viewed as looking over the visual templum from the north—Livy’s Numa occupies the same position—so that the quarters of a templum in terra would possess different names from Jupiter’s perspective and that of the auspicant. Frontinus’ and Hyginus Gromaticus’ haruspices—and the surveyors who followed their example—used a westward orientation. This apparent deviation from augural practice could be taken as a sign that the two authors were mistaken, engaging in what might be viewed as antiquarian speculation, or that the technique that served as a model for centuriation really was Etruscan, without links to Roman augury. Yet this need not have been the case. Both Frontinus and Hyginus Gromaticus explicitly link the westward orientation of a centuriation network not only with the Etruscan haruspices but also with Roman architecti setting the orientation of temples. Festus gives to the heavens an orientation from north to south—just as in Varro’s templum in caelo—and he goes on to note that, because of this, the parts of the temple were named according to an eastward orientation—the same combination of celestial and human orientations found in Livy’s templum from the arx.28 Vitruvius (4. 5. 1–2) adds an additional layer of complexity. For him, temples should be constructed, if possible, so that the structure and the cult statue in its cella both face the west, in order that those who sacrifice and pray at the altar in front of the temple might look toward the temple, its statue, and the eastern portion of the heavens. Here, then, officiants look to the east, while the temple reverses this orientation. The divergent practices of augurs, haruspices, architects, and surveyors clearly form part of a reasonably unified system. The Romans, after all, encountered the haruspices as foreign experts, possessing some techniques that overlapped with Roman ones, who were summoned by the Senate to consult about matters of concern to it. In these circumstances,

28

Festus (Paulus) 244–5 L. For the text, see n. 25 above.

276

The Ritual of Centuriation

it is possible that the haruspices working as consultants adjusted their practices in some way to fit Roman demands. Now, east and south are the augural directions. The haruspices, who are not augurs or even Romans, face to the west, as do architects when designing temples. Temples, it should be noted, are connected to augury, but not fully part of it. Officials built temples within inaugurated spaces, so that the templa in which they stood largely determined the actual orientation of the temple, but the priestly college of the pontifices, not the augurs, had charge of the rules governing the dedication and use of temples. Roman experts conceivably gave a westward orientation to the haruspices’ spaces, the architects’ temples, and the surveyors’ field systems to separate them from augural templa as constructs related to templa, but separate from them in crucial ways. Perhaps for this reason Frontinus and Hyginus Gromaticus, or their sources, linked centuriation with haruspicy and with temples, not with augury. It should be noted that, whatever their preferred orientations, augurs, surveyors, and haruspices are described as setting their east-west axis first and doing so ‘from east to west’ (ab oriente ad occasum);29 while orientations varied, some of the conventions of the constituting rite apparently did not. Frontinus’ and Hyginus Gromaticus’ orientations, then, are consistent with known Roman practices. Writers on surveying, however, are sometimes inconsistent in their orientation. Frontinus and Hyginus Gromaticus orient their fields to the west, but Servius Sulpicius Rufus assumes an eastward orientation, since he claimed that the part of the field called ‘behind’ (postica) was away from the rising sun.30 Furthermore, Frontinus and Hyginus Gromaticus sometimes describe decumani as facing to the east, the favored orientation of augurs, but supposedly not for haruspices, architects, and surveyors.31 There are two possible solutions to this problem. When constructing their decumanus maximus, surveyors often established this line by sighting their groma at dawn toward the rising sun, the easiest way of determining an approximate east.32 Vitruvius’ temples 29

See Livy 1. 18. 8: regiones ab oriente ad occasum determinauit; Dolabella p. 224. 1–3 C: ab oriente in occidentem. See also Frontinus pp. 10.20–11.8 Th; Hyginus Gromaticus 131. 8–132. 12 Th. 30 See Festus p. 262 L (quoted above, n. 26). 31 See, for example, Frontinus 12. 1–4 Th: postea hoc ignorantes non nulli aliud secuti, ut quidam agri magnitudinem, qui qua longior erat, fecerunt decumanum. quidam non ortum spectant, sed ita aduersi sunt, ut sint contra septentrionem; 12. 13–14 Th: [limites] qui spectabant in oriente, dicebant prorsos; Hyginus Gromaticus 132. 18–19 Th: reliquos limites fecerunt angustiores, et qui spectabant in orientem prorsos . . . appellauerunt. 32 Indeed, Hyginus Gromaticus will include a lengthy digression on why this practice is inadequate for determining true east, and he will include what he claims to be a better method.

The Ritual of Centuriation

277

possessed two different orientations along the same axis: one that of the temple and its cult statue, which faced to the west, and the other that of worshipers at the temple, who faced east, as would have the augurs who created the templum in which it stood. Some of the confusion over the orientation of a centuriation grid may well be the result of a similar situation. Surveyors looked to the east as they set the first of their main axes, while the conventional orientation of the grid would have been in the opposite direction.33 Thus, Roman authors could speak of the orientation of a grid from two related perspectives. A second explanation involves a change in preferences. Hyginus Gromaticus (pp. 134. 15–21 Th) notes that centuriation networks face either to the east or to the west. The westward orientation, he claims, is the earlier, since at that time, architects held that temples should face in that direction. Later, however, it seemed desirable that temples should face to the east, the direction from which the earth is illuminated, so that limites too came to be set facing in this direction. Here, then, Hyginus Gromaticus attributes differences in orientation to changes in the preferred orientation, and he explicitly preserves the link between the orientation of limites and temples through the shift. The resemblances between centuriation and templa—or, perhaps more accurately, between centuriation and the practices around temples—work at a number of levels and mesh too closely to be dismissed easily. Some scholars, however, dismiss the claim of Frontinus and Hyginus Gromaticus, written long after centuriation first appeared, as antiquarian speculation, which, in these circumstances, means that they derive from a retrospective attempt to make sense of a practice by building on perceived resemblances to other practices. But links between centuriation and augury certainly were made in the first century , the age of Varro and of Sulpicius Rufus; they probably were being made in the time of the Gracchi, and, as we shall see, they possibly were made in the opening decade of the second century. But what is the nature and the significance of the resemblance? For some, the orthogonal axes signal that the land in question was under divine protection or under augural law.34 The augurs distinguished between Roman territory, the territory of allied states, and land that

33 Note that Frontinus (p. 14. 15–17 Th) has decumani face away from the rising sun, which is just the direction surveyors must have faced toward in order to set their groma: sic uti effectum est, ut decumani spectarent ex qua parte sol eo tempore, quo mensura acta est, oriebatur. 34 Gabba (1985: 267) suggests that the crossing axes were intended to signal that the arrangements founders made in their land distributions were under divine protection.

278

The Ritual of Centuriation

belonged to hostile communities, each with its own ritual requirements and legal forms. Building on the hypothesis that centuriae were restricted to settlements that would remain ager Romanus, now difficult to maintain, it has been suggested that the crossing axes of centuriation formed a visible sign that the tract was to remain within the augural category of ager Romanus—it was to form part of a citizen colony, not a Latin one.35 If this suggestion were to prove correct, and it is without explicit textual support, then one might suspect that the orthogonal axes of templa should have been, in some fashion, typical of all land that was legally Roman—which is clearly not the case. Now, at one level, surveying and augury could be said to resemble each other in a very basic manner since surveyors and augurs both sought to prepare spaces for further use by projecting defining and separating lines around and through them. For this reason, some scholars have seen the roots of the links between the two areas of activity as lying in perceived analogies between the ways that surveyors and augurs prepared their spaces.36 Over time, then, surveying would have come to resemble augural practices more closely in vocabulary and in form in order to make the parallels even clearer. Such a hypothesis has considerable plausibility, and, indeed, it is probably correct, but it goes only part of the way toward a solution. It is time to examine perceived similarities between the two processes at another level, one that focuses on the actions of the officials bearing ultimate responsibility.

MAGISTRATES, RITUAL, AND THE CREATION OF BOUNDARIES In the late third and second centuries, individuals from the same narrow group created templa for public purposes and directed the installation of limites. Correct performance of the rites required by the auspices formed an important element in determining the legitimacy of an official’s tenure when he first took up his office, and further performance of these rites legitimized their many acts while in office. Officials of high rank, moreover, usually were responsible for dedicating temples, and 35

See, for example, Behrends 1992: 192–280, esp. 212–42. Thus, Clavel Lévêque et al. (1996: 3–5 n. 2) suggests that surveyors possessed a laïcized version of the augurs’ knowledge, because both groups defined and categorized lands for public or sacred ends. O. Dilke (1988: 158–2) suggests that augury played a role in the training of republican surveyors. 36

The Ritual of Centuriation

279

thus also were associated with the definition of the templa in which they stood.37 The foundation of colonies was also a matter for prominent members of the governing elite. Colonies and viritane assignments each involved its own magistrates, specially elected for the occasion: triumvirs for colonies and decemvirs for viritane assignments. Between 218 and 167, when the evidence is best, over half of all triumviri were former consuls or praetors, while half of the decemviri had earlier reached these offices.38 Commissioners were directly involved in the preparation of land for settlement well into the first century . Livy (21.25.5) describes the founders of Placentia and Cremona as personally supervising the measuring of land, while he later (35.9.7–9) presents one commissioner at Copia (194) deciding on the scene the size of allotments only after all the land has been measured and divided into modules. And Varro supervised some land measurement while a vigintivir for settlements authorized by Caesar’s law of 59.39 One element of the relationship between these supervising magistrates and the preparation of land for assignments should be clarified at once. Scholars often assume the involvement of individuals at three markedly different social levels: the magistrates who had ultimate responsibility; the workers who actually used the groma and decempeda; and an intermediary group, with technical expertise, who supervised the work according to the rules of their science; these last would have been the guardians—and perhaps the creators—of the system and its associations. Under the Empire, surveyors formed a semi-autonomous group of experts, who addressed a range of problems and issues, some associated with limites and others with conflicts over private boundaries, over areas, and over the laws that applied to certain tracts of land. Under the Republic, however, the existence of this middle group, while often assumed, is highly unlikely.40 Determining the boundaries between public land and private, for example, was, under the Republic, the responsibility of the holders of offices that possessed the ius publicorum privatorum locorum—censors, consuls, and praetors, but also the duumviri who built or dedicated temples, the triumviri who founded colonies, and the decemviri who made viritane assignments.41 Thus, since those who actually used the surveyors’ instruments would always have possessed a relatively low status, too low to shape

37

For the dedicators of temples, see Tatum 1993: 319–28. See Gargola 1995: 58–67, 106–7. Varro, De Re Rustica 1. 2. 9–10; Pliny, Natural History 7. 176. 40 For republican surveyors, see Classen 1994: 161–70. On the lack of a specialized group of intermediaries in the second century , see Gargola, 1995: 90–5, 185–8. 41 Gargola 1995; 33–5. 38 39

280

The Ritual of Centuriation

the process and determine its associations, links between surveying and augural practices almost certainly would have formed around magistrates and the ways that they acted in matters of public importance. Polybius’ description (6. 41–2) of the establishment of a camp, written at the end of the interval in which centuriation first appeared, illustrates how elected officials may have organized their task. Like centuriation grids, Roman marching camps were laid out by constructing a network of roads meeting at right angles and forming a grid, although camps were not organized by orthogonal main axes nor were they divided into blocks that everywhere were of the same size. When an army neared the end of a day’s march, the commander selected a military tribune and centurions, and sent them ahead to the site. There the tribune and his staff first chose the place for the commander’s tent and the basic orientation of the camp. After this, they set out two parallel roads, one for the tribunes’ tents and the other marking the beginning of the area set aside for the legionaries’ tents, defining them with stakes and marking crucial points with flags of different colors. The remainder of the roads that formed the camp’s grid were determined from these points and lines. For Polybius, the chief advantage of the process was that it required only rather simple operations that were familiar to all. In Polybius’ account, individuals with only a minimal technical expertise make crucial decisions. Neither tribunes nor centurions, temporary occupants of positions whose responsibilities centered on the leadership of soldiers in legions and centuries, were experts at surveying, nor is there a sign of the presence of other experts.42 When setting up camps, the tribune made crucial decisions: deciding on the location of the camp; choosing the site of the commander’s tent and other important places associated with it; establishing the orientation of the gates; setting out, if only roughly, the placement of legionaries and auxiliaries. Having made these decisions, surveyors need only set their network of roads according to a predetermined pattern; although they would have need some skill with their instruments, there would have been little need for any supervisory body of experts. Now scholars studying Polybius’ account suggest that it derived from a military tribune’s commentarii of an early secondcentury date.43 If so, and it appears likely, then military tribunes at the 42 Classen (1994: 167) notes that there is no indication in Polybius’ account or in other accounts of centurions supervising camp layout that they possessed any technical training, but he still supposes that those who used the instruments possessed an expertise that went beyond manual skill. 43 Thus Fraccaro 1934: 154–61, esp. 158; Rawson 1971: 134–3, esp. 14–15; and Salvatore 1996: 5–18. The various difficulties with Polybius’ account largely concern the varying layouts of camps for two or four legions; they do not involve the manner of the camp’s construction.

The Ritual of Centuriation

281

time when centuriation first appeared would have had access to the necessary technical expertise, embodied in books of instruction, to direct surveyors in the field. In what ways, then, could the process of installing limites acquire augural associations? Colonial commissioners may have begun their limites like military tribunes, choosing the site for the groma and setting the general orientation. In republican Rome, ritual was closely associated with public ways of acting, especially by high-ranking officials. Polybius (6. 56. 6–15) and Dionysius of Halicarnassus (Roman Antiquities 2. 63. 2), two Greek writers separated by over a century, both found worthy of comment the vast number of rites and the degree to which they had penetrated Roman public and private life. One consequence of this pervasiveness of ritual in the Roman civic order is the difficulty in separating, in time and space, ritual from other forms of public activity that may seem more utilitarian in nature. As a concept, ritual has proven remarkably difficult to define. Scholars usually have associated the term with oftrepeated actions according to some formula. Some investigators, however, wish to focus less on ‘ritual’ than on ‘ritualization,’ a strategy that transforms some otherwise mundane actions into symbols of much greater power.44 Such an approach is useful in the present context. The process of constructing a centuriation grid brought together high-ranking magistrates and the working groups who would actually construct the limites, especially at the very beginning of the operation. Just this moment may have been deemed suitable for ritualization, as a mark of its importance. Templa existed at the center of an elaborate, and often-performed, body of ritual. Priests and officials created templa each time they intended to search for signs of divine favor. They established templa when they inaugurated spaces for temples and altars or for the performance of official acts. Creation of a templum involved a sequence of ritual gestures and formulaic phrases. The officiant first drew out with a gesture an east/west line and named the two parts he had just created ‘right’ and ‘left’. He next defined the north/south axis with another gesture and named the parts ‘this side’ and ‘beyond’. At this point, he called for a specific sign, if he was taking the auspices, or proclaimed the rules that were to govern the place. In the rite, then, the officiant defined by gesture his first line and named the parts it formed and then set out the second line, naming the sections that the new line had created, and, while doing these things, spoke the formulas of a prayer and of the legum dictio. As we will see, the dedicators of temples performed virtually the same sequence of acts.

44

See Bell 1992.

282

The Ritual of Centuriation

At this point, we should return to the defining moment in creating a camp—or a centuriation grid: the first placement of the groma, the result of which was not only the beginning of the process of constructing a camp or a grid, but also the setting of their locations, their orientations, and the lines of the major roads that would define them. For Polybius, this task took place under the personal direction of the military tribune to whom the army commander had assigned the task. Hyginus Gromaticus (p. 135. 1–6 Th) envisions this act as taking place in the presence of the colony’s founder or conditor, who had just taken the auspices, a sign of the event’s importance and an indication that everything that would follow depended ritually upon it: ‘When the groma had been positioned after taking the auspices, perhaps in the very presence of the conditor, they sighted accurately the next sunrise, and sent out limites in each direction’ (posita auspicaliter groma, ipso forte conditore praesente, proximum uero ortum conprehenderunt, et in utramque partem limites emiserunt).45 It is just at this moment, I would suggest, that the performance by the conditor of a rite analogous to the ritual creation of a templum may well have seemed an appropriate way of emphasizing the importance of the official and of the defining act of centuriation. When surveyors established their first limes in the presence of the conditor, they would have created a line that, from the perspective of the nearby founder, would have had a clear left and right, and any comparison between the founder’s act and the actions of the maker of a templum or of the dedicator of a temple would have been especially appropriate at this time. Indeed, it is possible that surveyors added the kardo maximus, with the decumanus maximus one of the defining features of centuriation, in order to make this analogy between the actions of the two sets of officials at just this moment even closer. After this rite, those who actually used the groma and measuring rods, men far lower in status, would have operated according to pattern, perhaps without much official supervision until a decision was needed on important points. There is a possible objection to such a view. At the beginning of centuriation, the conditor would have stood to one side of the decussis, the point where the axes crossed, which would have been occupied by the groma and by those who manipulated it. From this perspective, he or his

45 Instead of regarding Hyginus Gromaticus’ auspicaliter as a sign of the importance of the actor and of the act, Dilke. (1992: 337–47, esp. 337–8), regards taking the auspices as an indication of the importance of the groma itself, declaring that Hyginus ‘takes it for granted that the groma . . . was regarded with some religious awe and was in use from very early times.’ There is no evidence for this transfer of the significance of the auspices from a magistrate’s actions to an object used in the actions.

The Ritual of Centuriation

283

attendants would have defined right, left, near, and far. When creating a templum for auspication, however, the augur clearly stood no closer than the margins of the field that he intended to define, so that the entire templum would lie before him in his field of view. Yet it is possible that augurs, on occasion, did take up their position at or near the decussis. Cicero, himself an augur, depicts Attus Navius as defining his field of vision from the middle of the vineyard, so that ‘behind’ would actually be behind the auspicant, while Festus names the parts of a temple in relation to a person occupying the decussis.46 Boundaries of a more secular nature sometimes were defined in a ritual fashion similar to the rites of augurs and the dedicators of temples. Plautus may have preserved traces of just such a rite in the prologue to his Poenulus, first performed at some uncertain date before his death in 184, perhaps in the decade of the 190s.47 On the surface, the prologue falls into two broad parts. In the first, the prologus parodies a number of official roles, while in the second, and longer, part he sets out the argumentum of the play.48 First, the prologus identified himself as a commander (imperator), which he immediately qualified as being of actors not of soldiers. Then, in his role as imperator histricus, he gave orders to his praeco or herald and then issued an edict—edicta is the word used—to the curatores ludorum, the lictors, and to the audience telling them where to sit and how to behave, instructing his listeners as if he were a magistrate addressing an assembly. Scholars have detected traces of official language in his announcements, and the words of the prologus’ edict do parallel fairly closely the forms of a magisterial edict.49 Magistrates addressing assemblies would have been a common occurrence in Rome; the audience, familiar with the forms, would have recognized the parody. The lines that form our primary concern make up the transition from the prologus as imperator to the prologus as narrator of the plot of the play. Immediately after completing his edict and reminding the audience of his imperium, the prologus announced (lines 46–9): ‘Now I wish to revert again to the argumentum, so that you might be as well-informed as I am. I shall now define its regiones, limites, and confinia; in this 46 Cicero, De divinatione 1.31 (quoted above, n. 18); Festus (Paulus) pp. 244–5 L (quoted above, n. 25). 47 Jocelyn (1969: 96–123) suggested that the prologue, as we have it, is the work of several authors, while Zwierlein, (1990: 206–24) defends Plautus’ authorship. In any case, for our purposes, it is necessary only to note that Jocelyn holds that the text as we have it was fixed in the first third of the second century. On the issue of ‘contamination’ in the Poenulus in general, see Fraenkel 1960: 253–67. 48 For the nature of the prologue, see Jocelyn 1969: 106–11; Slater 1992: 131–46. 49 On his language, see Jocelyn 1969: 106–11.

284

The Ritual of Centuriation

matter, I have been made finitor’ (ad argumentum nunc vicissatim volo remigrare, ut aeque mecum sitis gnarures. eius nunc regiones, limites, confinia determinabo: ei rei ego finitor factus sum). In these lines, the actor adopts yet another office role; from our current perspective, the nature of the underlying action and the office of the official who performed it are our central concerns. First, the office. Under the Empire, the word finitor was a substitute for mensor or surveyor, a rather low-ranking functionary: according to Nonius (pp. 11–12 L), ‘those who we now call agrimensores were called finitores because they separated fines’ (finitores dicebantur, quos nunc agrimensores dicimus; dicti quod fines diuiderent). Now, the words agrimensor and mensor were used under the Empire to describe a range of technical experts, and Nonius seemingly claims that finitor was used in the same way under the Republic, although there are no traces of such a body of experts at the time.50 In republican literature, the word finitor, in the sense of a person who establishes boundaries, occurs only in Plautus’ passage and in one oration of Cicero’s. According to Cicero (De lege agraria 2. 32, 34), Rullus’ proposed agrarian law of 63  would permit its decemviri to appoint finitores to make a number of decisions about the location of boundaries in a manner that imperial surveyors later would make. But Rullus’ surveyors are not experts: Cicero claims that they were to be young men chosen ex equestri loco, which, at the time, almost certainly meant that they were sons or grandsons of Romans holding the public horse, hardly a suitable status for technical experts.51 Cicero, moreover, presents their position as a highly undesirable innovation. Plautus’ finitor almost certainly occupies a higher status than did Rullus’ young equites: after all, the prologus first took the role of an imperator and then that of a finitor without any sign that the prologus was now adopting a role markedly lower in status than the one he had taken up just a few lines earlier.52 At core, finitor denotes a function and not an office, in much the same way as imperator or conditor do. For the uses of finitor, see Nicolet 1970: 72–103; Gargola 1995: 179–88. Nicolet (1970: 77–83) argues that Rullus wished to have young men of equestrian status act as finitores because his surveyors would possess judicial powers that they would exercise under the direction of an agrarian commissioner but at some distance from him; the provision, then, would represent an extension of the judicial powers that members of the equestrian order had received since the tribunate of C. Gracchus and a step on the way to the mensores of the Empire. 52 Note that Plautus’ prologus also mentions curatores ludorum, lictores, a praeco, and the iuratores who took citizens’ declarations in the census, all but the first are nonmagistrates. But the prologus does not take these roles for himself—he only assumes the positions of an imperator and of a finitor—and he assigns these lesser positions to the audience. 50 51

The Ritual of Centuriation

285

Consuls and praetors, holders of imperium, must have frequently issued edicts to the assembled soldiers or citizens, but what action of the finitor, beyond the mere establishment of boundaries, was being parodied? Just as was the case with the imperator, here too the prologus probably followed some official act that was public, formal, and perhaps even formulaic, and often-performed, for otherwise the audience would have failed to recognize the parody. In the prologue, the prologus as finitor announced his intention to set the lines that defined the play’s argumentum, and it is possible that the various words for boundaries denote the argumentum itself, the defining limits of the coming play. Slightly later (line 60), however, the prologus gives the argumentum a clear locus or place, the stage at which the play is to be performed (locus argumentost suom sibi proscaenium). When acting as an imperator or as a finitor, the actor certainly accompanied his words with gestures, which also may have been official in their inspiration. Here, gestures would have been most appropriate in the definition of a physical space, so that the finitor’s regiones, limites, and confinia would have defined the stage, which would limit in turn the argumentum and then the play. Now, Plautus’ imperator parodied real words, and probably real actions, of actual imperatores, and it is likely that his finitor did too. First, the prologus declared that he had been authorized to act as finitor in the matter: ei rei ego finitor factus sum. A number of public boundary markers, dating to and after the late second century, record the office and the authority under which the magistrate—under the Republic, it is always the holder of some magistracy—set the boundaries.53 It is conceivable that the official, in the act in which he determined the position of these boundaries, also proclaimed his authority in some manner. And with the words determinare regiones, we are again within the sphere of official language: their primary meaning is the formal definition of a space, and the words are often used to describe an augur’s setting of the defining lines of his templum by word and by gesture.54 But there are closer parallels to Plautus’ passage. His finitor proclaimed in the future tense his intention to mark out his boundaries: eius nunc regiones, limites, confinia determinabo. When dedicating a temple or an altar, the dedicator defined his space verbally and with gestures and then set

53 See, for example, ILS 9376: C. Caninius C. f. pr. urb. de sen. sent. poplic. ioudic. (late second or early first century); 24: C. S[e]mpronius Ti. f. Grac., Ap. Claudius C. f. Pole., P. Licinius P. f. Crass. IIIvir. a. i. a. (132 ); for a slightly later example, see ILS 5922a: M. Valerius M. f. M’. n. Messal., P. Serveilius C. f. Isauricus ces. ex s. c. termin. 54 For the use of determinare regiones to describe the definition of a templum, see Valeton 1892: 370–3.

286

The Ritual of Centuriation

out the laws that would govern it in a legum dictio. The opening part of the verbal formula, preserved in a number of inscriptions partially replicates the words of Plautus’ finitor, for both announce their intention, in the future tense, to perform a certain action: ‘When I will give and dedicate this altar to you today, I will give and dedicate it by these laws and by these regiones, which I will say openly here today.’55 Now it is true that Plautus used the verb determinare, while dedicators used the verbs dare and dedicare, but this need not be a fatal difference: his finitor was not announcing a gift or a dedication, but merely the intent to define a space and was parodying an official acting in the same manner. One can even go a step farther. When an augur was defining his templum or when a dedicator was consecrating his temple, both pronounced the rules that were to govern the use of the space that they had just created. Indeed, in the case of temples, we know from later inscriptions that the laws that governed temples, which presumably were included in the dedicator’s legum dictio, could be rather lengthy. In any case, in the texts of the inscriptions, the provisions of the law are written down immediately after the recording of the formalities of the ritual of consecration. Interpreters of the prologue have often viewed it as divided into two broad sections, the parody of official actions by imperator and by finitor and the recounting of the play’s argumentum. Yet one could regard the prologue as divided differently, with the first section centered around the prologus issuing instructions to the audience in the role of imperator and the second section beginning with the prologus as finitor defining formally a space and setting out in the argumentum the rules that would govern its use. If Plautus’ prologus was imitating the creation of boundaries in a way that exhibited broad similarities with certain official and augural practices, the question still remains as to the nature of the boundaries and the space that underlies the parody. Since the prologus was probably adopting a magisterial role, the boundary in question should have been one that a high-ranking magistrate would have established early in the second century. In the second century and into the first, censors, consuls, praetors, and special commissioners charged with establishing temples and colonies or making viritane assignments defined boundaries between public land and private or carved out a section of public property for some special use; private citizens appointed as recuperatores resolved disputes over boundaries between private citizens. Plautus certainly 55 Thus ILS 112: quando tibi hodie hanc aram dabo dedicabo, his legibus hisque regionibus dabo dedicaboque, quas hic hodie palam dixero. For other examples, see ILS 4907; CIL 12.756; 3.1933; 10. 3513.

The Ritual of Centuriation

287

would not have parodied the dedication of a temple or the creation of a templum, for such an act would have been inappropriate. The underlying action, then, should have been a matter of public law, rather than of sacred or augural law. To be recognizable, however, the rite should have been performed either very frequently or in a prominent manner recently. It is possible that Plautus had no particular type of boundary in mind: in other words, the process of formally naming and pointing out boundaries while giving a legum dictio was used to define and create formally a range of different spaces with sufficient overall frequency. Yet the period from the end of the Second Punic War to 184, the years in which the play was probably performed, saw the foundation of a substantial number of colonies, and some scholars have assumed that Plautus’ model here was a colonial commissioner and that the words regiones, limites, and confinia all denote the dividing lines of centuriation.56 This hypothesis is possible, and perhaps likely, but it is far from certain. At the very least, however, the passage shows that Romans of a time near to the development of centuriation would have found it plausible and possible to define secular spaces in rites with parallels in augural practices and in the dedication of temples. If this hypothesis is correct—and the similarity in actions and in vocabulary suggest that it is—then centuriation represents a ritualized elaboration of one of a colonial commissioner’s defining roles, forming another of the many rites clustered around high-ranking officials and their most important actions. As the rank of the founding commissioners shows, Romans of the third and second centuries attached great importance to the founding of colonies. Virtually every major element in the founding possessed its own defining rites: founders set the urban core by plowing a furrow around it in just the way Romulus was believed to have established the Roman pomerium; they formed the new citizenry through rites that resembled the Roman census, which accomplished at Rome the same end; they created sacred spaces in much the same way that officials did at Rome.57 Those elements of surveying that were most closely connected to the presence of a founder and the proclamation of his decisions over location and general orientation too would have come to be elaborated in a rite. The forms of this rite then shaped and organized the ways surveyors went about their tasks long after the performance of the opening ceremony; in other words, the initial ritual act would have given structure to the operation that followed it, while the axes of the network, enshrined in large roads, provided a visible reminder of the founding act.

56

See, for example, Jocelyn 1969: 110.

57

See Gargola 1995: 71–101.

288

The Ritual of Centuriation

Which would have come first, centuriation or the rite? If surveyors and augurs were seen as performing analogous operations when they projected lines through spaces, then this analogy would have been most visible at the point in the creation of a field system when the first line or lines were set. In this case, the analogy could be made closer by adding a crossing axis, the kardo maximus, for imitation of augury is the most plausible reason for its appearance. Indeed, it is possible that a rite, in some form, antedated centuriation itself, or at least the appearance of the orthogonal axes that were one of its defining features. There are some signs that the beginning of a project was regarded as an act of more than utilitarian significance before the Romans developed centuriation. At the colony of Cosa (273 ), where the fields were organized by scamna, the surveyors appear to have first placed their groma just outside one of the town’s gates.58 Officials would not have chosen this location, on top of a high hill with no arable land in the immediate vicinity, for strictly utilitarian reasons, so that the choice may well indicate that the initial act of a survey was deemed of sufficient importance to warrant performing it in close proximity to an important location in the future city (see below).

FRONTINUS AND HYGINUS GROMATICUS ON LIMITES Returning to the texts that began this investigation, there are clear signs that Frontinus and Hyginus Gromaticus did have rites in mind when they began their descriptions of centuriation, that they used these rites to give structure to their works, and that concerns about the rite, its proper location, and its proper orientation permeate much of their texts. The connections that they made between surveying and rites, in other words, are central to their works, not marginal or casual. Indeed, Hyginus Gromaticus began by calling centuriation ‘the most distinguished among all the rites and practices of measurement’ (inter omnes mensurarum ritus siue actus eminentissima).59 58

Thus Castagnoli 1956: 147–65; Hinrichs 1974: 29–30, 48, 82. The word ritus has the primary meaning of rite or formal religious observance, but it can also denote some customary way of acting, and both fit here. Thus, Campbell (2000: 135) translates Hyginus Gromaticus’ ritus siue actus as ‘observances and practices’, while Clavel Lévêque et al. translate it as ‘rituels ou operations’—their rituals, however, are centuriation itself, which they view as a ritualized (i.e. formalized) way of giving a religious character to the land that was being divided. Hyginus Gromaticus, however, moves immediately from his characterization of centuriation as a ritus siue actus to describing an actual rite, making ‘rite’ the preferred translation. 59

The Ritual of Centuriation

289

As we have seen, Frontinus and Hyginus Gromaticus both began with a description of the way that the Etruscan haruspices divided the world, and then they proclaimed that the Roman maiores or antiqui adapted these practices to surveying. These descriptions can be read in two ways. The more usual view is to take them as straightforward descriptions of Etruscan cosmology and its associated organization of spaces.60 But they are also descriptions of a series of words and actions that together constitute a rite. Thus, the haruspices divided (diuiserunt) the world with an east/west line and called (appellauerunt) the north ‘right’ and south ‘left’. Then, they divided or drew (diuiserunt or duxerunt) another line from north to south and named (nominauerunt) the far side antica and the near side postica. Finding the pattern for centuriation in an Etruscan rite, both authors went on to describe its adaption to Roman circumstances with a Roman version of the Etruscan model. For Frontinus, the maiores drew (duxerunt) first their east/west limes and then their north/ south one. Each line divided (divisit) the field into a dextra and sinistra or a citra and ultra. Frontinus, moreover, gave each line not only an orientation but a direction, perhaps as if it was drawn by a gesture: his haruspices go first from east to west and then from north to south, while his Romans set their lines from east to west and then from south to north. Hyginus Gromaticus’ haruspices too go from east to west and then from south to north, while his Roman lines are ‘directed’ to lead in the same way. At the ends of their renditions of the Roman adaptations, both present their decumanus and kardo as actors, performing actions that form part of the ritual division of spaces: ‘naming’, ‘directing’, and ‘dividing’.61 Several passages that focus on limites or their creation illustrate the continued importance of the ritual elements present in the opening act or of the vocabulary associated with it. In their descriptions of lesser limites, following the rites of haruspices and of surveyors in their texts, Frontinus (pp. 11. 15–13. 12 Th) and Hyginus Gromaticus (pp. 132. 13–135. 14 Th) first set out etymologies of decumanus and kardo and note deviations from the proper orientation. Lesser limites follow. Both describe them in a way that emphasizes the main axes and their augural associations. Frontinus, for example, does not call these roads decumani or kardines, preserving these terms at this point for the main axes. For him, limites, which ‘looked to’ (spectabant) the east, are called

60

Thus Hübner 1992: 140–70. For the use of dirigere for the act of marking out the lines of a templum, see Cicero, De divinatione 1. 30: nempe eo Romulus regiones direxit tum cum urbem condidit; for the use of dividere in the same context, see Cicero, De divinatione 1. 31 (quoted above, n. 18). 61

290

The Ritual of Centuriation

‘straight ahead’ (prorsi), while those running south are called ‘transverse’ (transversi); Festus (264L) confirms that limites directed to the east were called prorsi. Hyginus Gromaticus too stresses the proper orientation of lesser limites and noted that ‘looking to the east’ should have the same sense for limites that it possessed in the orientation of temples. Hyginus Gromaticus (pp. 140. 16–146. 8 Th) provides the clearest evidence for the continuing importance that he gave to the act of placing the groma for the first time. He does not frame his discussion on choosing land for division in a strictly utilitarian fashion. Instead, he structures his examination of the principles guiding the choice of land in terms of where to place the decumanus maximus and kardo maximus. Stated in this fashion, Hyginus Gromaticus’ discussion necessarily involves the choice of the location where surveyors, founders, and their attendants were to gather for the first placement of the groma, the taking of the auspices, the performance of any other opening rites, and the formal identification of the axes that would guide the construction of the grid. Ease of construction is sometimes presented as a reason for deviating from proper practices, but it is not presented as a primary concern. To express a desire that assigned lands be close to the city, he recommends that the point of origin of the two main limites actually be in the city, if possible, or close to its walls if not. Then he notes that in some later colonies the kardo maximus and the decumanus maximus begin in the town and leave it through the four gates; the earliest known instance of this practice is at the Roman colony of Carthage, either Gaius Gracchus’ or Julius Caesar’s.62 He next compares this pattern, which he calls the most attractive, with the camps, where, he says, the groma is erected at a crossroads where men can assemble, as if in a forum. The belief that the first placement of the groma was more than a merely utilitarian act was not peculiar to Hyginus Gromaticus. In measuring out a camp, the surveyors first set out the area for the praetorium, then placed the groma at the praetorium’s entrance and used it to mark out a road—the only one that ran completely through the camp—which was to be the site for the tribunes’ tents. Now, Pseudo-Hyginus called the site for the groma the locus gromae, and placed near it the commander’s tribunal, an altar, and an auguratorium for taking the auspices; Polybius calls the same place the gramma, showing that the same term was current in the mid-second century.63 In the colony at Cosa (273), before the 62

For Carthage, see Wightman 1980: 29–46. For the locus gromae, see Hyginus 12; for the gramma, Polybius 6. 41. 6. PseudoHyginus’ text survives as an appendage to Hyginus Gromaticus, but it is clearly by a different author and is usually dated to around  100; see Lenoir 1979: vii–viii. 63

The Ritual of Centuriation

291

development of regular centuriation, the surveyors placed their groma to lay out their first decumanus just outside one of the town’s gates.64 Several interpretations of the significance of this concern that surveyors first erect their groma at an appropriate place are possible. In the camps, the locus gromae becomes a place of authority, so that the focus of the process of creating the camp becomes in turn the focus of the camp itself. The recommendation that the founders of colonies begin to survey from the new city’s forum or from one of its gates may well mark a similar effort to link the technical operation to the city itself and to its central institutions.65 The same places, for the same reasons, would also be suitable for the performance of significant public acts, especially if they were encompassed in a dramatic and highly visible rite.

FRONTINUS, HYGINUS GROMATICUS, AND ANTIQUARIANISM In this essay, we have examined the links between centuriation and augury at two levels. In the first, which concentrates on ritual acts performed by members of the Roman elite at specific times and places, we have argued that augural rites served as a pattern for the ritualization of the beginnings of a land-division project. The founding of colonies was a highly formalized process, and commissioners performed a range of rites marking its essential stages. Centuriation, too, was assimilated into the ritual order of colonial foundation and of the Roman state, in a way that emphasized the actions of high-ranking magistrates and their links with augury, essential to the legitimization of all public acts in Rome. The further elaboration of this rite would have set the pattern for field-division, and probably would have continued to influence the development of its forms long after the first links between surveying and augury had been made. At a second level, the rite, and the organization of spaces that resulted from it, provided a way for some technical writers later to describe the process of installing limites and the rules that governed it, in which they emphasized the rite, its proper placement, and its orientation—all nonutilitarian aspects of the process—and used the rite, and the terms set

64

Castagnoli 1956: 160; Hinrichs 1974. Note that Hyginus Gromaticus (p. 144.16–17 Th) says that the groma is placed at the center of the camp, where men can assemble as in a forum. 65

292

The Ritual of Centuriation

forth in its description, as bases to organize their discussions of related matters. These conclusions lead to further questions. In what other areas of Roman public life, if any, can a similar relationship between rite and public actions be found? In what other texts? From the late second century  and well into the Empire, Roman authors produced a range of texts, now existing only in fragments, that described rites, their origins, and the rules that guided their proper performance.66 Modern scholars have often characterized such works as antiquarian, a term that can carry the connotation of nostalgia for the past, often the very distant past, and a lack of concern for present-day matters. Indeed, the opening passages of Frontinus and Hyginus Gromaticus, with their descriptions of rites and their attributions of these rites to an indefinite but seemingly distant past, have been called antiquarian by those who wish to deny any connection between augury and surveying.67 Not too long ago, Jerzy Linderski suggested that Roman works on public and sacred law— essential to the Romans’ view of the proper functioning of their state— may have been antiquarian in nature.68 This investigation has set out some of the ways that this link between antiquarianism and administration may have operated, not only in the description of processes but in their formation.

B I B L I OG R A P H Y Beard, M., J. A. North, and S. R. F. Price (1998). Religions of Rome. Cambridge. Behrends, O. (1992). ‘Bodenhoheit und privates Bodeneigentum im Grenzwesen Roms’, in O. Behrends and L. Capogrossi Colognesi (eds.), Die römische Feldmeßkunst: Interdisziplinäre Beiträge zu ihrer Bedeutung für die Zivilisationsgeschichte Roms. Göttingen, 192–280, esp. 212–42. Behrends, O., M. Clavel Lévêque, D. Conso, et al. (1998). Frontin: L’oeuvre gromatique. Corpus Agrimensorum Romanorum 4: Iulius Frontinus. Luxembourg. Bell, C. (1992). Ritual Theory, Ritual Practice. New York. Blume, F., K. Lachmann, and A. Rudorff, eds. (1848–52). Die Schriften der römischen Feldmesser. 2 vols. Berlin. (Lachmann’s edition of the text is in vol. 1.) Campbell, B. (2000). The Writings of the Roman Land Surveyors: Introduction, Text, Translation and Commentary. London. Castagnoli, F. (1956). ‘La centuriazione di Cosa’. MAAR 24: 147–65, esp. 160. 66 68

See Rawson 1985: 233–49. Linderski 1990: 42–53, esp. 50.

67

See n. 7 above.

The Ritual of Centuriation

293

Castagnoli, F. (1971). Orthogonal Town Planning in Antiquity (trans. V. Caliandro). Cambridge, MA. Catalano, P. (1960). Contributi allo studio del diritto augurale. Turin. Catalano, P. (1978). ‘Aspetti spaziali del sistema giuridico-religioso romano: Mundus, templum, urbs, ager, Latium, Italia’. ANRW II.16.1: 440–553, esp. 467–79. Chouquer, G. and F. Favory (1991). Les paysages de l’antiquité: Terres et cadastres de l’occident romain, IVe s. avant J.-C./IIIe s. après J.-C. Paris. Classen, C. J. (1994). ‘On the training of the agrimensores in Republican Rome and related problems: some preliminary observations’. ICS 19: 161–70. Clavel Lévêque, M., D. Conso, A. Gonzales, J.-Y. Guillaumin, and P. Robin (1996). Hygin l’arpenteur: L’établissement des limites. Corpus Agrimensorum Romanorum 4. Naples. Dilke, O. A. W. (1988). ‘Religious mystique and the training of Agrimensores’, in D. Porte and J.-P. Néraudau (eds.), Homages à Henri Le Bonniec: Res sacrae. Brussels, 158–62. Dilke, O. A. W. (1992). ‘Insights in the Corpus Agrimensorum into surveying methods and mapping’, in Behrends and Capogrossi Colognesi, 1992, 337–47, esp. 337–8. Dumézil, G. (1970). Archaic Roman Religion (trans. P. Krapp). Chicago. Fraccaro, P. (1934). ‘Polibio e l’accampamento romano’. Athenaeum 12: 154–61, esp. 158. Fraenkel, E. (1960). Elementi Plautini in Plauto. Florence. Gabba, E. (1978). ‘Per la tradizione dell’heredium romuleo’. RIL 112: 253–8. Gabba, E. (1985). ‘Per un’interpretazione storica della centuriazione romana’. Athenaeum 63, 265–84. Gargola, D. J. (1995). Lands, Laws, and Gods: Magistrates and Ceremony in the Regulation of Public Lands in Republican Rome. Chapel Hill—London. Hinrichs, F. T. (1974). Die Geschichte der gromatischen Institutionen: Untersuchungen zu Landverteilung, Landvermessung, Bodenverwaltung und Bodenrecht im römischen Recht. Wiesbaden. Hübner, W. (1992). ‘Himmel und Erdvermessung’, in Behrends and Capogrossi Colognesi, 1992, 140–70. Jocelyn, H. D. (1969). ‘Imperator histricus’. YCIS 21: 96–123. Lenoir, M. (1979). Pseudo-Hygin: Des fortifications du camp. Paris. Lewis, M. J. T. (2001). Surveying Instruments of Greece and Rome. Cambridge. Linderski, J. (1986). ‘The augural law’. ANRW II. 16. 3: 2146–312. Linderski, J. (1990). ‘Mommsen and Syme: law and power in the Principate of Augustus’, in K. Raaflaub and M. Toher (eds.), Between Republic and Empire: Interpretations of Augustus and His Principate. Berkeley, 42–53, esp. 50. Nicolet, C. (1970). ‘Les finitores ex equestri loco de la loi Servilia de 63 av. J.-C.’ Latomus 29: 72–103. Norden, E. (1939). Aus altrömischen Priesterbüchern. Lund. Prosdocimi, A. L. (1991). ‘La “stella” del templum augurale e la ‘stella’ dei gromatici: una stella augurale da Alba Fucens’. PP 46: 37–43.

294

The Ritual of Centuriation

Rawson, E. (1971). ‘The literary sources for the pre-Marian army’. PBSR 39: 13–31, esp. 14–15. Rawson, E. (1985). Intellectual Life in the Late Roman Republic. Baltimore. Rudorff, A. (1992). ‘Gromatische Institutionen’, in Behrends and Capogrossi Colognesi, 1992, 2: 227–464, esp. 296–7, 419. Salvatore, J. P. (1996). Roman Republican Castrametation: A Reappraisal of Historical and Archaeological Sources. Oxford. Scheid, J. (1985). Religion et piété à Rome. Paris. Schubert, C. (1996). Land und Raum in der römischen Republik: Die Kunst des Teilens. Darmstadt. Slater, N. (1992). ‘Plautine negotiations: the Poenulus prologue unpacked’. YCIS 29: 131–46. Tatum, W. J. (1993). ‘The Lex Papiria de dedicationibus’. CP 88: 319–28. Thulin, K. (1913). Corpus Agrimensorum Romanorum. Leipzig. Torelli, M. (1966). ‘Un templum augurale d’età repubblicana a Bantia’. RAL 21: 293–315. Torelli, M. (1969). ‘Contributi al supplemento del CIL IX’. RAL 24: 9–48. Vaahtera, J. (2001). Roman Augural Lore in Greek Historiography. Stuttgart. Valeton, I. M. J. (1892). ‘De templis Romanis’. Mnemosyne 20: 370–3. Weinstock, S. (1946). ‘Martianus Capella and the cosmic system of the Etruscans’. JRS 36: 101–29, esp. 129. Wightman, E. M. (1980). ‘The plan of Roman Carthage: practicalities and politics’, in J. G. Pedley (ed.), New Light on Ancient Carthage. Ann Arbor, 29–46. Zwierlein, O. (1990). Zur Kritik und Exegese des Plautus. Vol. I: Poenulus und Curculio. Stuttgart.

11 The Lucus Pisaurensis and the Romanization of the Ager Gallicus Filippo Coarelli

The group of inscribed stones excavated in the seventeenth century by Annibale degli Abati Olivieri Giordani, in the district of Santa Veneranda (Fig. 11.1) a mile south-west of Pesaro and now preserved in the Museo Oliveriano,1 undoubtedly constitutes the most important complex of sacred dedications to have come down to us from the Middle Republic. In making this claim, I am obviously using the higher dating of the dedications and of the sanctuary to which they belong; this is in conflict with an established tradition of study stretching from Bormann to Degrassi2 and finally to Peruzzi,3 which attributes to them a date later than the foundation of the colonia civium Romanorum of Pesaro, which took place in 184 ;4 but my date is in agreement with another tradition, starting from Mommsen, who, in the first edition of CIL I (1863),5 expresses himself in the following terms: 1 CIL I2 368–81 = XI 6271; 6273–6; 6278; 6279; 6283; 6290; 6292; 6297; 6300–2 = ILS 2970–83 = ILLRP 13–26. Cresci Marrone and Mennella 1984: 89–150, nn. 1–14. 2 E. Bormann CIL XI. p. 942; A. Degrassi, ILLRP I p. 47. Bormann 1888–1926: 942; Degrassi 1957: 47. 3 4 Peruzzi 1990. Livy 39. 44. 10; Velleius Paterculus 1.15. 2. 5 Mommsen 1863: pp. 32ff. (but see already Mommsen 1850: 29). Dedicantium nomina reperiuntur virorum duo . . . mulierum quattuor . . . quarum nullam praenomine carere inter iudicia est remotissimae vetustatis. Idem de scriptura valet . . . ne hos quidem titulos quominus quinto saeculo tribuamur impedit absolute. Lingua denique meram vetustatem spirat. . . . Praeterea haec reliqua etiam omnia quae in sermone horum titulorum singularia inveniuntur vetustissimam aetatem clare arguunt, id est aetatem bello Hannibalico vel anteriorem vel certe supparem. Verum est anno demum u. c. 570 coloniam civium Romanorum deductam esse Pisaurum; neque tamen quicquam impedit—cum Senones exacti reperiantur anno 471 condita sit Sena, circa idem tempus pauloque post anno 486 Ariminum—quominus Pisaurum

Filippo Coarelli, The Lucus Pisaurensis and the Romanization of the Ager Gallicus In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0012

296

The Lucus Pisaurensis and the Romanization

Among the dedicators are found the names of two men . . . of four women . . . none of whom is missing a praenomen, which is one of the indications of a very early date. The same holds for the writing-style . . . which absolutely forbids us from dating these graffiti to the fifth century a.u.c. The language, finally, breathes antiquity. . . . Moreover, all the elements found to be singular in the language of these inscriptions clearly argue in favour of the earliest date, that is either earlier than or certainly contemporary with the Hannibalic War. But it was in 570 a.u.c. (186 ) that the colony of Roman citizens was sent out as Pisaurum. There is no reason not to decide that, since Sena . . . was founded in 471 a.u.c. (283 ), and Ariminum, around the same time and a little later in 486 (268 ), there would have been, for a whole century before Pisaurum was converted into a colony, a conciliabulum of Roman citizens occupying the location. But history does not allow the redating of these inscriptions to earlier than the end of the fifth century (i.e. not earlier than 254 ).

Lommatzsch and Bücheler expressed the same opinion in the second edition of CIL I (1918), as did Ritschl before them.6 But already in 1901 E. Bormann in Volume XI of the CIL had taken an opposite position:7 ‘The grove, in which bases 6290–6302 were located, since it belonged to the Matrons of Pisaurum would have been set up, I believe, at the time when the Roman colony of Pisaurum itself was deducta (led out), or a little after’. Subsequently, this opinion would be Degrassi’s view8 as well and, more recently that of Peruzzi,9 according to which it would have been inconceivable for an ordo matronarum to have existed in a conciliabulum before there was a colony, thus imposing a date that would necessarily have to be later than 184 . Moreover, the connection of the dedications with the prodigy that occurred in Pisaurum in 163 10—a wholly arbitrary connection in my opinion—would imply the progressive lowering of the chronology to that date.11 For my part, I cannot see what the incompatibilities are that make the presence of an organization of matrons in a conciliabulum inconceivable; after all, Mommsen could not see it either (and I do seem to be in the best company!). In addition, the fact that we are dealing with matronae

antequam in coloniae formam redigeretur per integrum saeculum conciliabuli civium Romanorum locum obtinuisse statuamus. Ante saeculum tamen quintum extremum titulos hos reicere rerum memoria non permittit. 6

7 Ritschl 1859: 400–2. Cf. n. 2. 9 Cf. n. 2. Peruzzi 1990: 29ff. 10 Obsequens 14: Nocte species solis Pisauri adfulsit. (In the night, a kind of sun shone out at Pisaurum). 11 Peruzzi 1990: 35. 8

The Lucus Pisaurensis and the Romanization

297

Fig. 11.1 The area S. of Pesaro (block IGM, F° 109 1 SE Pesaro). Within the circle is the approximate position of the lucus Pisaurensis.

Pisaurenses does not constitute an argument in favour of a date later than the foundation of the colonia given that the occupation of Pisaurum was already happening in the archaic age, as the recent excavations prove,12 and probably had the same name derived (by hydronymy) from the river that flowed in the vicinity.13 Additionally, there are elements decisively in favour of the higher chronology: in the first place, the palaeographic and linguistic considerations already put forward by Mommsen. As regards the second aspect, I do not have the competence to intervene. All the same, it seems to me that to claim that Pesaro, as organized by the Sabines, was ‘the colonial area of a colonial area’14 makes it difficult to explain the use in written 12

Luni 1984: 109–80; Braccesi 1984: 1–38, esp. 4–7. The name of Pesaro and of its river—of which the earliest accepted version is Isaurum—appears on the famous stele of Novilara in the form Isairon. Durante 1962: 66; Durante 1978: 393–400, esp. 397f. 14 Peruzzi 1990: 11. 13

298

The Lucus Pisaurensis and the Romanization

documents directed to a high-level audience (the local aristocracy) of out-of-date Latin—out-of-date by a century. All the more so, since there is no trace of any such obsolete usages in epigraphic documents dating to the second century from the Sabine area.15 I would like to insist instead on the palaeographic aspect of these documents (Figs. 11.2–11.3) an argument singularly undervalued by the more recent commentators (but not by Mommsen!) and simply disregarded by Peruzzi. Certain characteristics, from this point of view, stand out as still decidedly archaic: in general, the absence of ‘quadratura’ (squaring),

Fig. 11.2 Facsimile of the inscriptions of the lucus, taken from CIL I2 368–75.

15

See, for example, the inscription from Flamignano, Filippi 1984: 165–77 esp. 174f.

The Lucus Pisaurensis and the Romanization

299

Fig. 11.3 Facsimile of the inscriptions of the lucus, taken from CIL I2 376–81.

the bending of certain letters (typically the N), the total lack of apicature (apicature). More specifically, the presence in certain instances of an A with its cross-stroke disarticulated, and an O with either its bottom or top left open and also smaller than other letters; of a D open below; of an open P, occasionally composed of three lines; but above all, of a hooked L, of which there is no later attestation than the first years of the second century .16 In addition, the presence of the third declension dative in -e or in -ei and, above all, of the nominative or accusative of the second declension in -o lead one to exclude any date later than the final decades of the third century .17 Moreover, there is no shortage of variants

16

e Coarelli 1976: 157–79, esp. 160, n. 7.

17

Coarelli 1976.

300

The Lucus Pisaurensis and the Romanization

which seem to suggest through chronological variations even more decades between the dedications, a fact that contributes to confirming the impossibility of a date later than 184 for the more ancient ones, which would imply a date too late in the second century for the more recent ones. From this point of view, it is significant that there is at least one example that can be assigned to the first half of the second century: this is CIL I2 373, with a dedication to Salus, where there appears for the only time the L with a right-angle, but always in the presence of a dative in -e, which cannot in any case be lowered later than the middle of the second century.18 Paradoxically, if anything seems clear from an internal, but also rapid, examination of these dedications, it is that they seem to dry up just at the moment that coincides with the foundation of the colony; in fact, there are no inscriptions that have come down to us and can be dated to the end of the Republic or to the imperial age; this seems explicable only in the event of an abandonment, at least partial, of the sanctuary in these same years. But we shall come back to this point later on. Another source of information, systematically ignored down to recent years, is archaeological, consisting of material from the stips votiva belonging to the sanctuary.19 This particularly concerns a group of terracotta ex-voto’s (heads, anatomical parts, animals) which reflect a very well-known typology, securely datable between the second half of the fourth and the beginning of the second century ,20 in full agreement with the chronology of the inscriptions that we are proposing. One objection that could be brought up here can be disposed of speedily: the votive terracottas, while they prove the existence of the sanctuary in the middle republican period, would nevertheless not show that the Latin dedications also go back to the same period; it might be a question, in other words, of an indigenous sanctuary later absorbed into the ambit of the colony and ‘Romanized’. This seems impossible in any case: from the moment it was clear that this type of ex-voto was characteristic exclusively of the Latial culture, which provides in effect one of the securest fossil-guides for identifying the presence, outside their area of origin, of colonists coming from Rome and Latium. The presence of such objects in the grove of Pesaro proves, without any possibility of doubt, its use by viritane settlers, at a date that cannot in any case be after the first years of the second century —at the latest. On the contrary,

18

19 Coarelli 1986: 43–4. De Luca 1984: 71–84. Cf., inter alia, Pinna 1979: 205–6; Ferrea 1979: 2307–8; Ferrea and Pinna 1986: 89–144; Steingräber 1980: 215–53; Comella 1981: 717–803; Pensabene (ed.) 1980. 20

The Lucus Pisaurensis and the Romanization

301

certain objects (in particular the feminine heads with nimbi (clouds of light)) should rather be dated to the first half of the third century. Through the independent analysis of another type of material originating from the grove, we arrive at the same conclusion as that suggested by the examination of the inscriptions. We are thus in the presence of evidence of cult, explicable only through an early occupation by colonists coming from Rome or at least from areas close to the city, that cannot in any case be connected with the foundation of the colony of Pisaurum. It will be impossible, therefore, to be dealing with any others than coloni viritani (individual colonists) who should be those settled by C. Flaminius in 232 ,21 but perhaps also in part other colonists of an earlier settlement, immediately after the conquest of the ager Gallicus (therefore after 283 ),22 perhaps at the time of the acquisition of the ager Picenus (268 )23 and the time of the foundation of the Latin colony of Ariminum (likewise, 268 ).24 We shall examine later on some indications that seem to confirm such a hypothesis, which was already Mommsen’s. I would like now to put briefly under examination the more properly cultic of the inscribed stones and in first place the nature of the divinities mentioned in them. This concerns: Apollo, Fides, Iuno, Iuno Lucina, Iuno Regina, Mater Matuta (twice), Diana, Liber Pater, Salus, Feronia, Dii Novensides, Diva Marica.25 Beside the evidently ‘Servian’ divinities, such as Diana, Iuno Lucina, and Mater Matuta,26 we find others connected with Camillus and the conquest of Veii (Iuno Regina and again the Mater Matuta),27 still others introduced into Rome from the Sabina (Feronia, Dii Novensides)28 probably following on the victory of 290 , at the hands of Curius Dentatus29 (a presence, this, of particular significance in the ager Gallicus) or from the area of the Aurunci (Diva Marica). In this last case, we can prove a connection with the foundation of the colonia civium Romanorum at Minturnae, in 298 .30 The strongest plebeian connection of all these cults probably also involves Apollo, who is closely connected with the plebeian area of the Circus Flaminius,31 and with the same Salus. Particularly notable, from this point of view, is the presence of Liber Pater. We are dealing with a kind of plebeian Pantheon, which can be associated with a precise historical situation, namely that of Rome in 21 23 26 28 31

22 Bandelli 1988: 3ff.; Hermon 1988: 273–84. Bandelli 1988: 2. 24 25 Bandelli 1988. Bandelli 1988: 6. See n. 1. 27 Coarelli 1988: 215ff., 245ff. Coarelli 1988: 208ff. 29 30 Coarelli 1997: 197–209. Ibid. Trotta 1989: 11–20. Coarelli 1997: 363–74.

302

The Lucus Pisaurensis and the Romanization

the first decades of the third century , over which the guardian heroes can be recognized as Servius Tullius, Camillus, and Curius Dentatus; this is a cultural (and social) conjunction, quite appropriate for viritane colonists sent out in the course of the third century , soon after the permanent resolution of the age-old conflict between the plebs and the patriciate.32 If there are any surprises here, they concern certain absences, amongst which the absence of Fortuna is particularly obvious: in my view, this absence is totally accidental, caused by a gap in the documentation, which we are in the process of bridging. In the first place, Fortuna’s presence seems presupposed by the evidence for Mater Matuta, a sister cult.33 In the second place, the goddess’ presence is evident in analogous contexts inside the same ager Gallicus; from the nearby settlement of Fanum Fortunae to the many scattered sites along the course of the via Flaminia there is evidence for the cult, derived from epigraphic documents such as the one coming from the Spina district and connected with the already famous one of the Galicos colonos.34 But, above all, the cult of Fortune is definitely attested in the same territory of Pesaro: the evidence consists of a sandstone cippus in the shape of a truncated cone, just like the ones in the lucus, coming from the area of Candelara35 (from where comes also the funerary cippo of a certain M. Pleturius,36 dateable to the third century , from which we can once again see proof of the early presence of viritane colonists in the Pesaro area). On the cippus in question, which can probably be attributed to the last decades of the second century , there is a very rare inscribed dedication to Fortuna Respiciens, which is to say the characteristically ‘Servian’ form of Fortuna, thus demonstrating the presence in the area of Pesaro of the goddess’ cult, in an invocation entirely fitting in with the rest of the documentation in the grove. Another document—a quite well-known one this time—is able to confirm further such a conclusion: this is about the lot in the shape of a cobblestone, currently in the Fiesole Museum37 (Fig. 11.4), whose text reads: 32

33 Cassola 1962. Coarelli 1988: cit. n. 26. CIL I2 2877b. Uggeri 1977: 127–37, esp. 133–5; Bandelli 1988: 15. On the cult of Fortuna along the via Flaminia: Cresci Marrone and Mennella 1984: 157–60; Cenerini 1995: 129–42, esp. 141. 35 Cresci Marrone and Mennella 1984: 157–62, n. 18. For Fortuna Respiciens, cf. Coarelli 1988: 258–60; Anselmino, Ferrea, and Strazzulla 1990–1: 193–262; Anselmino and Strazzulla 1995: 276–8. 36 Cresci Marrone and Mennella 1984: 392f., n. 160. 37 CIL I2 2841 = ILLRP 1070. Guarducci 1949–51: 23–32; Guarducci 1960: 50–3; Guarducci 1972: 183–9; A. Degrassi 1967: 124–8; Peruzzi 1959: 212–20; Mariotti 1959: 220. 34

The Lucus Pisaurensis and the Romanization

303

Fig. 11.4 The ‘sors’ (the lot), preserved in the Museum of Fiesole. Se ceduas perdere nolo; ni ceduas Fortuna Servios perit. If you yield, I have no wish to destroy; if you do not yield, Servius perishes by Fortuna.

The object cannot be dated later than the third century  and bears witness, at a period not in doubt, certainly earlier than the first annalists, to the strong connection between Fortuna and Servius Tullius; it originates quite certainly from the Marche.38 It is already recorded in the early 38

Degrassi 1967: Guarducci 1960: 188f.

304

The Lucus Pisaurensis and the Romanization

1800s by a scholar from the Romagna, Girolamo Amati, who was closely connected to the cultural environment of Pesaro—many of his letters to scholars of the locality are preserved in the Biblioteca Oliveriana.39 Margarita Guarducci suggested Fanum Fortunae as the lot’s provenance, but the context and date of the discovery, extremely early, might suggest that it originally derived from the lucus Pisaurensis. This is, of course, just a matter of a hypothesis, which if proved, might be able to confirm definitely the Servian character of several of the cults of the Grove. These data and the evidence collected so far now seem to justify a preliminary conclusion: the higher chronology of the dedications seems definitively confirmed and likewise the consequent attribution of the grove to a concili abulum (community) of viritane colonists, while the suggested connection with the colony of Roman citizens at Pesaro is ruled out. It is especially significant, from this point of view, that the most relevant documentation deriving from the sanctuary seems to dry up just at the moment of the colony’s foundation. It is, in fact, hard to imagine that such a number of cults, and such significant ones, should ever have been established in the country, not in the city, if the city had already been in existence. In the same way, the foundation of the city must have involved transferring the same cults into the new entity. The structure of the lucus—with that characteristic concentration of cults, especially feminine cults and cults of salvation, localized in a village (while at least one other analogous cult, that of Fortuna Respiciens, found its place in a second village)—seems thoroughly typical of preurban situations of which a better-known example is the one documented by the Agnone Tablet.40 An illuminating confirmation can be supplied by the parallel case of the sanctuary of Macchia Grande at Veii,41 a temple long since identified as a cult-place linked with the viritane colonists sent out shortly after the capture of the city through Camillus (him again!) and where there appears, in addition to Apollo, the characteristically plebeian cult of Iuppiter Libertas. Mario Torelli42 has rightly associated this episode of colonization with the terra cotta statuettes of Aeneas and Anchises, symbol of the foundation at Veii of a ‘New Troy’ and therefore of a plebeian ‘New Rome’. A subsequent series of inferences flows from the consideration of some of the names attested by the inscriptions of the grove, when compared 39 40 41 42

Guarducci 1960. Vetter 1953: n. 147. See the Acts of the Colloquium: Del Tutto Palma (ed.) 1996. CIL I2 2628–32 = ILLRP 27–31. NSA 1922. 387f. M. Torelli 1988: 65–72, esp. 69ff. Torelli 1984: 227; Torelli 1987: 128f.

The Lucus Pisaurensis and the Romanization

305

with the specific historical moment, which saw the colonization of the ager Gallicus, and when compared also with one particular version of the myth-historical tale of the Gallic gold and again with the foundation at Rome of the collegium matronarum (College of Matrons). The pre-eminence of the feminine element in the Grove seems certain; it is recognizable from the dedicants (four women, to two men), but also from the divinities (ten feminine, two masculine—plus the Novensides). The matronae Pisaurenses (matrons of Pesara), mentioned explicitly in one of the inscriptions, recall the existence on the spot of an ordo of the matrons, in which we should probably identify the main organizing body for the sanctuary.43 The festival of the Matronalia ought to have been one of the principal manifestations of this college: we know that the Festival happened on the first of March, the birthday of Iuno Lucina,44 whose temple at Rome [was built] on the site of an archaic fanum, as we know from the fasti Praenestini (which is to say from Verrius Flaccus);45 that it was carried out by the Matronae and vowed by the daughter (or wife) of Albinius, the plebeian who [rescued?] the Vestals, by taking them to Cerveteri at the moment of the Gallic conflagration.46 We are in the presence of a cult that tradition connected to Servius Tullius47 and which was renewed after the Gallic conquest of Rome, and probably in relation to that event. Also significant in this context is the matronal festival of the Matralia,48 connected to the Mater Matuta: and that is to say, once again, to a Servian cult, renewed by Camillus. Now, both Iuno Lucina and Mater Matuta are attested as among the cults of the grove. We shall see further on what the significance can be of the presence at Pesaro of the ordo matronarum and of the principal cults with which the ordo was linked. In the first place, then, I would like to linger over the most important of the cippi from Pesaro, the one containing a dedication to the Mater Matuta from two matrons,49 who must have played senior roles within the College. The women in question were one Mania Curia and one Polla Livia, both, as was already underlined by Mommsen,50 using the praenomen, that certain index of great antiquity. The name Curia, especially in conjunction with the praenomen Mania, cannot fail, in this particular context, to evoke the name of the conqueror of the ager Gallicus, Manius Curius Dentatus, all the more so, since we are dealing here with a gentilician name of the greatest rarity at Rome 43 45 47 48

44 Peruzzi 1990: 29ff. Degrassi 1962: 418ff. 46 Ibid. Gagé 1963: 70ff. Dionysius of Halicarnassus, Roman Antiquities 4. 15. 5; Giannelli 1996: 122ff. 49 50 Degrassi 1962: 468f. CIL I2 379 = ILLRP 24. See n. 5.

306

The Lucus Pisaurensis and the Romanization

(or anywhere else).51 These remarks, taken together with the inscription’s chronology (which in my opinion goes back to the first half of the third century) seems to suggest a direct connection between the consul of 290 and 275  and at least one eminent member of the local aristocracy. The possibility cannot be excluded that the person concerned might have been a relative of Curius Dentatus. In this case, the reality of a first viritane distribution would be confirmed; it would have occurred soon after the conquest of the ager Gallicus and been brought into effect by the author of the conquest himself, a transaction of which no record survives, owing to the disastrous state of the evidence relating to these years.52 Just as interesting is the name of the second authoress of the dedication, Polla Livia, who deserves a somewhat more extended discussion. Generally, it has been M. Livius Salinator who has been connected with the conquest, as one of the victors of the Metaurus;53 but the locality where this took place does not, in my view, provide an argument to use for connecting the Livii Salinatores with the colonization of the ager Gallicus. On the other hand, another somewhat neglected tradition, re-evaluated not very long ago by Lorenzo Braccesi,54 seems able to provide a more reliable answer. As we have mentioned already, the foundation of the collegium matronarum was attributed to Camillus and connected to a donation of golden ornaments (the aurum matronarum, to put it precisely), explained in our sources in two ways:55 either through the production of a gold crater to send to Delphi after the conquest of Veii, or else through the necessity of redeeming the city of the Galli Senones. Now, a curious tradition explained the name of Pisaurum in direct relation to the recovery of the gold handed over to the Galli by the Romans: ‘Pisaurum was so-called because it was there that the gold was weighed out’ (pensatum aurum).56 This forced the unidentified annalist from whom this version derived, to push Camillus’ pursuit of the Galli as far as Pesaro. The absurdity of this tradition, evident in itself, calls for no refutation. But the question remains: what was the problem to which the

51 Apart from the most famous Curius Dentatus, we only know one Curius in the Roman fasti before the first century —one M’. Curius, tribune of the plebs in 198 : RE IV. ‘Curius’ (4); Broughton 1950–60: II. 558. 52 The sources from 293  onwards are collected in M. R. Torelli 1978. 53 See Münzer 1926: cols. 853–5. 54 Braccesi 1982–3: 77–98; Braccesi 1984: 1–38; Gagé 1963: 202ff. 55 Gagé 1963: 154ff. 56 Servius, ad Aeneid 6. 825; cf. Isidore, Etymologiae 17. 4. 10.

The Lucus Pisaurensis and the Romanization

307

response was inserting the name of Pisaurum into the narrative of one of the episodes related to the Gallic conflagration? As we all know, the appearance of Camillus in this episode is the work of the annalistic tradition, while Polybius, for instance, totally ignores it.57 This alternative version is in relation to the recovery of the ‘Gallic Gold’, as reported by Suetonius in his Life of Tiberius:58 Drusus by killing the enemy leader, Drausus, gained the honorific name (cognomen) for himself and his descendants. It is also told of him that, while he was the Governor of the Province of Gaul, he brought back the gold long ago handed over to the Senones, during the siege of the Capitol; this contradicts the traditional story that it was extracted from them by Camillus. It was the great-great-grandson of this Drusus who was named ‘patron of the Senate’ because of his outstanding work against the Gracchi.

So, the cognomen ‘Drusus’ conferred on one branch of the gens Livia would have been derived from the fact that, in the course of the conquest of the ager Gallicus (that is, around 283 ), Drusus had killed Drausus, the chieftain of the Gauls, with his own hand. It would have been appropriate for him to have been the man who recovered the Gallic gold, about a century after Rome’s conquest by the Senones. The reliability of this notice (and its antiquity) is already implied by the fact that it seems incompatible with the vulgate version, which attributes the recovery to Camillus. We are undoubtedly in the presence of a version that we can attribute to an annalist. Confirmation is provided by a passage of a Ciceronian letter, generally misinterpreted,59 which makes it apparent that the recovery of the Gallic Gold by a Livius Drusus was already proverbial in the Late Republic: (ad Atticum 2. 7. 3 = 27.3 (Shackleton)): illa opima ad exigendas pecunias Druso, ut opinor, Pisaurensi an epuloni Vatinio reservatur. (the fat mission (‘fat’, as in pulling in the money), is reserved for Drusus, I guess, the man of Pesaro; or else for Vatinius, the feast-priest).

This is not, as has been supposed, referring to a Livius coming from a branch of the Livii originating in Pesaro, but simply to an ancestor of Tiberius, recalled by Suetonius—an ancestor who had recovered at Pesaro the gold taken by the Galli at Rome. This character ought to be historical: according to Suetonius, we are dealing with the great-grandfather (abnepos) of the tribune of the plebs 57 58

Polybius 2. 18. 2–6. Cf. Ogilvie 1965: 727ff.; Piccirilli 1983: xxxi ff. 59 Suetonius, Tiberius 3. 2. Cicero, ad Atticum 2. 3. 7.

308

The Lucus Pisaurensis and the Romanization

of 122 , M. Livius Drusus, the adversary of Gaius Gracchus and father of the famous tribune of 91  (RE XIII, Livius 12–13). This inevitably takes us back to the first half of the third century , that is to say back to a period, which at least for Suetonius if not for us, was historically documented in full. We should be dealing here with an associate (a legatus?) of Curius Dentatus, at the very moment of the latter’s conquest of the ager Gallicus.60 The presence of a Polla Livia alongside a Mania Curia in an inscription that belongs in the period of the first viritane settlements in the territory of Pesaro thus seems to take on its full significance. It should also be a fact not without importance that this involves two members of the local ordo matronarum, a copy of the prestigious model at Rome. That the tradition of the aurum matronarum is foregrounded in both cases, at Rome as at Pesaro, is perhaps—once again—not just a random fact.

A FT E R W O R D Since this article, I have published two other papers on similar arguments (Coarelli 2005; 2017). The first illustrates a group of inscriptions with the names of divinities from a new sanctuary from the viritane colonization of the ager Pomptinus (which previously belonged to the Volscan city of Privernum), the second re-examines and develops the study of the sors of Fiesole, already discussed in here.

B I B L I OG R A P H Y Anselmino L., L. Ferrea, and M. J. Strazzulla (1990–1). ‘Il frontone di via San Gregorio ed il tempio della Fortuna Redux sul Palatino: una nuova ipotesi’. RPAA 63: 193–262. Anselmino, L. and M. J. Strazzulla (1995). ‘Fortuna Redux, aedes’. LTUR II: 276–8. Bandelli, G. (1988). Ricerche sulla colonizzazione romana della Gallia Cisalpina. Rome. Bormann, E., ed. (1888–1926). In Corpus Inscriptionum Latinarum XI, p. 942. Braccesi, L. (1982–3). ‘Pesaro romana, moribunda e felix’. SOliv n.s. 2–3: 77–98. Braccesi, L. (1984). ‘Lineamenti di storia pesarese in età antica’, in Pesaro nell’antichità. Storia e monumenti. Venice, 1–38, esp. 4–7.

60

Peruzzi 1990: 67; Gagé 1950: 141–76, esp. 162–5.

The Lucus Pisaurensis and the Romanization

309

Broughton, T. R. S. (1950–60). The Magistrates of the Roman Republic. New York. Cassola, F. (1962). I gruppi politici romani nel III secolo a.C. Trieste. Cenerini, F. (1995). ‘Caio Flaminio: uomo politico, homo religiosus’, in A. Calbi and G. Susini (eds.), Pro poplo ariminese. Epigrafia e Antichità 14. Faenza, 129–42, esp. 141ff. Cenerini, F. (1976). ‘Un elmo con iscrizione latina arcaica al Museo di Cremona’, in L’Italie préromaine et la Rome républicaine (Mélanges Jacques Heurgon) I. Rome, 157–79, esp. 160, n. 7. Cenerini, F. (1986). ‘Le iscrizioni’, in Coarelli (ed.), Fregellae 2: Il santuario di Esculapio. Rome. Cenerini, F. (1988). Il Foro Boario. Rome. Cenerini, F. (1997). Il Campo Marzio. Rome. Coarelli, F. (1976). ‘Un elmo con iscrizione latina arcaica al museo di Cremona’, in L’Italie préromaine et la Rome républicaine (Mélanges J. Heurgon). Rome, 157–79 esp. 160, n. 7. 245ff. Coarelli, F. (1986). ‘Le iscrizioni’ in Fregellae 2: Il santuario di Esculapio. Rome, 43–4. Coarelli, F. (1988). Il Foro Boario. Rome, 215ff; 245ff. Coarelli, F. (1997). Il Campo Marzio. Rome, 197–209. Coarelli, F. (2005). ‘Un santuario medio-repubblicano a Posta di Mesa’, in W. V. Harris and E. Lo Cascio (eds.), Noctes Campanae. Studi di storia antica e di archeologia dell’Italia pre-romana e romana in memoria di M. W. Frederiksen. Naples., 181–90. Coarelli, F. (2017). ‘La sors di Fiesole e il culto di Fortuna nelle Marche’, in O. Mei and P. Clini (eds.), Fanum Fortunae e il culto della dea Fortuna. Venezia, 19–29. Comella, A. (1981). ‘Tipologia e diffusione dei complessi votivi in Italia in epoca medio e tardo repubblicana’. MEFRA 93: 717–803. Cresci Marrone, G. and G. Mennella (1984). Pisaurum I: Le iscrizioni della colonia. Pisa. Degrassi, A. (1957–63). Inscriptiones Latinae Liberae Rei Publicae I–II. Florence. Degrassi, A. (1962). Inscriptiones Italiae XIII, fasc. 1–3. 418ff. Degrassi, A. (1967). ‘Epigraphica III Notizie sulla sors del Museo di Fiesole’. MAL s. 8. 13. 1. 30–3 = Scritti vari III. Venezia-Trieste, 124–8. Del Tutto Palma, L., ed. (1996). La tavola di Agnone nel contesto italico. Florence. De Luca, M. T. (1984). ‘Il lucus Pisaurensis’, in Pesaro nell’antiquità. Venice, 712–84. Durante, M. (1962). ‘Commento all’iscrizione di Novilara’. Ricerche linguistici 5. Durante, M. (1978). ‘Nord Piceno: la lingua delle iscrizioni di Novilara’, in Popoli e civiltà dell’Italia antica VI. Rome, 393–400, esp. 397f. Ferrea, L. and Pinna A. (1979). ‘Il deposito votivo’, in Fregellae 2, 89–144. Ferrea, L. (1979). ‘Teste votive da Fregellae’. Arch. Laz. 2: 207–8. Filippi, G. (1984). ‘Recenti acquisizioni su abitati e lughi di culto nell’ager Aequiculanus’. Arch. Laz. 6: 265–77, esp. 174f.

310

The Lucus Pisaurensis and the Romanization

Gagé, J. (1963). Matronalia. Collection Latomus 60. Brussels. Gianelli, G. (1996). Art. ‘Iuno Lucina: aedes’, in LTUR III: 122f. Guarducci, M. (1949–51). ‘La Fortuna di Servio Tullio in un’antichissima Sors’. RPAA 25–6: 23–32. Guarducci, M. (1960). ‘Ancora sull’antichissima sors con il nome di Servio Tullio’. PP 15: 50–3. Guarducci, M. (1972). ‘Ancora sull’antichissima sors della Fortuna di Servio Tullio’. RAL s. 8 27, 183–9. Hermon, E. (1988). ‘La lex Flaminia de Agro Gallico Dividundo—modèle de romanisation au IIIe siècle av. J.-C.’ Mélanges Lévêque II. Paris, 273–84. Luni, M. (1984). ‘Topografia storica di Pisaurum e del territorio’, in Pesaro nell’ antichita. Storia e monumenti. Venice, 109–80. Mariotti, S. (1959). ‘Postilla’. PP 14, 220. Mommsen, Th. (1850). Die unteritalischen Dialekte. Leipzig. Mommsen, Th., ed. (1863). Corpus Inscriptionum Latinarum I1 32ff. Münzer, F. (1926). RE XIII 1: 853–5: s.v. Livius (12–13). Palma, L. Del Tutto, ed. (1996). La tavola di Agnone nel contesto italico. Florence. Pensabene, P., ed. (1980). Terrecotte votive dal Tevere. Studi miscellanei 25. Rome. Peruzzi, E. (1959). ‘Un’antichissima sors con iscrizione Latina’. PP 14: 212–20. Peruzzi, E. (1990). I Romani di Pesaro e i Sabini di Roma. Florence. Pinna, A. (1979). ‘Il deposito votivo di Fregellae’. Arch. Laz. 2: 205–6. Pinna, A. (1986). ‘Il deposito votivo, in Fregellae 2, 89–144. Ritschl, F. (1859). ‘Epigraphische Briefe 5’. RhM 14: 400–2 = 1862 Opuscula philologica IV, Leipzig, tabb. XLIII ff. Ritschl, F. (1862). Priscae Latinitatis monumenta epigraphica. Bonn. Steingräber, S. (1980). ‘Der Phänomen der etruskisch-italischen Votivköpfe’. MDAI(R) 87: 215–43. Torelli, M. (1984). Lavinio e Roma. Rome. Torelli, M. (1987). La societá Etrusca. Rome Torelli, M. (1988). ‘Aspetti ideologici dell colonizzazione romana più antica’. Dial. Arch. 3a ser. 6.2: 65–72 (esp. 69ff.). Torelli, M. R. (1978). Rerum Romanarum fontes. Pisa. Trotta, F. (1989). ‘Minturnae preromana e il culto di Marica’, in F. Coarelli (ed.), Minturnae. Rome, 11–20. Uggeri, G. (1977). ‘Nuovi testi epigrafici dell’antico delta padano’. A & R 22: 126–37. Vetter, E. (1953). Handbuch der italischen Dialekte. Heidelberg.

12 Rome and the Great Places of Worship in Italy John Scheid

In 32  Octavian declared war on Cleopatra, in accordance with the ritual of the fetiales.1 As there had been no evidence of this priesthood since the second century , the likelihood is that he restored it around this date.2 In fact, this was not an isolated measure. Its main purpose was to demonstrate that the war against Egypt was a just war. Nonetheless, a brief look at the sources throws up an array of apparently connected measures which provide a more complex picture of this initiative. It reveals the Romulean background to the procedure and shows it, indeed, to belong into an already ancient Roman tradition which consists in one way or another of taking ownership of the great Italic places of worship. 1. Octavian recreated the college of the fetiales around the same time that he restored the temple of Jupiter Feretrius,3 who was closely linked to treaties and to the fetiales. Clearly Octavian could not restore one without the other. These were not isolated measures. It may be tempting to consider in the same context the restoration of another priesthood, that of the Caeninenses. Of course, the first known Caeninensis, Q. Trebellius Maximus, of Toulouse, dates from the Flavian period.4 But from the functions belonging to this priesthood of equestrian rank, it is possible to reconstruct a plausible Augustan context. Judging by the Caeninenses we know about, this was the highest-ranking equestrian priesthood.5 What was their function? The sources do not say. The Caeninenses clearly had some relation to Caenina, an ancient city 1

2 Dio Cassius 50. 4. 4–5. Flower 2000: 34–64, spec. 46, n. 75. See n. 22 for the sources. The decision to restore the temple was taken before March 32, as Atticus gave Octavian the idea for it. 4 5 IG II–III², 4193; AE 1947, 69. Granimo Cecere 1999: 1–112, esp. 99ff. 3

John Scheid, Rome and the Great Places of Worship in Italy In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0013

312

Rome and the Great Places of Worship in Italy

situated not far from Rome. With Georg Wissowa, then, we might conjecture that following the conquest of Caenina by Romulus, the rites of this city were transferred to Rome and entrusted to the sacerdotes Caeninenses, in the same way as the cults of Lavinius (Laurentes Lauinates) or of Monte Cavo (Cabenses).6 Judging by the etymology of their name, the function of the Caeninenses does indeed resemble that of the Laurentes Lauinates, whose duty it was to worship the gods of Lavinium with the magistrates and priests of Rome. They formed the elite of the people of Lavinium, which had long since disappeared. Thus, among the Laurentes Lauinates were found the moneylenders, the pontiffs, the flamines, and the Salii of the Laurentes. In fact, the Athenian inscription that mentions the priesthood of Trebellius Maximus7 reveals it also to include the function of Καινείνσις etc., that is to say, consul Caeninensis sacrorum populi Romani, which recalls the titles of the Laurentes. Another inscription tells us that the Caeninenses were creati by the Roman pontiffs.8 There is a clear relation between these priests and Romulus, as according to Roman myth it is the latter who conquered and destroyed Caenina. This connection suggests that the creation of the Caeninenses should be placed in the period prior to the year 27, when Octavian was thinking of adopting the agnomen of Romulus. From 27 on, Augustus abandoned any too direct reference to Romulus. Caenina was also connected to the temple of Jupiter Feretrius by the history and institution of spolia opima, which Romulus had taken from the king of Caenina, Acro, whom he had killed with his own hand.9 The spolia, in other words the king’s armour, were said to have been placed by Romulus in this temple. An anecdote10 concerning the restoration of the building makes reference precisely to the problem of the spolia opima, in which the future Augustus showed a keen interest, judging by the case of Licinius Crassus, to whom Octavian denied the privilege of rich spoils in 28 .11 To understand the interest in rich spoils, it is important to recall that a senatus consultum had in 44 conferred on Julius Caesar the right to place rich spoils in the temple of Jupiter Feretrius,12 as if he had killed the enemy general with his own hand. Octavian in all likelihood took inspiration from this Caesarean privilege. Putting all these accounts together, we may in one respect consider that the restoration of the fetials and of the temple of Jupiter Feretrius, and the creation or revival of the Caeninenses were most probably all part of the same initiative; and in another that Octavian’s intention was to 6 8 11

Such, for example, was the opinion of Wissowa 19122: 520. 9 10 CIL XI, 3103. Flower 2000: 42ff. Livy 4. 20. 12 Flower 2000: 49–53. Dio Cassius 44. 4. 3.

7

See n. 4.

Rome and the Great Places of Worship in Italy

313

achieve a splendid and archaic triumph—a triumph over Romulus, the founding king. For this to occur, all the rites connected with the temple of Feretrius, the ancient rites for the declaration of war as well as those commemorating in some manner or other the feat of Romulus—rites which perhaps involved the Caeninenses—had to be restored. The initiative of 32 was essentially a matter of internal politics. No less was it part of an already ancient Roman tradition, which consisted in exploiting the reputation of the great Italic places of worship. This tradition was furthermore exploited more widely by Augustus. 2. Over the course of the first century , relations between Rome and the Italics fluctuated between extreme suspicion—as eloquently attested by Sulla’s words on the Samnites—and the integration that took place over the decades following the Social War and especially at the start of the Empire. The evolution of these relations is etched into the history of the great places of worship of Italy, places that enjoyed regional and supra-regional influence and played a ‘federal’ role in the Italic world. These were mainly sanctuaries lying outside the city or on its outskirts, such as the one overlooking the city of Praeneste. The sanctuary of Fortuna Primigenia at Praeneste is not as symbolic as the one in Pietrabbondante, but in the aftermath of the Social War and the conflicts between Marius and Sulla, it came close to knowing the same fate. For Cicero,13 the sanctuary of Fortuna retains its fame amongst the uolgus due to its age and beauty. He is of the view, though, that no man with any kind of public status would ever consult the famous Praenestine oracle again. Even if we interpret this assertion in the context of the Ciceronian dialogue and thus of the critique of divination by sortes, we have to conclude that in the middle of the first century , the magistrates and the Roman elite no longer consulted the oracle of Fortuna, in spite of its fame. Such scorn, however, was shortlived. Under the Empire, people of quality made dedications there, and furthermore the oracle was not lacking in potential political influence. This is the apparent implication behind Tiberius’s desire to outlaw the Praenestine oracle or to reserve it for his own use.14 Be that as it may, the shutting down of the oracle is certainly connected with the treatment to which in 83 the Romans had subjected the sanctuary and the city of Praeneste, where Marius the Younger had taken refuge with the Samnites. As much as the massacre of the Praenestines when the city was sacked, the setting up of a colony profoundly altered the elite of Praeneste. After the colony was founded, the sanctuary and its oracle

13

Cicero, De divinatione 2. 86.

14

Suetonius, Tiberius 63. 1.

314

Rome and the Great Places of Worship in Italy

functioned in accordance with Roman sacred law, under the control of the Roman settlers, who did not have the social rank of the oracle’s previous patrons. This change was in all likelihood at the root of the oracle’s decline in prestige amongst the Romans and certainly the Italics. Nonetheless, the site had lost nothing of its symbolic power, for it is not only for its riches and its topographical qualities that the sanctuary of Fortuna served as a refuge for Fulvia, wife of Mark Antony, and for the troops of L. Antonius during the Perusine War. One has the impression that it was just as much the symbolic power of the sanctuary that attracted the interest of those who claimed to be in support of the Italics in their anger at the brutal settling of the veterans freed after Philippi. The example of Praeneste is, however, an unusual one, concerning as it does a city as much as a temple. Other sites are of greater interest for our purposes here. The case of Pietrabbondante does not yield any further information, as excavations show it to have been abandoned after the Social War. Its abandonment was the result of the definitive damnatio memoriae handed down to a place of worship that for a while served as a political centre for insurgents, as the inscriptions attest.15 But this is an extreme case. Other sanctuaries had a less troubled history. Let us come back once more to Sulla. The way he handled the sanctuary of Diana Tifatina near Capua appears to have pointed the way for his successors. The temple of Diana on Mount Tifata was incontestably the most famous sanctuary in Campania. Following the victory over Norbanus on Mount Tifata, Sulla gave waters and grounds to the sanctuary.16 This gift was registered by Augustus and reaffirmed by Vespasian.17 Together with the privilege of the right to own property, the attestation in the Imperial period of a pr(aefectus) i(ure) d(icundo) is proof that the sanctuary and its grounds thenceforth formed an independent district.18 The place of worship evidently enjoyed considerable symbolic power, a repute which Hannibal may also have exploited, as his camp was set up on Mount Tifata.19 Sulla and his successors therefore took this site away from Capua so as to make it an autonomous entity, akin to a municipium or a prefecture. We do not know exactly what kind of status it held, but the principle behind the initiative is clear: to remove this famous sanctuary and site from any outside influence, so as to make it autonomous, that is to say solely dependent on Rome. The same initiative handed over, in some 15 16 17

Poccetti 1979: 13, 16, 19. Velleius Paterculus 2. 25. 4; Pobjoy 1997: 59–88; see also Keppie 1983: 143. 18 19 ILS 251, 3240. Heurgon 1942: 300. Livy 23. 36. 1.

Rome and the Great Places of Worship in Italy

315

sense, the goddess to Rome and its interests. Sulla may have drawn inspiration from the particular set of connections between the Romans and the Venus of Mount Eryx in Sicily. Over the course of the Second Punic War, the temple of Venus on Mount Eryx, which had already proved favourable to the Romans during the First Punic War, became as a part of Rome’s territory. When a consul or general came to Sicily, he would generally first come to Mount Eryx to offer sacrifice there.20 At the same time, in 215 the goddess received a temple in Rome. We might note that Fortuna Primigenia also enjoyed the benefit of a chapel or an altar on the Capitoline,21 which may go back to the time when the colony of Praeneste was founded.22 These were not isolated interventions, as is shown by a number of other examples. Following the final surrender of the Latins, in 338 , three great places of worship were under the permanent control of the Romans. The federal sanctuary of the Latin League on Monte Cavo was thenceforth a place of Roman worship and the Latin feria were likewise celebrated in Rome while the gathering of Monte Cavo was in progress.23 The temple of Juno Sospes at Lanuvium became a cult common to both Romans and Lanuvians.24 And at Lavinium, several shared annual rites were celebrated amongst the Lavinates and the magistrates and priests of Rome.25 A further example is provided by the Umbrian place of worship at the sources of the Clitumnus. The sources of the Clitumnus were manifestly one of the great regional sites of the Valle Umbra. While not attested by any source from before the Empire, the fame held by this place in Augustan poetry shows that it was not of recent renown. As they are situated very near Mevania, we may take it that in the third century they were managed by the Umbrians who had settled in this town or the surrounding area. When in 241 the Latin colony of Spoletium was founded,26 the sources of the Clitumnus were clearly attributed to the new colony, as later on they lay at the perimeter of Spoletium.27 It may be that the Umbrians’ great ethnic sanctuary of the Villa Fidelia, at Spoletium,28 was likewise given to the new colony, but there is no 20 Schilling, R. (1982): Préface à la deuxième édition de La religion romaine de Vénus. Paris, 239–64. 21 Schilling 1954: 264ff. 22 CIL XIV, 2852; Plutarch, De Fortuèna Romanorum 10.e. 23 24 Wissowa 1912²: (1982). 124ff. Livy 8. 14. 2. 25 26 Livy 8. 11. 15; Macrobius, Saturnalia 3. 4. 11.la. Pietrangeli 1939. 27 Philargyria, On Virgil’s Georgics 2. 146: Clitumnus et deus et lacus in finibus Spoletinorum (‘Clitumnus is a god and a lake in the lands of the Spoletini’). See Pietrangeli 1939. 28 Coarelli 2001: 39–52.

316

Rome and the Great Places of Worship in Italy

evidence for this. Whatever the case may be, Rome thereby controlled through its colony the supra-regional places of remembrance and of gathering in the Umbrian Valley. 3. Octavian’s initiative, which was connected to the prospect of an archaic, ‘Romulean’ triumph, therefore belonged to an already ancient tradition. The gods of enemies were either evoked or annexed. In one way or another they became Roman deities or deities controlled by Rome. A series of later initiatives shows that this was a systematic and deliberate policy. Augustus thus turned the sacred forest of Lucus Feroniae, in meridional Etruria, into a Roman colony, perhaps also assigning to it the neighbouring sanctuary of Apollo Soranus on Mount Soracte. Caesar appears already to have been attracted to the site and its history, as around 46 he set up a colony at Capena, another abandoned ancient city, which formerly administered the sacred forest of Feronia. This endeavour on Caesar’s part must doubtless be interpreted as a demonstration of interest in this symbolic site. If this assumption is correct, Caesar no doubt planned to get back the previous judicial organization by reuniting Capena with the neighbouring forest of Feronia. But the attempt appears to have failed, since after Actium a Colonia Iulia Felix Lucoferoniensis was founded at Lucus Feroniae itself, with the sacred forest as a sort of urban centre. Like the sanctuary of Diana Tifatina, which became a kind of municipium or prefecture, the sacred forest of Feronia and its estates were transformed into a city, one which lasted until at least the third century , as it still offered honorary dedications to the tetrarchs Gallienus, Constantius Chlorus, Galierius, Flavius Valerius Serverus, and Maximinius Daia.29 The colony-temple possessed, or rather consisted of, a monumental centre, which formed more of a sort of annex to the great villa of the Volusii than a headquarters. This highly artificial colony, essentially no more than a place of worship, offers a good example of Octavian’s intentions. In the main, the initiative consisted in ensuring that this famous site, which had once been plundered by Hannibal and whose supraregional value is beyond question, had an autonomous, Roman status.30 This place of remembrance thereby came to be restored and integrated into the Roman community. At the same time, it undoubtedly enjoyed the benefit of public funds, and received a share, at least, of settlers, as a way of keeping up the fiction. It is possible to conjecture that during the same period, or in any case at the end of the first century , or a little later, the sanctuary of Apollo 29

AE 1988, 554, 559.

30

Dionysius of Halicarnassus 3. 32. 1; Livy 26. 11. 8–10.

Rome and the Great Places of Worship in Italy

317

Soranus on Mount Soracte was assigned to the new colony. The sanctuary was well known. In his description of Lucus Feroniae, Strabo in fact refers to the rites of Soracte.31 For his part, Pliny the Elder also describes the rites of the Hirpi Sorani,32 noting that a senator-consul had granted them immunity. Yet even if the immunity granted to the families of the Hirpi was an already ancient privilege, the likelihood is that it had to be given within an institutional context. The Hirpi were immunes due to their sacerdotal responsibilities, as were the priests of Rome, or those of the Roman colonies. Unless there was a direct requisition of the local cult of the Hirpi Sorani by the city of Rome—we might think here of the Caeninenses—this immunity may have been granted to the Hirpi within the framework of a neighbouring city. The hypothesis of a local cult promoted to Roman priestly rank is a suggestive one, as it would once more bring us back to the restorations of Octavian or Augustus, and to the priesthoods associated with these restorations. But no specifically Roman function of this kind is known. By contrast, the proximity of Lucus Feroniae and Mount Soracte, combined with Strabo’s confusion of the rites of the Hirpi with those of the festival of Feronia, would tend to indicate that it was within the framework provided by the foundation of the new colony of Lucus Feronia that the sanctuary of Apollo Soranus was allocated to this colony. Strabo’s error would therefore not come from one of his sources but from the context in which he himself discovered the cult, that is to say, in a document about the foundation of the colony of Lucus Feroniae. Of course, this is only a hypothesis, but given what happened around the same period at Hispellum, it is not an absurd one. The colony of Hispellum was set up by Octavian, no doubt after the Battle of Philippi; after 42 , then, at a time when land was being distributed to freed veterans. While we cannot be sure exactly, experts date its foundation from this year or the years immediately after. In any case, it is still before 28 . First of all here, the large sanctuary situated at Hispellum (Spello), in the villa Costanzi, might according to archaeologists have taken over from an important supraregional sanctuary of the Umbrians.33 Once more, the reason for founding a colony on this site, where apparently no town had previously been, was the presence of a renowned supraregional Italic place of worship. Unfortunately, we do not know any more at present, other than that under the Empire this sanctuary became the collective place of worship for the region of 31

32 Strabo 5. 226. Pliny, Natural History 7. 19. Manconi, Cruciani, and Camerieri 1997: 375–429; Sensi 1998: 457–77; Sensi 1999: 365–73; Coarelli 2001: 39–52. 33

318

Rome and the Great Places of Worship in Italy

Umbria:34 the famous rescript of Hispellum was discovered there.35 Clearly the site was very important. But it is another decision by Octavian, taken, it would appear, in the same context, that is of particular interest to us here: he assigned to the Hispellates the famous sanctuary at the sources of the Clitumnus,36 with the task of managing it and receiving visitors. In other words, he took away from Spoletium the famous place of worship which this city had upon its foundation been given to administer; and likewise he may have withdrawn from it the supraregional sanctuary of the Villa Fidelia, in order to make it into a colony and an urban centre. We do not know for what reason Octavian was punishing Spoletium. But we may venture the hypothesis that it may have involved a settling of scores connected to the troublesome context of the battles of Modena and Perugia.37 In Lucania, the sanctuary of Mefitis Utiana, which originally belonged to an Italic community, was allocated at the beginning of the Empire to the municipium of Potentia.38 Strictly speaking, then, it is not that the Italic place of worship was transferred to the urban centre of the municipium, as Michel Lejeune puts it, but that the Roman community took on the Italic place of worship. This is clearly attested by inscriptions from the Imperial period made by magistrates and found at Rossano di Vaglio. These can only have been the magistrates of Potentia. Octavian may have taken the same initiative at Fanum Fortunae (Fano) in Roman Umbria, where a colony was established after the Battle of Actium, on a site where there had manifestly been a sanctuary of Fortuna, but no town. Despite the close proximity of the mouth of the Metaurus, neither the temple nor the city are mentioned at the time of the Battle of the Metaurus. It has nonetheless been surmised that the origin of what to start off with was only a conciliabulum situated at the end of the via Flaminia was to be sought in the viritane assignments of the Lex de agro Piceno et Gallico uiritim diuidendo of 232. This conciliabulum would then have become a municipium and finally, in the Augustan period, a colony. But we cannot be certain in the matter. The first elements of certainty come from the Bellum ciuile. There Caesar39 relates that his troops occupied Fanum—or the fanum?—; this is all we can say. It might be added that if not the Latinized name of a local deity,

34

35 Gascou 1967: 609–59; Amann 2002: 1–28. CIL XI, 5265 (ILS 705). Pliny, Letters 8. 8; Scheid 1996: 246ff. 37 The only clue we possess lies in the favourable signs received at Spoletium by Octavian ahead of the Battle of Modena (Pliny, Natural History 11. 190). This clearly does not help us in our inquiry. 38 39 Lejeune 1999. Caesar, Civil Wars 1. 11. 4. 36

Rome and the Great Places of Worship in Italy

319

Fortuna was an exclusively Latin and Roman deity.40 If such were the case, we would be dealing with a Roman foundation, connected to the construction of the road and/or to the first colonization. Whatever the case may be, it is sufficiently unusual for a municipium or a colony to bear the name of a place of worship and of a goddess for us to imagine that the situation was that of a sanctuary transformed into a city. Another example might be provided by Cupra Marittima. This important place of worship, already frequented by the Etruscans, became a Roman colony after 49  or later, under Augustus. Strabo simply calls it the sanctuary of Cupra, Kúpra etc., without mentioning a city.41 The sacred forest of Angitia, near Lake Fucinus, knew the same fate. Its fame is evidenced by the cult’s spreading out into Marsia and beyond. For its part, the inscription of Caso Cantovius42 provides proof that the place of worship played a ‘federal’ role. We do not know anything about the status of the sacred wood at the time when this inscription was engraved. According to two inscriptions mentioning local magistrates and a boundary cippus43 between the lands of Alba Fucens, Marruuium, and the goddess Angitia, the sanctuary became a municipium after the Social War. Discovered in situ in the grounds recovered by Claudius’s draining of Lake Fucinus, the boundary stone provides a terminus ante quem for the transformation of the temple into a city. Cesare Letta44 suggests that the event should be dated from the time of the work carried out by Claudius. On this basis, he attributes inscription CIL IX, 3894 to Alba Fucens. This mentions a IIIIuir on the site of the sanctuary, thus removing a serious objection to its reconstruction. In fact, according to an observation made by Umberto Laffi,45 no municipium created with certainty after 49  was managed by the IIIIuiri. This magistrate cannot therefore have belonged to the municipium of Lucus Angitiae if the municipium was in the reign of Claudius. However, this line of thinking strikes me as questionable, for one has to ask whether a magistrate from Alba Fucens would hold his title in another city without indicating the place in which it had its function. There is some doubt here, then, and if we go along with Laffi we have to conclude that the quattuorvirate of this figure may also indicate that the lucus Angitiae was turned into a municipium before the middle of the first century , and once more

40 Champeaux 1982: vol. 1, 190, 454. As Champeaux notes (p. 190, n. 220), Fortuna was a typical Romano-Latial deity. The evidence of his cult, outside Latium, corresponds to the map of the Latin colonies and the roads connecting them. 41 42 Strabo 5. 4. 2. AE 1991, 567. 43 Letta and D’Amato 1975: no. 172; AE 1975, 369. 44 45 Letta 1972. Laffi 1973: 48.

320

Rome and the Great Places of Worship in Italy

no doubt by Sulla. This hypothesis is supported by context and there is no reason a priori to oppose it, especially since Alba Fucens and Marruuium were governed by quattuorvirs. The measure was in any case confirmed under the Empire, and Lucus Angitiae provides a good example of a city-sanctuary, endowed with grounds and free from the control of neighbouring cities. In spite of gaps in the documentation, these examples show how Octavian recovered for Rome—and himself, of course—these Italic places of remembrance in their full importance, demonstrating at the same time his profound piety, in comparison to the impious, who had dared despoil or neglect these sacred sites. In the Res Gestae,46 he likewise insists that the sacred possessions confiscated by Mark Antony from the temples of all cities in the province of Asia be returned. The Roman interventions in Capppadocia and Galatia47 provide further evidence of these transformations. The four great temples inhabited by hierodules and governed by priests were here turned into city-states with annually elected priesthoods, held by the local Hellenized aristocracy. All these initiatives were part of measures which Augustus’s policy of religious restoration extended across Italy and the Roman world. Another interesting aspect of this process of recovery for the great sanctuaries of Italy is the expansion of the notion of Roman public territory beyond the limits of the third century . We know that specifically Roman cults were celebrated at Ostia Antica, and that from the middle of the fourth century cults and Roman public temples were situated in Latium, quite a long way from Rome. The end of the Social War and the concession of Roman citizenship to all Italics had a profound impact on the status of a number of Italic or Roman places of worship. For this reason, it is worth looking at a text by Tacitus48 concerning a temple in Antium and a votive dedication made by the Roman knights. The equestrian order had made a vow for the health of the empress. An ancient tradition connected the knights to Fortuna equestris, a public deity who in 173  had received a temple near the Theatre of Pompey. This temple, though, no longer existed. A vow was a contract that could not be altered, even on points of detail. It had to be carried out just as it was, which meant finding a public deity who bore this name. Then again, a vow formulated by the equestrian order did not automatically apply in a place other than Rome or Roman territory: it concerned a deity of Rome who possessed an officially established temple, and so a public

46 48

47 RGDA 24. 1. Beard et al. 1998: vol. 1, 341. Tacitus, Annals 3. 71. 1–2.

Rome and the Great Places of Worship in Italy

321

deity of Rome. Noting the problem, the senate deliberated on the question of whether this votive obligation could apply to the equestrian Fortuna of Antium. The answer came in the affirmative, for since the Social Wars, all of Italy—that is to say the municipia and the colonies of Italy—were situated on the ager Romanus and were peopled by Roman citizens. They were Roman cities. As a result, a deity established in one of these municipia or colonies was both a Roman and local deity. There was therefore no need to request authorization from the Antiates in order to make this offering, and the dedication was legitimate according to Roman sacred law. In one sense, all these initiatives reveal an old tradition on the part of the Romans of taking over the Italic places of remembrance, and the full use made of this tradition by Octavian/Augustus at the start of the principate. Nonetheless, the decisions taken by Augustus go beyond simply wanting to have control over symbolic and potentially dangerous sites. Augustus took advantage of the state of dilapidation in which many of these places of worship lay as a result of the acts of violence and other disturbances which had affected Rome, Italy, and the Roman world between the Social War and the end of the Civil Wars, in order to give a spectacular demonstration of his piety and of the restoration of the traditions of old. Behind all these initiatives we can clearly see a new vision of Roman space, as the religious restorations were carried out in Rome, Italy, and the provinces alike. But it suffices to recall the debate aroused by the vow of the Roman knights, or later by the expansion of the pomerium by Emperor Claudius,49 to appreciate that one and a half centuries after the end of the Social War, the Romans had not yet found a means of representation for the new space that centuries of conquest had brought to Rome.

B I B L I OG R A P H Y Amann, P. (2002). ‘Das konstantinische “Reskript von Hispellum” (CIL XI, 5265) und seine Aussagekraft für die etrusko-umbrischen Beziehungen’. Tyche 17: 1–28. Beard, M., J. North, and S. Price (1998). Religions of Rome. Vol. 1: A History. Cambridge. Champeaux, J. (1982). Fortuna. Recherches sur le culte de la Fortune à Rome et dans le monde romain des origines à la mort de César. I. Fortuna dans la religion archaïque. Rome.

49

See Giardina 1995: 123–40.

322

Rome and the Great Places of Worship in Italy

Coarelli, F. (2001). ‘Il rescritto di Spello e il santuario “etnico” degli Umbri’, in Umbria cristiana. Dall a diffusione del culto al culto dei santi (secc. IV–X). Atti del XV Congresso internazionale di studi sull’alto medioevo (Spoleto, 2000). Spoleto, 39–52. Flower, H. I. (2000). ‘The tradition of the spolia opima: M. Claudius Marcellus and Augustus’. Cl. Ant. 19: 34–64 (esp. 46 n. 75). Gascou, J. (1967). ‘Le rescrit d’Hispellum’. MEFRA 79: 609–59. Giardina, A. (1995). ‘Seneca, Claudio e il pomerio’, in Alla Signorina. Mélanges Noëlle de la Blanchardière. Rome, 123–40. Granino Cecere, M. G. and J. Scheid (1999). ‘Les sacerdoces equestres’, in S. Demougin and Th. Raepsaet-Charlier (eds.), Ordo Equester. Histoire d’une aristocratie. CÉFR 257. Rome. Heurgon, J. (1942). Recherches sur l’histoire, la religion et la civilisation de Capoue préromaine des origines à la deuxième guerre punique. Paris. (1970, new edn.) Keppie, L. (1983). Colonisation and Veteran Settlement in Italy 47 .. – .. 14. Rome. Laffi, U. (1973). ‘Sull’organizzazione amministrativa dell’Italia dopo la guerra sociale’, in Akten des VI. Internationales Kongresses für griechische und lateinische Epigraphik. Munich. Lejeune, M., (1999). Méfitis d’après les dédicaces lucaniennes de Rossano di Vaglio. Leuven. Letta, C. (1972). I Marsi e il Fucino nell’antichità. Milan. Letta, C. and S. D’Amato (1975). Epigrafia della regione dei Marsi. Centro studi e documentazione sull’Italia romana 7. Milan. Manconi, D., V. Cruciani, and P. Camerieri (1997). ‘Hispellum. Pianificazione urbana e territoriale’, in Assisi e gli Umbri nell’antichità. Atti del convegno internazionale, Assise (18–21 December 1991). Assisi, 375–429. Pietrangeli, C. (1939). Spoletium (Spoleto). Italia romana: municipi e colonie. Rome. Pobjoy, M. (1997). ‘A new reading of the Mosaic Inscription in the temple of Diana Tifatina’. PBSR 65: 59–88. Poccetti, P. (1979). Nuovi documenti italici: a complemento del Manuale di E. Vetter. Orientamenti linguistici 8. Pisa. Scheid, J. (1996). ‘Pline le jeune et les sanctuaries d’Italie. Observations sur les Lettres IV, 1, VIII, 8 et IX, 39’, in A. Chastagnol, S. Demougin, and C. Lepelley (eds.), Splendidissima civitas. Études d’histoire romaine en hommage à François Jacques. Paris, 246ff. Schilling, R. (1954). La religion romaine de Vénus. Rome. Schilling, R. (1982). La religion romaine de Vénus. Paris, 239–64. (deuxième édition). Sensi, L. (1998). ‘Sul luogo del ritrovamento del rescritto constantiniano di Spello’, in Atti dell’Accademia romanistica constantiniana. XII Convegno interènational in onore di Manlio Sargenti. Naples, 457–77. Sensi, L. (1999). ‘In margine al rescritto constantiniano di Hispellum’, Annali della Fondazione per il Museo ‘Claudio Faina’, VI. Rome, 365–73. Wissowa, G. (19122). Religion und Kultus der Römer. Munich.

V Late Republican Transformations

13 Cicero and Divination The Formation of a Latin Discourse Mary Beard

This article is intended to be read in association with that of Schofield (1986: 47–65), which followed it in the original publication. The two articles share a common outlook—for we both believe that an understanding of the literary form of De Divinatione is integral to an understanding of its philosophical and historical point. But in detail our approaches are rather different. My own paper is the work of an historian and is concerned principally with the intellectual and cultural context of De Divinatione. My analysis of the text, highlighting its tensions and unresolved contradictions, follows from my analysis of that broader context. Schofield, by contrast, studies De Divinatione as an example of Hellenistic philosophical argumentation and explores the ways Cicero translates this not merely into Latin, but into a specifically Roman rhetorical mode. Other differences—in particular some disagreement as to how far it is possible to identify a ‘Ciceronian position’ on religion—are signalled in the text and notes of what follows. Both papers were originally given in a series of seminars on Cicero’s De Divinatione which we organized together at the Institute of Classical Studies, London. The other contributors to this series were Nicholas Denyer (with a paper now published in PCPhS 1985), and Elizabeth Rawson, who presented material now published in her book, Intellectual Life in the Late Roman Republic. The Roman élite in the last century  were sceptical about divination, augury, prodigies, and haruspicy; or, at least, that has been the view of

Mary Beard, Cicero and Divination: The Formation of a Latin Discourse In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0014

326

Cicero and Divination

most modern scholars.1 One text repeatedly taken as evidence of this scepticism is Cicero’s dialogue On Divination, a work in two books—the first, put into the mouth of Quintus Cicero, formulating along Stoic lines the arguments in favour of divination; the second, in the mouth of Marcus himself, attacking those arguments and undermining the principles of divinatory science.2 The spirited tone of this second book has often captured the attention (and admiration) of modern rationalist scholars. Marcus ridicules, with all his rhetorical skill, the supposed examples of divination’s success: he dismisses celebrated prodigies— rivers flowing with blood or statues dripping with sweat—as physically impossible;3 he writes off so-called ‘prophetic’ dreams as merely reflections of man’s daytime preoccupations, not as true warnings from the gods;4 he pours scorn on such famous portents as cocks crowing in forewarning of military victory—‘you talk, (Quintus), as if a fish and not a cock had done the crowing! But come; is there any time day or night, when they are not liable to crow?’5 At first sight, it seems that Cicero is expressing here his own personal scepticism on such religious practices: everywhere ‘rationalism’ seems to win the day; divination is to be continued only for reasons of political expediency.6 Other works of Cicero appear to contradict their author’s scepticism on the validity of divination. In De Legibus, for example, he affirms his support for traditional Roman religious practices (including augury);7 and his letters and speeches include many explicit references to the tenets of state religion.8 Yet this apparent contradiction has not, even when recognized, stimulated any serious challenge to the view that Cicero himself was a sceptic on divination and other matters of religion. The problem has been side-stepped quite simply by privileging the

1 The classic formulation of this view is that of Taylor 1949: 76–97 and it remains the dominant view. For more recent expressions, see Dumézil 1970: 549–50 and Momigliano 1984b: 873–92. For Malcolm Schofield’s companion piece to the present article, see Schofield 1986: 47–65. 2 Throughout this article ‘Marcus’ (like ‘Quintus’) refers to the character in De Divinatione, ‘Cicero’ to the author of the dialogue. 3 4 De Divinatione 2. 27. 58. De Divinatione 2. 68. 140. 5 De Divinatione 2. 26. 56. 6 For a clear and balanced account of the orthodox position on De Divinatione and the other theological works, see Rawson 1975: 241–5. The bibliography on the De Divinatione is now vast; the important article of Linderski (1982: 12–38) contains a useful collection of references, as does also Troiani 1984: 920–59. The most recent treatment, by Denyer (1985: 1–10), takes a different line from most; he argues convincingly for the philosophical inadequacy of Marcus’ demolition of the Stoic case. 7 De Legibus 2. 13. 32–3. 8 For example, Cicero, Ad Atticum 16. 6; Catiline 2. 13. 29; Pro Sulla 14. 40.

Cicero and Divination

327

philosophical writing (especially that of the last period of Cicero’s life) above all other areas of the Ciceronian corpus. Either, it is commonly suggested, Cicero’s sceptical philosophy reflects his ‘true opinions’, in contrast to the insincere or merely conventional appeals to religion in his public speeches.9 Or, more cautiously, it is argued that De Natura Deorum (of 46 ) and De Divinatione (written in 45–44 ) as a supplement to the earlier work) represent a change in Cicero’s considered religious outlook since his earlier writing, at least since the writing of De Legibus in the late 50s. Linderski and Momigliano have, for example, recently suggested that this shift from faith to scepticism may be ascribed not so much (as has previously been argued) to Cicero’s shock at the death of his daughter Tullia, but rather to his disillusionment with state religion, whose forms he saw increasingly exploited by the Caesarian faction.10 In each case, whether a shift of opinion is proposed or the late philosophy regarded as more truly ‘Ciceronian’ than any other work, the genuine scepticism of Cicero himself, at the time of his writing De Divinatione, is hardly called into question. There are several objections to the usual deduction that De Divinatione, and particularly its second book, is an expression of Cicero’s disbelief in the practice of divination. Some of these objections are general, or explicitly theoretical. It seems doubtful, for example, that ‘belief ’ and ‘disbelief ’—with their suggestion of the personal commitment characteristic of modern world religions—are appropriate terms for the analysis of traditional Roman religion. Likewise it seems highly controversial whether the ‘real views’ of Cicero could ever be traced off from the views of the character of Marcus in the dialogue. For much recent literary theory has called into question the notion of any easy relationship between the author and his text. This is not the place to discuss such problems in detail, though an awareness of these and related issues underlies much of what follows in this chapter.11 A more particular objection to the standard view stems from the structure of the work as a whole and the relationship between the second and first books. Cicero did not write the second book of De Divinatione in isolation, as a partisan tract against Roman traditions of divination. The second book is balanced by the first: the arguments against divination must be seen alongside the earlier arguments in favour of the practice. Both positions are laid out, and no conclusion, supporting 9

For example, Latte 1960: 285; Le Gall 1975: 143–4. Momigliano 1984a: 199–211; Linderski 1982. 11 Difficulties with the application of the concept ‘belief ’ are fully discussed by Needham 1972. For a clear introduction to the debates on the status of the author, see Belsey 1980. 10

328

Cicero and Divination

one side or the other, is offered; instead, in the traditions of the Academic school of philosophy, the reader is left to make up his own mind on the most convincing case: ‘It is characteristic of the Academy [says Marcus Cicero, at the very end of the work] to put forward no conclusions of its own, to approve those which seem most like the truth, to compare arguments, to draw forth all that may be said on behalf of any opinion, and without asserting its own authority to leave the judgement of those listening entirely free. We shall hold to this method, inherited from Socrates, and if it is agreeable to you, my dear brother Quintus, we shall follow it as often as possible in our future discussions.’ ‘Nothing could please me better’, Quintus replied. When this was said, we arose. ‘Cum autem proprium sit Academiae iudicium suum nullum interponere, ea probare quae simillima veri videantur, conferre causas, et quod in quamque sententiam dici possit expromere, nulla adhibita sua auctoritate iudicium audientium relinquere integrum ac liberum. Tenebimus hanc consuetudinem, a Socrate traditam, eaque inter nos, si tibi, Quinte frater, placebit, quam saepissime utemur.’ ‘Mihi vero’, inquit ille, ‘nihil potest esse iucundius’. Quae cum essent dicta, surreximus.12

Whatever the force of the individual arguments that have gone before, these final words in the dialogue explicitly suspend judgement.13 Those who deduce Cicero’s personal scepticism from the second book of De Divinatione ignore this clear denial of a directed conclusion and neglect to treat the dialogue as a whole, as a balance of arguments for and against divination. No weight is added to the arguments against divination merely because they are spoken in the dialogue by the character of Marcus himself. Although it may be tempting (whatever the theoretical problems) to equate the words of a writer apparently speaking in propria persona with the ‘correct meaning’ of his work, that temptation here proves elusive; the ‘authorial voice’ in De Divinatione and the related De Natura Deorum constantly evades definition. Indeed. Cicero himself in the introduction to the latter dialogue specifically states his opposition to

12 De Divinatione. 2. 72. 150. This is reiterated at De Fato 1. 1, with Cicero’s explicit statement that in De Natura Deorum and De Divinatione he laid out both sides of the question at issue ‘so that each (reader) might more easily adopt the view that seemed to him the most probable’ (quo facilius id a quoque probaretur quod cuique maxime probabile videretur). 13 I lay great stress on the fact that these words are (literally) the conclusion of the dialogue—in contrast to Schofield 1986, who locates the ‘authorial conclusion’ in the immediately preceding chapters and their denunciation of ‘superstition’.

Cicero and Divination

329

attempts to identify his own personal opinions from his philosophical writing: Those, however, who seek to learn my personal opinion on various questions show an unreasonable degree of curiosity. Qui autem requirunt quid quaque de re ipsi sentiamus curiosius id faciunt quam necesse est.14

This statement encourages me to go further than Schofield in my rejection of any notion of a ‘Ciceronian viewpoint’ emerging from De Divinatione; but it is not my only argument. The lack of a clear authorial voice is yet more forcefully demonstrated by the apparent shift in the opinions of the character of Marcus between De Natura Deorum and De Divinatione. In the earlier work Marcus takes only a small role, largely as narrator and listener; but at the end of the dialogue he supports the Stoic views of the nature of the gods that had been put forward by the character of Balbus: ‘I felt that (the discourse) of Balbus was a closer approximation to the truth’ (ita discessimus ut . . . mihi Balbi (disputatio) ad veritatis similitudinem videretur esse propensior).15 This contrasts markedly with Marcus’ lengthy attack on the Stoic views of divination in De Divinatione. In part this contrast may be explained by reference to the traditions of Academic philosophy: the Academic philosopher (as Cicero claimed to be) would count himself free to espouse some individual doctrines of rival schools, while rejecting others; he might think it reasonable to accept, say, the Stoic views of the general character of the divine, while at the same time objecting to Stoic theories of divination. But there is a significance here beyond simple adherence to Academic traditions: the contrast in the views of Marcus between De Natura Deorum and De Divinatione also warns the reader against any attempt to trace an authorial voice from the various statements of the ‘author-as-character’ in the philosophical dialogues. It is easy enough to cast doubt on the usual reading of De Divinatione and the standard assumption that Cicero expressed in that work his own scepticism on the validity of divination. Yet the arguments put forward so far raise as many problems as they solve. If Cicero’s dialogue on divination offers the reader no directed conclusion, how are we to interpret this suspension of judgement? In many other of his philosophical treatises—De Senectute, for example, or De Officiis—the reader seems to be left in no doubt on the tenor and message of the work; why is De Divinatione different in this respect? How are we to interpret the strategies

14

De Natura Deorum 1. 5. 10.

15

De Natura Deorum 3. 40. 95.

330

Cicero and Divination

which in this and the related theological works so effectively efface the authorial voice? A broader approach is needed. I shall attempt in this paper to throw light on the peculiar characteristics of De Divinatione by investigating the wider cultural and intellectual context within which the treatise was written. The first main section of the paper considers the history of philosophical inquiry in Rome before Cicero and shows how Cicero himself was an important innovator in attempting for the first time an active integration between Greek philosophy and traditional Roman practice and thought. The second section discusses the difficulties that necessarily stand in the way of such a cultural integration between two different systems of thought; it suggests that some of the tensions, problems, and evasions characteristic of Cicero’s theological works are a direct consequence of the particular difficulties of integrating the traditions of Roman state religion with a Hellenizing, ‘scientific’ approach. The final section draws on the earlier conclusions to argue that the importance of De Divinatione for the historian of religion lies not in the evidence it provides for the supposed scepticism of the Roman élite in the late Republic, but in its position as a specifically religious treatise; for as such it represents an important stage of cultural development at Rome—the definition of ‘religion’, for the first time, as an independent subject of discourse. The arguments I shall present, unlike those of Schofield, are not philosophical in the technical sense, but they have necessarily involved me in certain decisions on method which may seem important, if not contentious, to a specialist in Hellenistic and Roman philosophy. Two of these should be noted at this point: (a) In setting De Divinatione in its literary context, I have not considered in any detail the fragmentary works of Ciceronian philosophy. Much of my argument centres on the structure of Cicero’s dialogues and treatises; hence any great stress on works surviving only in fragments, where the overall structure is lost, could well be misleading. The most notable omission that results from this principle is De Fato. This work was written, like De Divinatione, as a supplement to De Natura Deorum—but it survives only in some twenty pages, probably less than a quarter of the whole.16 16 For the place of the De Fato in the programme of theological works, see De Divinatione 2. 1. 3. I stand by this justification for leaving the fragmentary treatise out of my consideration; but I cannot help but be struck by Schofield’s observation that De Fato is a ‘philosopher’s work’, constantly avoided (for one reason or another) by historians and literary critics. See below, p. 000.

Cicero and Divination

331

(b) I have not concerned myself with the Greek sources of Cicero’s philosophy. While I recognize that the treatises and dialogues do not constitute ‘original’ philosophical thought, in the modern sense of the word, I assume that Cicero exercised some choice in drawing on the Greek authorities available to him. That act of choice, in my view, makes De Divinatione and the related works understandable within a first-century context, whatever the ‘source’ of the ideas contained within them.17 These preliminary decisions, while avoiding some of the traditional problems of Ciceronian philosophy, have enabled me to proceed with a broader approach than is usually adopted in this subject. In this way, I believe, Cicero’s De Divinatione (and particularly its second book) can be set in a new and more helpful context than hitherto.

PHILOSOPHY AT ROME: THE PLACE OF CICERO Rome in the late Republic was the site of much philosophical activity. Elizabeth Rawson’s recent study of intellectual life in Ciceronian Italy has amply documented the wide range of philosophical writing of the period that survives in fragments or passing references in later writers.18 This section offers a brief review of the earlier history of Roman philosophical activity, in order to point up the new and distinctive features of the surviving treatises of Cicero: although Cicero was by no means the first Roman to study and write philosophy, he was the first (or at least among the first) fully to integrate Hellenizing philosophy with traditional Roman practice. Cicero was, in short, an innovative figure in the development of philosophy at Rome.

Philosophy before Cicero Philosophical enquiry was known at Rome long before the mid-first century  As early as the fourth century  there was, according to Pliny, a statue of Pythagoras in the Roman comitium—an indication of

17 Although the search for Cicero’s sources is no longer at its height, the simple fact that Cicero is the source for so much of Hellenistic philosophy, necessarily focuses the interests of philosophers rather differently than those of historians. See, for example, recently Schäublin 1985. 18 See Rawson 1985: esp. 282–97.

332

Cicero and Divination

some contact between Rome and one of the major figures of Greek philosophy, though probably not of any significant degree of Roman philosophical activity in the strictest sense.19 From the second century, however, there is considerable evidence of active philosophical interest among the Roman élite. It is unnecessary here to detail this fully, but the main strands of evidence can usefully be summarized. The earliest surviving Latin literature often shows clear influence of the Greek philosophical tradition and contains precise references to Greek philosophers and schools. Ennius, for example, produced translations or adaptations of Greek philosophical works, including Euhemerus’ tract on the human origins of the gods.20 More striking still, the popular, comic, dramatic works of the late third and early second century  display an obvious acquaintance with Hellenizing philosophy. In part this may be ascribed to the Greek origin of Roman comedies; but only in part. Both Plautus and Terence not only refer to philosophers (the former sometimes to individual schools),21 but also on occasion assume some common, if stereotyped, perception on the part of their audience of the nature of philosophical activity. So, for example, in Plautus’ Captivi a pretentiously clever response to an apparently simple question (‘Is his father alive?’) brings forth the ironic comment: All is well now; the man’s not only lying, he’s philosophizing too. Salva res est, philosophatur quoque iam, non mendax modo est.22

There is also ample documentation of the presence of Greek philosophers at Rome. Even before the famous sojourn of Panaetius in the second half of the second century some philosophers came to Rome, principally as ambassadors, but also engaging in teaching during their stay. In 155 , for example, Critolaus, Carneades, and Diogenes, who came to petition the Romans to remit a fine imposed upon the Athenians, also gave lectures to a (reputedly) large audience;23 and in 169 the unfortunate Crates of Mallos, head of the Pergamene library, Stoic, and ambassador of Attalus, used his enforced convalescence in Rome—he 19 For the statue, see Pliny, Natural History 34. 6. 26. In my view, Jocelyn (1976: 323–66) seriously overestimates the degree of active Roman philosophical interest indicated by this statue. 20 Ennius, Varia 45–146 (Vahlen); and for a convenient compilation of other references to Greek philosophy in Roman Republican writing, see Garbarino 1973. 21 See, for example, Plautus, Pseudolus 465 (Socrates); Rudens 1003 (Thales); Persa 123 (Cynics); Terence, Andria 55–9; Eunuchus 262–4. 22 Plautus, Captivi 282–4. 23 See, for example, Cicero, De Oratore 2. 37. 155; Aulus Gellius, Noctes Atticae 6.14. 8–10; Plutarch, Cato Maior 22. 1–5.

Cicero and Divination

333

had broken his leg falling down a sewer on the Palatine—for teaching (presumably Stoic) philosophy.24 Others no doubt came to Rome for more specifically philosophical purposes. Among these perhaps were the two Epicureans present in the city to be expelled in 173 (or 154),25 and the philosophers and rhetoricians again victims of expulsion in 161.26 Members of the Roman élite formed personal contacts with these Greek scholars. So, for example, in the second half of the second century Scipio became closely associated with Panaetius and, perhaps a little earlier, there is evidence of connections between L. Marcius Censorinus and the Academic philosopher Clitomachus.27 More frequently, though, Romans met Greek philosophy and philosophers in Greece itself, away from Italian soil. Aemilius Paullus came into contact with philosophers at Athens at the time of the campaign of Pydna,28 and during the late second and early first centuries several Romans are attested either to have studied philosophy there or to have taken the opportunity to attend lectures while in Athens for other reasons. Amongst these were not only Cicero himself, but also L. Licinius Crassus (cos. 95), and M. Antonius (cos. 99)—both of whom took advantage of service in the East to visit Athens and listen to the leading philosophers of the day.29 By the end of the second century a number of Romans were considered philosophical experts in their own right. Among early Roman Stoics, Spurius Mummius, who served as ambassador in the East in 140/139, was described by Cicero as ‘learned in the doctrines of the Stoics’ (doctus ex disciplina Stoicorum)30 and, with even greater stress, P. Rutilius Rufus (cos. 105 ..) was regarded as a man ‘devoted to philosophy’ (philosophiae deditus) and ‘widely read in Greek literature, a pupil of Panaetius, virtually perfect in Stoic doctrines’ (Graecis litteris eruditus, Panaeti auditor, prope perfectus in Stoicis).31 More widely popular, however, and probably earlier in its vogue was Epicureanism. Cicero again discusses the works of one C. Amafinius, the earliest attested Latin philosophical 24

Suetonius, De Grammaticis et Rhetoribus 2. 1–2. Athenaeus 12. 68. 547a; Aelian, Varia Historia 9. 12; Suidas, s.v. Epikouros, 2405. The consul of the year is recorded only as L. Postumius, whence the confusion of dates. See further Garbarino 1973: 374–9. 26 Suetonius, De Grammaticis et Rhetoribus 25. 1. 27 For the relationship of Panaetius and Scipio, see Astin: 1967, 294–306. For Censorinus and Clitomachus, Cicero, Lucullus 32, 102. 28 See, for example, Pliny, Natural History 35. 11. 135. 29 De Oratore 1. 11. 45–7 (Crassus); 1. 18. 82 (Antonius). And in general, see Rawson 1985: 6–7. 30 Brutus 25. 94. 31 De Oratore 1. 53. 227; Brutus 30. 114. Rutilius Rufus was, it seems, sufficiently expert in philosophy to be cited by Posidonius, in Cicero, De Officiis 3. 2. 10. 25

334

Cicero and Divination

treatises. Cicero is disparaging but admits that Amafinius’ writing drew many adherents and imitators.32 One such adherent was perhaps T. Albucius, praetor towards the end of the second century, who is described by Cicero in the following terms: Titus Albucius was learned in all things Greek, or rather almost completely Greek . . . he spent his youth at Athens and turned out a perfect Epicurean. . . . Doctus etiam Graecis T. Albucius vel potius paene Graecus . . . fuit autem Athenis adulescens, perfectus Epicurius evaserat. . . . 33

We must imagine with these early Roman philosophers a small group of men prepared to exploit fully both the opportunities for philosophical study in Greece (as T. Albucius), and also, no doubt, the increasing flow of philosophical writing into Italy and the resources of the well-stocked libraries of the Roman upper class.34

Cicero’s Innovation It is tempting to regard Ciceronian philosophy as the direct continuation of this earlier philosophical activity, representing no particular new or problematic development. In part the works of Cicero themselves give this impression; for not only does Cicero’s own survey of the history of Roman philosophy in the fourth Tusculan Disputation relate his own philosophical achievements closely to those of his predecessors,35 but the choice of characters and the dramatic setting of several of the dialogues in the second-century ‘Scipionic Circle’ may suggest to the reader that the arguments of Ciceronian philosophy are appropriately put into the mouths of Scipio and his contemporaries. This is illusory; the contribution of Cicero himself to the history of Roman philosophy was far more innovative than this would suggest. Cicero for the first time Romanized Greek philosophy, tackling Roman problems, with Roman exempla, in a Roman setting. For his predecessors, by contrast, philosophy had remained essentially Greek, even if practised by Romans. The first argument in justification of this distinction is a negative one. There is no reason to believe that the arguments put by Cicero into the mouths of the so-called ‘Scipionic Circle’ accurately reflect the type of

32 Tusculanae Disputationes 4. 3. 6–7. No precise date is given for Amafinius, but the implication is that these Epicureans had chronological priority over the early Stoics. For Amafinius’ imitators, see Academica Posteriora 1. 2. 5; Ad Familiares 15. 16. 1; 15. 19. 1. 33 34 Brutus 35. 131. For libraries in Italy, see Rawson 1985: 39–42. 35 Tusculanae Disputationes 4. 1. 1–3, 7.

Cicero and Divination

335

philosophical debate in second-century  Rome. Cicero no doubt aimed at historical accuracy up to a point; there are, for example, no glaring anachronisms in the historical allusions incorporated into the dialogues, and the characters assembled for each work represent plausible groups of contemporary interlocutors. But in terms of the philosophical argument of the dialogues, the second-century characters are principally convenient vehicles for essentially first-century debate.36 This is clearly indicated on one occasion by Cicero’s uncertainty, quite late in the process of composition, as to the identity of his interlocutors. Writing to Quintus in 54 , on the progress of his De Republica, he reports on his original scheme to set the work in 129  in the company of Scipio, Laelius, and their associates—characters, he argues, who gave weight to the speeches because of their high rank. Yet, he continues, when he had parts of the dialogue recited to one of his friends, it was suggested to him that it would be more appropriate if he himself were to take the part of the main speaker. So Cicero changed the speaking characters of the dialogue— temporarily, as we now know: Marcus Cicero himself and Quintus took the place of Scipio and Laelius.37 This is clear evidence that those characters were part of the literary scheme of the dialogue, not in any sense external voices reported by Cicero. The second argument concerns the character of the earliest phases of Roman philosophy. There is no evidence that any philosophical writing at Rome before the age of Cicero went beyond the exposition of the major tenets of the Greek philosophical schools or beyond translations of Greek treatises. This point must be expressed negatively and, to some extent, tentatively, for the simple reason that only the scantiest fragments of early Roman philosophy survive. But such material as is preserved stands firmly against any suggestion that earlier writers had, as Cicero, constructively used Greek philosophy to deal with Roman problems. The theoretical works of Ennius, for example, were clearly translations of Greek treatises; while the writing of Amafinius and his followers seems (as far as can be gathered from the brief testimony of Cicero) to have offered little more than didactic accounts of Epicureanism: everything you ever wanted to know about Epicurus—and never dared to ask.38

36

In this I broadly follow the arguments of Astin 1967: 9–10. Ad Quintum Fratrem 3. 5 (5–7). 1–2, with Garbarino 1973: 18–20. 38 See above, n. 32. An apparent exception to this would be the second-century Q. Mucius Scaevola (whose views on the tripartite division of the divine are quoted by Augustine, City of God 4. 27). I have, however, been convinced by the views of Cardauns (1960) that Augustine is in fact quoting not from any philosophical work of Scaevola, but from the words of the character of Scaevola in a Varronian dialogue. 37

336

Cicero and Divination

This suggested contrast between Cicero’s philosophy and the work of his predecessors may at first sight appear at odds with the impression given by Cicero in the Tusculans and elsewhere that his own philosophical activity had direct precedents in earlier Roman history. But here one should bear in mind Cicero’s defensive stance in most of the philosophical works. He is concerned to justify his own activity in philosophy and its suitability for a Roman statesman by tracing back its roots into earlier Rome and by claiming it, almost, as a traditional activity for the Roman élite. This is clearly a procedural tactic and one that tends to elide any differences between Cicero and his predecessors. It would be unwarranted to assume that the traditional colours in which Cicero from time to time paints his work necessarily indicate that it was itself a traditional part of Roman intellectual life.39 On occasion Cicero’s treatment of the work of his predecessors implicitly defines it as Greek, despite his overall attempt to assimilate their work to his own. This emerges most clearly from Cicero’s description of the Epicurean T. Albucius. It is striking that the man is designated not merely ‘learned in all things Greek’ (doctus Graecis), but also ‘almost completely Greek’ (potius paene Graecus).40 It is as if a passionate interest in Hellenizing philosophy served to dissociate the philosopher from traditional Roman culture: he became de facto Greek. No one could speak of Cicero in these terms. His brand of philosophy, while including from time to time exposition of Greek theory, is distinctive for its integration of Greek philosophy with Roman practice—with Roman political institutions in De Republica, with traditional Roman moral values in De Officiis and with Roman divinatory practice in De Divinatione. It must remain uncertain how far the work of Cicero stands apart from that of his direct contemporaries. I have treated Cicero himself as the philosophical innovator in the late Republic—largely because there is no firm evidence of other writing of precisely his kind. Lucretius’ brilliant exposition of Epicurean philosophy hardly engages deeply with Roman practice; while the surviving parts of Varro’s Hellenizing systematization of Roman traditions leave it unclear to what extent he had undertaken a constructive integration of the two systems. We can, however, do little more than guess at the character of the rest of Varro’s output or the other

39 As Rawson (1985: 57) notes, this defensiveness on Cicero’s part should warn us against overestimating the amount of early Roman philosophical activity on the basis of those early Roman philosophers he cites. It is likely that, in self-justification, Cicero is not selecting, but parading all such characters that he can find. 40 Brutus 35. 131. For Athens as the obvious location of philosophical debate, see De Oratore 1. 11. 45–7; 1. 18. 82.

Cicero and Divination

337

first-century philosophy so ably resurrected by Rawson. It may be that ‘Cicero’ should stand as shorthand for his whole generation.41

THE PROBLEMS OF CULTURAL INTEGRATION: CICERO’S SUSPENSION OF JUDGEMENT The development of philosophical or ‘scientific’ thought is a wellrecognized problem in cultural history and anthropology. Scholars in many different fields have investigated the complex stages by which individual societies, either through internal change or outside influence, adapt their traditional ‘symbolic’ inheritance to a ‘scientific’ world view. Anthropologists, in particular, have highlighted the problems associated with the introduction of ‘scientific’ thought into an essentially ‘prescientific’ society.42 Students of the ancient world also have recognized many of the same issues in the origins and early history of Greek philosophy. They have long appreciated that the rise of pre-Socratic philosophy was embedded in a complex set of relations between the mythical, symbolic inheritance of the early philosophers and their increasingly scientific modes of thought;43 and recent work on such developed scientific studies as Aristotle’s Historia Animalium has shown that there are still latent conflicts between traditional systems of classification and a strictly scientific approach.44 Curiously, this method of approach has never been applied to the growing ‘intellectualization’ of Roman life in the first century .

41 Lucretius was surprisingly (in modern eyes) un-influential in the history of Roman philosophy; he may almost be seen as the final flowering of the tradition of Epicurean writing started by Amafinius. See Rawson (1985: 285) and—for his Greek, rather than Roman, character—Boyancé 1963: 7–32. The assessment of Varro’s theological works is made particularly difficult by the fact that almost all the substantial fragments are preserved in the Christian polemic of Augustine. 42 Amongst the many modern studies of ‘pre-scientific’ thought and transition in modes of thinking, note especially: Goody 1977; Horton and Finnegan (eds.) 1973. The terminology in this area is far from standard. I have used the following pairs of opposites interchangeably: traditional/non-traditional; mythological/non-mythological; pre-scientific/scientific; symbolic/ encyclopaedic. I have used inverted commas wherever these technical terms might be confused with popular usage. 43 Gernet 1968: 405–14 (Eng. trans., 1981: 343–51); Lloyd 1979: 1–8; Vernant 1965: 285–314. Even the most traditional studies of the pre-Socratics discuss such issues; see, for example, Guthrie 1962: 58–62 (on Thales). 44 Lloyd 1983: 7–57.

338

Cicero and Divination

Cicero’s innovation in integrating Hellenizing systems of thought with traditional Roman practice and institutions can well be understood in similar anthropological terms. There are, of course, obvious differences between the philosophical developments of first-century  Rome and those of early Greece: Roman philosophy largely involved the adoption of external (Hellenizing) models of thought; the origins of Greek philosophy are conventionally seen as the consequence of internal revolution in ways of thinking. Yet the underlying cultural clash in each case was very similar. Cicero, like the pre-Socratic philosopher (or like the tribesman adapting Western medicine to his traditional knowledge of witch-doctoring), was necessarily faced with conflicts between his ‘pre-scientific’, traditional ways of understanding the world and the ‘scientific’ world view implicit in the Greek philosophical modes of thought he was attempting to deploy. These conflicts provide an important background to any understanding of the characteristics of Cicero’s philosophical output. The most acute problem in integrating traditional Roman practice with Hellenizing philosophy arose in those areas where Greek philosophical discourse clashed with the institutional framework of Roman politics and religion. Roman institutions were not Greek. They could not convincingly be interpreted by a straightforward application of Greek theory, especially where Roman traditional knowledge (whether implicit in ‘common sense’ or explicit in a literary tradition) offered its own independent, often competing, interpretation. Roman augury provides a clear example here. It was deeply embedded at the centre of Roman political and religious life, defined and regulated in the written tradition of the priestly books; it made sense according to the logic of the symbolic inheritance of Rome and Roman views of the operation of the gods in the world; its traditional interpretation could not simply be passed by in favour of a Greek philosophical scheme of divine involvement in the world.45 So, for example, as Cicero states in De Divinatione, Roman augury involved no predictive element, but was concerned rather simply to ascertain divine approval for the undertaking contemplated;46 by contrast, its Greek ‘equivalent’, mantike, whose philosophical aspect forms the basis of much of the dialogue, was defined by Cicero himself as ‘the foresight and foreknowledge of future events’.47 The conflict is immediately apparent. In the case of such central Roman institutions as augury, a Greek theoretical approach thus necessarily involved a 45 On the formal traditions of augury, see Linderski 1986. For the augural books, Regell 1878; Regell 1882. De augurum publicorum libris (1878); Fragmenta auguralia (1882). 46 47 De Divinatione 2. 33. 70. De Divinatione 1. 1. 1.

Cicero and Divination

339

complex process of active reinterpretation of the Roman inheritance within an overall Hellenizing model and also, no doubt, a rethinking of the theory itself in the light of Roman practice. Some areas of Roman life were more easily susceptible to a Greek philosophical interpretation. In some aspects of political philosophy, for example, the precedent of Polybius no doubt eased the integration between Roman practice and Hellenizing philosophy; but there were other particular areas where Roman cultural attitudes were either (contingently) relatively easy to reconcile with aspects of Greek theory, or where those attitudes were not closely defined within any institutional or written tradition. This was most obviously the case with ethics, oratorical principles, and some aspects of religious thought. Subjects such as the fate of the body and soul after death, for example, could more readily be treated in Greek philosophical terms than augury, simply because (unlike augury) they were not a central element of the institutional framework of Roman religion. Greek philosophy was not here liable constantly to conflict with the established ‘traditional’ knowledge of Rome. Yet even in these more tractable subjects the problems of integration between the two cultural systems were not negligible. At the most fundamental level of language, a simple exposition of Greek theory in Latin involved greater difficulties than simply forming a dictionary of equivalent terms; it involved, in part, a rethinking and reworking of such theory, so that it could make some sense in the Latin language and in the world view incorporated in that language.48 Cicero’s expertise in De Officiis, for example, at constructing an apparently close fit between Greek and Roman ethical outlooks should not mislead us into supposing that the two systems somehow ‘naturally’ overlapped.49 Cicero’s theological works can best be understood in the context of these problems of cultural integration. I shall highlight in the following pages some of the tensions, constraints, and evasions within, particularly, De Divinatione and De Natura Deorum, and I shall argue that these, as a whole, may be explained by reference to the underlying confrontation between traditional Roman symbolic knowledge of the workings of the world and the developed Hellenizing encyclopaedic rules for comprehending the same phenomena. The tenor of these arguments will be generalizing and structural in two particular respects. First, while 48 For a rare appreciation of the extent of the difficulty involved in translating Greek philosophy into Latin, see Douglas 1962: 41–51. 49 For a clear illustration of the fact that Greek and Roman ethical systems could be perceived as strikingly different, see Dionysius of Halicarnassus, Roman Antiquities 5. 8 (Greek distaste for the traditional severity of a Roman father).

340

Cicero and Divination

recognizing that certain features of tension that I identify can, individually, be explained by a variety of piecemeal explanations, my argument will focus on the whole structure of tensions and constraints in Cicero’s theological works. This is not to deny the validity of such piecemeal explanations as have been offered, but to suggest rather that a different type of explanation is appropriate when the individual features are considered as part of a broader pattern. Secondly, my primary interest lies in the tensions and contradictions between two systems of thought, rather than in the particular handling of these tensions by Cicero as author. In this respect my paper differs from that of Schofield: whereas he is concerned principally with the author’s handling of the difficult subject of divination, I focus on the more general context within which divination necessarily becomes a difficult subject. In the course of this discussion the Academic school of philosophy will sometimes figure prominently, the school to which, in the writing of the last phase of his life, Cicero frequently claimed adherence.50 Although the history of the Academy is complex and in places obscure, it is clear that it was known for most of antiquity as the leading school of scepticism, that it regarded firm knowledge as impossible to obtain, and that its traditional practice was to lay out both sides of any question under discussion without attempting to reach a firm conclusion. Cicero’s method was no doubt influenced by this philosophical school. But it is not enough to explain the characteristic features of uncertainty present in the theological works simply by reference back to the principles of the Academy; for that explanation itself raises the more pertinent and interesting question of why the principles of the Academy should prove an attractive theoretical stance in Cicero’s philosophical writing. Schofield suggests that the Academic philosophical style was peculiarly appropriate for educating the Roman reader in the subject of philosophy. I would add that behind Cicero’s Academic stance there were strong structural pressures against the formulation of firm opinions—namely, the problems necessarily encountered in any attempt to integrate Hellenizing and traditional Roman thought.

The Position of Priest and Philosopher One of the main characters in De Natura Deorum is both a priest of Roman state religion and a philosopher—a combination of roles that 50 See, for example, De Divinatione 1. 3. 6–4. 7 (‘nos’ referring to the Academics) and De Natura Deorum 1. 5. 11–12 (both passages from the authorial statement in the introduction, not from the dialogue proper).

Cicero and Divination

341

exemplifies the tension between traditional Roman religious practice and Greek philosophical thought. C. Aurelius Cotta, consul in 75  and pontifex from perhaps as early as 91  is depicted as host for the discussion on the nature of the gods and also acts as mouthpiece for the arguments of the sceptical Academic, attacking in particular the Stoic view of the gods put forward by the character Balbus. The fundamental contradiction of his role is plain: Cotta, the pontifex and representative of Roman state religion, takes a radically sceptical stance on the nature of the gods and, paradoxically, mounts a particular attack on the principles of Stoicism—the one ancient philosophical system that we now perceive to have been relatively easily compatible with the traditional symbolic inheritance of Roman religion.51 The ambiguities in Cotta’s role are made particularly prominent twice in the dialogue. The first occasion is at the beginning of the second book where Balbus, in trying to persuade Cotta to give some firm opinion on the nature of the divine, draws attention to his dual role: It is the mark of a philosopher and a pontiff and a Cotta to possess not a shifting and unsettled conception of the immortal gods, like the Academics, but a firm and definite one like our school [sc. the Stoics]. Est enim et philosophi et pontificis et Cottae de dis immortalibus habere non errantem et vagam ut Academici sed ut nostri stabilem certamque sententiam.52

Cotta refuses to comply with Balbus’ request but, despite his paraded role as representative of state religion, adopts an Academic stance that it is easier on such subjects to express negative views and disagreements than positive opinions. At the beginning of the third book the ambiguous role of Cotta is made explicit for a second time. Before embarking on an Academic deconstruction of the Stoic position put forward by Balbus, the character of Cotta is made to speak as pontifex and explicitly to accept the inherited traditions of Roman state religion, as handed down by his predecessors: For my part I shall always uphold [the ceremonies and duties of religion] and always have done so, and no eloquence of anybody, learned or unlearned, shall ever dislodge me from that belief on the worship of the immortal gods that I have inherited from our ancestors.

51 I would emphasize relatively here. Although we readily perceive apparent similarities between Stoic and traditional Roman views of the gods (the pervasiveness of the divine; divine benefits for man), even here, as I suggested above (p. 41 = 000), an active reinterpretation of both was required before they could convincingly be seen to overlap. 52 De Natura Deorum 1. 1. 2.

342

Cicero and Divination

Ego vero eas defendam semper semperque defendi, nec me ex ea opinione quam a maioribus accepi de cultu deorum immortalium ullius umquam oratio aut docti aut indocti movebit.53

This statement of explicit support for Roman state religion lies uneasily beside the sceptical arguments used by the character of Cotta in the rest of the book. A similar ambiguity is inherent in the character of Marcus Cicero in De Divinatione; for his own position as a member of the augural college is hardly compatible with his role in the dialogue as sceptic on the principles of divination in general and augury in particular. He attempts to negotiate this incongruity by separating his rational scepticism on the theoretical validity of divination from his practical commitment to its continuance at Rome for reasons of tradition and political stability: However, out of respect for the opinion of the masses and for its great service to the state, we maintain the augural practices, discipline, religious rites and laws, as well as the authority of the augural college. Retinetur autem et ad opinionem vulgi et ad magnas utilitates rei publicae mos, religio, disciplina, ius augurum, collegi auctoritas.54

But the strategy of appeal to expediency and tradition does not entirely remove the tension between Cicero’s prominent public role within traditional Roman religion and his denunciation of the principles of augury as a character in De Divinatione.55 Like the character of Cotta, Marcus in De Divinatione highlights the underlying problems in reconciling traditional Roman practice and Greek philosophical theory.

Dialogue Form and the Suspension of Judgement The dialogue form of Cicero’s philosophical treatises allows the possibility of further play of ambiguity. Schofield argues strongly for a close relationship between the rhetorical character and philosophical point of Cicero’s treatise, and suggests, in particular, that the traditional Stoic anecdotal defence of divination made ‘knockabout’ cross-questioning the appropriate sceptical response. I would argue more broadly that the dialogue structure itself could be used (though was not always) as a

53

54 De Natura Deorum 3. 2. 5. De Divinatione 2. 33. 70. As Schofield points out, appeal to tradition can also be seen as characteristic of the sceptical philosopher, in places where reason appears to be insufficient to establish the truth (below, 55–6 = 000–000). 55

Cicero and Divination

343

depersonalizing, distancing device, which obviated the need for Cicero, as author, to identify with any one expressed opinion.56 In both De Divinatione and De Natura Deorum the dialogue form allows the easy expression of uncertainty and suspension of judgement. This goes beyond the simple formula of De Divinatione, where each character offers a set of arguments on one side of the question without any final conclusion or evaluation of the two positions. In the more complex structure of debate in De Natura Deorum the interplay of characters is concluded by Cicero’s report of the ‘votes cast’ on each side—again an equal division: Here the conversation ended, and we parted, Velleius thinking Cotta’s discourse to be the truer, while I that of Balbus approximated more nearly to a semblance of the truth. Haec cum essent dicta, ita discessimus ut Velleio Cottae disputatio verior, mihi Balbi ad veritatis similitudinem videretur esse propensior.57

Likewise, the dialogue form, with the possibility of continuity (or discontinuity) of character from one fictional discussion to the next, allows a more sustained undercutting of a directed thread of thought and the effacement of, in particular, the authorial voice. Cicero’s own suspension of judgement is made more obvious to the reader by the fact that he appears as a character in both De Natura Deorum and its supplement De Divinatione, and expresses contradictory positions in each work. There is a striking contrast in form and function of the dialogue between the theological areas of Cicero’s philosophical corpus and those concerned with such topics as politics or ethics. In the latter, Cicero uses dialogue to direct the argument to a conclusion which the reader tends to equate with the position of the writer, even if the writer is not present in propria persona; and the shifting, uncertain perspective of the characters, typical of the theological works, is almost entirely absent. So, for example, in the Tusculan Disputations, the character of Marcus plays a didactic role, guiding the argument to clear conclusions on such issues as the endurance of pain or the despising of death; similarly in De Amicitia, Laelius is given a directing role which involves no problematic juxtapositions or obviously conflicting attitudes.58 The principal reason

56

See below, pp. 45–6. Note also the sophisticated approach offered by Levine 1957:

7–36. 57

De Natura Deorum 3. 40. 95. Only De Finibus approaches the theological works in formal character; but it is still hardly comparable, being much more an explicit exposition of the different ethical systems of the different Greek schools (De Finibus 1. 4. 12). 58

344

Cicero and Divination

for this difference in the character of the dialogue between one part of the corpus and another is fairly clear: its formal potential for allowing suspension of judgement was most fully exploited in that area of Cicero’s philosophy—namely the theology of state religion—where the difficulties of integration between Roman practice and Hellenizing thought were most acute.

The Choice of Interlocutors and Their Roles The identity, character, and interrelationship of the various interlocutors in the dialogues are important elements in the presentation of ambivalence and suspension of judgement. The particular conflicts highlighted by the choice of Cotta and Marcus—both priests and philosophers—as main speakers in the De Natura Deorum and De Divinatione, may be set against more general principles of character choice, which underline the problems of the theological works, in contrast to other areas of Cicero’s philosophical writing. A most important, though rarely recognized, feature of the speakers in De Natura Deorum and De Divinatione is their parity of status. All the chosen characters (with the possible exception of Velleius in De Natura Deorum) enjoyed high political and social prestige—in contrast to the position in (say) the Tusculans, where the role of the second interlocutor is definitely subordinate and almost that of a pupil.59 With Balbus and Cotta (in De Natura Deorum) and Quintus (in De Divinatione) the status of the characters gives weight and added point to the arguments they are said to espouse, making the reader’s assessment of their arguments all the more difficult. Indeed, in the case of De Divinatione, the choice of two brothers as the two interlocutors in the dialogue serves not, in my view, to mark out the inferiority of the arguments of the younger and less distinguished, but to equalize as nearly as possible the weight of each side of the debate.60 The contrast between the theological dialogues and many of the other treatises is clear, and the reason for it not hard to conjecture: where the subject allowed a clearly directed argument, that argument was often further highlighted by a marked difference in status between the main speaker and his interlocutors; where the problems of handling the subject in Greek philosophical terms made a directed argument For various solutions to the identity of ‘A’ in Tusculanae Disputationes, see T. W. Dougan, ed., ad 1. 5. 9. For my purposes it is sufficient to recognize that the interlocutor (whoever he is) is of clearly inferior status. 60 For a fuller discussion of the role of Quintus, see Schofield 1986: 60–1. 59

Cicero and Divination

345

impossible, the equal status of the opposing views put forward was underlined by the equality of social and political status between the various interlocutors. Cicero had considerable freedom of choice in selecting the interlocutors for his philosophical dialogues. With hindsight it is easy to forget that the characters were not predetermined and that, even within the necessarily limited circle of Romans who might be considered suitable mouthpieces for Greek philosophy, Cicero had a relatively wide range from which to choose. When, for example, Cotta or Marcus himself (as both priests and philosophers) are selected as main interlocutors, we must regard this as a conscious choice on Cicero’s part, and a conscious attempt to highlight the tensions inherent in their dual role. The difficulties we find in the contradictory position of these two characters are intentional ones; they could, after all, have been avoided simply by the choice of different interlocutors. Although it is difficult in most cases to be certain how far Cicero was aware of the problems and contradictions so evident in a modern reading of his work, his choice of characters represents one area of certainty: he was not only anxious to provide an integration between traditional religious practice and Hellenizing philosophy; he was also concerned to highlight the problems of such an integration. Many of these examples of uncertainty or tension I have cited from the theological works were, no doubt, individually influenced by other factors, such as Cicero’s imitation of Academic philosophical practice. When taken together, however, they form a much more telling picture, marking out Cicero’s theological dialogues (in contrast, say, to the ethical works) as a particular area of difficulty and uncertainty for their author. We can never be absolutely certain what the ultimate cause of that difficulty was; but the place of Cicero in the history of philosophy at Rome and his innovation in integrating Roman and Hellenizing traditions strongly suggest that the difficulty is to be primarily explained by reference to the predictable (yet hard to resolve) cultural clash between different systems of thought.

DE DIVINATIONE II: THE CASE AGAINST SCEPTICISM This paper began with simple expressions of doubt on the conventional assumption that Cicero’s personal scepticism can be deduced from the second book of De Divinatione. These doubts have been reinforced by a consideration of the work in its wider literary and intellectual context.

346

Cicero and Divination

I have shown that it is not justifiable to extract one part of one work and to claim for that part the status of Cicero’s ‘real views’; nor is it justifiable simply to compare an expression of scepticism in, say, De Divinatione with support for state religion in De Legibus and deduce from that either a change of opinion or insincerity in the earlier work. For each element of Ciceronian argument—whether tending towards ‘scepticism’ or ‘faith’, whether in the mouth of Marcus or any other of the chosen interlocutors—must be seen not in isolation, but as part of a broader whole, where firm views in one place find their contradiction or erosion in another, in the same or a related work. The identification of a clear authorial standpoint can be made only at the cost of blindness to the many devices within the theological works by which that standpoint is constantly evaded or sidestepped. It is not my purpose just to reinforce the negative arguments against Ciceronian religious scepticism, but rather to demonstrate different, positive, ways of understanding Cicero’s theological dialogues. Two particular conclusions may be suggested: the first, a reformulation of the arguments from cultural integration that have emerged at various points during the paper; the second, following from that, an attempt to evaluate the significance of De Divinatione as a document of the history of religion. (a) Cicero’s philosophical works demonstrate a remarkable mastery of a wide range of Greek philosophical theory and a sophistication in the (almost excessively) expert deployment of Greek philosophical arguments. Nonetheless, the theological works in particular are also tentative in the sense that they represent the first attempts at the formation of a scientific discourse on Roman religion. A comparison between Ciceronian philosophy and that of the generation or two following him makes clearer this tentative quality: while Horace and Seneca, for example, deftly used an established discourse of Greco-Roman philosophy to explore freely such notions as otium or divitiae, Cicero, by contrast, was still (at least in the area of theology) attempting to establish the discourse within which such philosophical argument might be possible. In this sense, Cicero’s handling of state religion in his philosophical works does not constitute the argued presentation of an opinion or a view; it constitutes rather the process of formation of a discourse on theology. (b) De Divinatione is a work of much greater importance for the history of religion than any exclusive concentration on its author’s views has ever revealed. Its most striking positive feature is the fact that it is a dialogue about religion; that, over two books, an argument is sustained specifically on the subject of divination. This amounts to a clear indication of one of the most important religious developments of

Cicero and Divination

347

the late Republic. Not only was it a period characterized by intense interest in religion, as has been so well documented by Momigliano and Rawson;61 but it was, more crucially, the period when ‘religion’, as an activity and a subject, became clearly defined out of the traditional, undifferentiated, politico-religious amalgam of Roman public life. The differentiation of religion is evident in many aspects of late Republican life. A classic example is found in the history of the Bacchic cult in the early second century . The unprecedented nature of this cult lay in its status, for the first time in Roman Italy, as a specifically religious organization; and it was partly this novelty which led the senate in 186  severely to curtail its activities.62 Later, however, individual members of the traditional Roman governing class are themselves found participating in, rather than prohibiting, such developments. Appius Claudius Pulcher in the first century , for example, assumed the role of (selfproclaimed) ‘expert’ in augury. He has often been treated as a conservative traditionalist, vociferously upholding the rites of state religion in the face of growing neglect or scepticism amongst his peers. In fact, both the missionary zeal of Pulcher and the doubts or unconcern of his contemporaries are part of the same, new, phenomenon. It was now possible for members of the Roman élite to proclaim a particular stance in relation to religious activity; it was no longer simply ‘something they did’.63 So with De Divinatione (and its related dialogues), for the first time, we find developed verbal arguments specifically on state religion. ‘Religion’ had now been defined as a subject of Roman discourse.

B I B L I OG R A P H Y Astin, A. E. (1967). Scipio Aemilianus. Oxford. Belsey, C. (1980). Critical Practice. London. Boyancé, P. (1963). Lucrèce et l’épicurisme. Paris.

61 See the articles of Momigliano 1984a: 199–211; Momigliano 1984b: 873–92; Rawson 1985: 298–316. 62 The clearest account of this aspect of the incidents of 186 is given by North 1979: 85–103. 63 Note Appius Claudius’ book on augury (Cicero, Ad Familiares 3. 4. 1), his debate with C. Marcellus on the nature of the augural discipline (De Divinatione 2. 35. 75; De Legibus 2. 13. 32) and his popular (almost humorous) image as the keen augur par excellence (Varro, De Re Rustica 3. 2. 2). Other contemporary religious experts include P. Nigidius Figulus and Aulus Caecina (for whom, see Rawson 1985: 309–12 and 304–6). They are wrongly called ‘traditionalists’ by, for example, Wardman 1982: 46–7.

348

Cicero and Divination

Denyer, N. (1985). ‘The case against divination: an examination of Cicero’s De Divinatione’. PCPhS n.s. 31: 1–10. Douglas, A. E. (1962). ‘Platonis Aemulus’. G&R 9: 41–5. Dumézil, G. (1970). Archaic Roman Religion. Chicago—London (1st French edn, as La religion romaine archaïque (1966). Paris). Garbarino, G. (1973). Roma e la filosofia greca dalle origini alla fine del II secolo A. C.: Raccolta di testi con introduzione e commenti. Turin. Gernet, L. (1968). ‘Choses visibles et choses invisibles’, Anthropologie de la Grèce antique, 405–14 (Eng. trans. in The Anthropology of Ancient Greece (1981), 343–5I). Goody, J. (1977). The Domestication of the Savage Mind. Cambridge. Guthrie, W. K. C. (1962). A History of Greek Philosophy I. Cambridge—New York. Horton R. and R. H. Finnegan, eds. (1973). Modes of Thought: Essays in Thinking in Western and Non-Western Societies. London. Jocelyn, H. D. (1976). ‘The ruling class of the Roman Republic and Greek philosophers’. BullRylandsLib. 59: 323–66. Latte, K. (1960). Römische Religionsgeschichte. Handbuch der Altertumswissenschaft, 5. 4. Munich. Le Gall, J. (1975). La religion romaine de l’époque de Caton l’Ancien au règne de l’Empereur Commode. Paris. Levine, P. (1957). ‘The original design and the publication of the De Natura Deorum’. HSCPh 62: 7–36. Linderski, J. (1982). ‘Cicero and Roman divination’. PP 36: 12–38. Linderski, J. (1986). ‘The Augural Law’. ANRW II.16.3. Berlin—New York. Lloyd, G. E. R. (1979). Magic, Reason and Experience: Studies in the Origin and Development of Greek Science. Cambridge. Lloyd, G. E. R. (1983). Science, Folklore and Ideology: Studies in the Life Sciences in Ancient Greece. Cambridge—New York. Momigliano, A. D. (1984a). ‘Religion in Athens, Rome and Jerusalem in the first century BC’. AnnScNormPisa Series 3, Vol. 14, fasc.3: 873–92. Momigliano, A. D. (1984b). ‘The theological efforts of the Roman upper classes in the first century B.C.’ CPh 79: 199–211. Needham, R. (1972). Belief, Language and Experience. Oxford. North, J. A. (1979). ‘Religious toleration in Republican Rome’. PCPhS n.s. 25: 85–103. Rawson, E. (1975). Cicero: A Portrait. London. Rawson, E. (1985). Intellectual Life in the Late Roman Republic. London. Regell, P. (1878). De augurum publicorum libris. Diss. Vratislaviae. Regell, P. (1882). Fragmenta auguralia. Progr. Hirschberg. Schaublin, C. (1985). ‘Cicero, “De Divinatione” und Poseidonius’. MusHelv 42: I57–67. Schofield, M. (1986). ‘Cicero for and against divination’. JRS 76: 47–65. Taylor, L. R. (1949). Party Politics in the Age of Caesar. Berkeley.

Cicero and Divination

349

Troiani, L. (1984). ‘La religione e Cicerone’. RSI 96: 920–52. Vernant, J.-P. (1965). ‘La formation de la pensée positive dans la Grèce antique’, in Mythe et pensée chez les Grecs, 285–3I4 (trans. in his Myth and Thought among the Greeks (1983), 343–74). Wardman, A. (1982). Religion and Statecraft among the Romans. London, 46–7.

14 The Peace of Augustus, the Equinox, and the Centre of the World Alfred Schmid

The transformation of a society into a micro-cosmos when the society in question is already formed and has been institutionalized (and so has no foundational need for a ‘cosmic identity’), is certainly a paradoxical undertaking.1 But post-Actium it must at least have been made easier thanks to two circumstances. First of all, for a long time Rome’s ruling class had been on the way towards a ‘more global’ form of political identity,2 it had followed in the ‘ecumenical’ footsteps of Alexander the Great and had surpassed him,3 and thereby the globe—using it in a geographically incorrect sense—became the symbol of a hegemony that could encompass at most one of the four possible Oikoumenai of the globe, even going by a euphoric interpretation.4 It is plausible, then, to regard this ‘symbolic’ globe not as the earth’s geographical globe, which continued to depict the problematic and irregular arrangement of land and sea as a grid that reflected the sublime geometry of the celestial sphere (from Plato onwards itself the form of the periodos (cycle) of the principles of all that could be conceived). The Platonic kosmos-ouranos was the symbol of a mastery that wanted to be taken rather as ‘universal and cosmic’ than as ‘geographical’.5 The palpably reified form of this ‘Urano-centric’ globality was in accordance with an intellectual milieu that was accustomed to follow the verses of Aratus haptically, ‘from the 1 A shortened version of this thesis was presented by the author on the occasion of a congress in Malaga (October 2001) and published in 2002 (Schmid 2002). 2 See Nicolet 1991: 29ff.; Hardie 1986: 368, 377. 3 Polyb. 1. 1. 5 (for other texts, see Nicolet 1991: 30). 4 Nicolet 1991: 35. For the ancient state of knowledge on the structure and suitability for habitation of the earth-globe, Abel 1974. 5 Nicolet 1991.

Alfred Schmid, The Peace of Augustus, the Equinox, and the Centre of the World In: The Religious History of the Roman Empire: The Republican Centuries. Edited by: John North, Oxford University Press. © Oxford University Press 2023. DOI: 10.1093/oso/9780199644063.003.0015

The Peace of Augustus

351

outside’, with the help of a celestial globe made of stone or wood.6 Another factor was helpful: the Civil War and its trauma must at least have made the Romans more ready for a symbolism that imparted meaning with a fresh intensity. It was precisely images of unbreakable unity, of consensus, concordia, and pax7 that must have been welcome to a society behind whose everyday life there lay a ‘point zero’ of sorts8 and whose system of political values and self-identity had failed depressingly in respect of the ability of its leading representatives to achieve consensus. From Naulochos onwards at least the promise of a peace that was supposed to turn into a fundamental constant of new political order becomes an element in Rome’s leading power-holders presenting their intentions. Thus, the poets Virgil, Horace, and Tibullus, where they sang the praises of peace, were complying, according to Stefan Weinstock, with an official request made by Augustus to create a cult of peace.9 An altar of the Peace of Augustus (Ara Pacis Augustae: RGDA 12. 2) was decided upon by the Senate in the year 13  and it was dedicated on the Field of Mars four years later.10 The literature about this monument that is so imposing from an artistic viewpoint as well has in the meantime become impossible to assess;11 what mainly comes into focus here is the fact that new questions arise with Edmund Buchner’s discoveries:12

6 On this, see Le Boeuffle 1983: x f.; id. 1975: xxii f.; on this, the reading cited in Nicolet 1991: 51 n. 30. A splendid example of this is Geminos (ed. Manitius 5.63) who can be ‘seen’ sitting in front of a globe, where he notices that the location of the horizon is marked on the support-frame (sphairotheke) of the globe. 7 On pax, together with salus publica and Concordia as a central theme of the Augustan ‘ideology’, see Dio Cassius 54. 35. 2. Koch 1949: 2430–6, on the Augustan origin of the Paxcult; Thome 2000: 93ff., on the development of the pax-project in the late Republic; Richard 1963: 330ff. on Caesar’s new Concordia and passim; Weinstock 1960: 45f.: Caesar brought the goddess Pax to Rome ‘to give expression to this political philosophy’. (The reference is to Cicero’s presentation of the Roman Empire as the peace-bringing protective power.) 8 See Bleicken’s conclusion on the situation in 35  (Bleicken 1998: 235). 9 Weinstock 1960: 47 with supporting evidence. That the new and enduring state of peace, the pax perpetua, was a pax Victoria parta (a peace born of Victory) and that it had to be based on the comprehensive ‘victory by land and sea’ (terra marique) was undisputed. See Thome 2000: 98ff.; Gruen 1985; Glei 1991. But it was ‘total’ imperial domination that maintained the wide-ranging claim of this peace—see Propertius 4. 65. 37, where Apollo addresses Augustus—before Actium: Mox ait ‘O Longa mundi servator ab Alba, / Auguste, Hectoreis cognite maior avis, / vince mari; iam terra tua est’ (Soon he (Apollo) spoke: ‘O saviour of the world, sprung from Alba Longa, known a greater man than your ancestors under Hector’s command / conquer by sea—the land is yours already’). But that in Rome the dominant conception was anyway ‘The claim of Rome to represent on its own the order of law and peace in the world and to be allowed to impose it on all others’: Klingner 1965: 618. 10 Constitutio on 4 July 13  (Fasti Amiterni CIL I2, 244) and dedicatio on 30 Jan. 9  (Fasti Praenestini, CIL I2, 212; Ovid, Fasti 1. 709). 11 12 An overview in Kienast 1999: 239f. n. 115. Buchner 1982.

352

The Peace of Augustus

Fig. 14.1 The solarium with the Ara Pacis, after E. Buchner.

the integration of the Ara Pacis within an arrangement (through the so-called solarium Augusti) that is now supposed to express the ‘cosmic’ self-understanding of the great establisher of peace13 (and so to express a cosmic concept of peace). The image of the obelisk, the peak of whose shadow is supposed to move across the west portal of the Ara Pacis in a gigantic grid of lines on the evening of September 23, has gone into most of the standard works on Augustus (Fig. 14.1).14 The gnomon of this huge ‘sun-dial’ was specially brought over from Egypt and dedicated to Sol,15 so that it was itself part of a consecrated area. That the works of the gnomon must have a connection with the Ara Pacis was plausibly suggested by Buchner, given the chronological co-ordination of the construction of both monuments.16 And it is convincing that for a viewer the conspicuous shadow cast by the obelisk will have been so dominant that the ara which lay nearby must have been influenced by it visually. Conceptual co-ordination seems a reasonable suggestion. Buchner then accepted that on 23 September (according to Buchner not just the birthday of the princeps,17 but also the date of the equinox)18 the top of the obelisk’s shadow, which was adorned with a sphere (that was meant to symbolize the world) lay in a dead straight line 13 See Buchner 1982: 37: ‘This construction is, so to speak, the horoscope of the new ruler, vast in scale and explaining cosmic correlations.’ 14 Buchner 1982: 43, figs. 13 and 14. 15 CIL VI 702: Imp. Caesar Divi f. Augustus pontifex maximus imp. XII trib. Pot. XIV Aegypto in potestatem populi Romani redacta Soli donum dedit. According to Halsberghe (1972: 28f.), Sol should here mean the Roman Sol, who had, so to speak, overwhelmed the Egyptian one. 16 Buchner 1982: 10, 48ff. 17 It is certain that the birthday of Augustus was highlighted in the Calendar of Festivals. According to Gros 1976: 32f., six temple-birthdays fell on 23 September, and so did the Festivals of the corresponding deities: Apollo in the circus; Iuppiter Stator; Mars in the circus; Neptune in the circus; Iuno Regina; the temple of Felicitas in the campus—but see Gradel 2002: 131 n. 49: the attribution of the temple-birthdays was just Degrassi’s hypothesis—what was meant was that these were the deities to whom offerings were made on Augustus’ birthday. 18 Buchner 1982: 36.

The Peace of Augustus

353

to the entrance to the altar of peace and crept inside. Thus, the birth of the princeps was associated with the (altar of) peace and both of them were associated with the cosmos (the order of which was actually demonstrated here by the sun’s regular path).19 Now, however, in an article of 1990, Michael Schütz called Buchner’s results into question in a very fundamental way. The fact that Schütz’s objections have obviously not had any far-reaching impact on the debate is quite possibly to be put down to the details of the critical points, often purely mathematical, astronomical, and also calendrical. Augustus’ cosmic birthday at the equinox has in any case made its way without comment into the Feuilletons in the daily papers. Perhaps Buchner’s geometrical constructions, in parts complex, are simply not comprehensible to most of the historians who have been called to pass judgement on them—but the situation presented nowadays by the debate over these startling correlations seems strange and unsatisfactory. For as a rule the research on Augustus is marked, on the one hand, by the acceptance of Buchner’s thesis (with the ingenious connection between the Peace of Augustus, the birth of Augustus, and the cosmic date of the equinox)20 and on the other by making a reference to Schütz’s divergent opinion in a footnote.21 This is somewhat disconcerting, because taken seriously the latter’s critique would destroy the whole ‘cosmic conception’ of the arrangements on the Field of Mars.22 So what were Schütz’s most serious objections? I will put together what seems most relevant to me: the shadow of the sphere on the top of the gnomon was already invisible before it could have reached the Ara Pacis.23 Indeed this objection calls into question a spectacular detail in Buchner’s original version, but in my opinion the symbolic significance of the sphere (as a globe) on top of the gnomon would one way or

19

The association of this arrangement with the Mausoleum as well, which Buchner (1982: 54f.) and in his turn Schaldach (1997: 90ff.) want to explain, I will set aside here. 20 The equinox oddly enough gets no further attention. As if it were totally clear that the one born on the equinox would also be natus ad pacem (born for peace). Because the ‘Peace-allusion’ is established here through the Altar of Peace, the equinox (corner-stone of the solar year) is obviously relevant here as the cosmic background for Augustus’ birthday. See Wallace-Hadrill 1987: 227 (about Augustus): ‘cosmically he is born on the equinox to bring peace to the world’. What the equinox is supposed to contribute here, is not stated. 21 Some representative examples: Kienast 19993: 240f.; Bleicken 1998: 516; Galinsky 1996: 146; Castriota 1995: 131f.; Buchner’s thesis was adopted without criticism by Eck 1998: 114. 22 See e.g. Barton 1984: 46. 23 Schütz 1990: 450ff.; Schaldach 1997: 85f. Buchner now seems (tacitly) to admit the point himself: Buchner 1996: 35, where the visibility is related to a figure (fig. 22, p. 392), which in any case does not reach the Ara Pacis.

354

The Peace of Augustus

another be a ‘redundant’ statement: in the underlying paradigm of the analemma the tip of the gnomon marks the place of the earth-globe (comparatively extensionless) in the middle of the heavenly-globe.24 This objection would therefore not refute Buchner’s connection if the Ara Pacis remains positioned on the exact line along which the peak of the shadow encroached on exactly two days in the year (the spring and autumn equinoxes). Therefore it is much more serious if there might be grave doubts about the height of the gnomon also25 (which determines the network of lines). Hence Schütz has labelled Buchner’s equinox line (it is the decisive element of the connection between the Ara Pacis and the ‘solarium’) as a mere working hypothesis.26 It may at any rate be valid that the possible variation in Buchner’s values for the Augustan equinox line is not indefinite (the excavation has in fact uncovered a month-line—the variable of the gnomon’s height can still be established in a way that the result comes out to Buchner’s value). For the higher-lying ‘Vespasianic’ level, the excavation indicated a variation of 2.5 m to the south in comparison with Buchner’s anticipated value for the equinox line. Karlheinz Schaldach for instance, who worked from different basic assumptions, calculated a value that differs from Buchner’s by about half a metre.27 For all that, Buchner has in the meantime claimed in two more recent publications that he has confirmed his estimated height for the gnomon—hence he refers to a new bore-hole that has confirmed the foundations of the obelisk almost ‘to the centimetre’,28 without however furnishing more precise information.29 With the material published so far it is therefore difficult to assess the situation (especially the exact position of the equinox line on the Ara Pacis), if it depends, to put it simply, on the connection between the equinox and the Altar of Peace then no one can count on an argument that is clearly comprehensible and can be proven exactly. The layman may in any case be struck by the fact that to date no one has put forward a definite equinox line that diverges from Buchner’s (and which does not in some way bisect the Ara Pacis): Schaldach, accepting generally Schütz’s critique, also considers the connection between the

24

Szabó 1992: 82. Schütz 1990: 436–42; Schaldach 1997: 87ff. Buchner (1996: 36) defends the precise gnomon height. It should be noted that it was not the original Augustan foundation, but a later one, earlier than Domitianic, at present marked as Vespasianic, that was excavated. So, it is at least a matter for debate which was the exact Augustan position of the gnomon. 26 27 28 Schütz 1990: 444. Schaldach 1997: 88. Buchner 1996: 163. 29 While in 1996 he fielded another argument, in order to defend his 29.5 m for the height of the gnomon. 25

The Peace of Augustus

355

obelisk, the Ara Pacis and Mausoleum to be ‘not disproven’.30 However, the exceptionally precise connections assumed by Buchner (according to which the line of the equinox bisects the exact centre of the ara in a sophisticated relationship to the openings of the door)31 cannot be considered proven (not least because here every estimate must contain a few variables). But the objections to Buchner’s network of lines and thus against the very existence of a sundial at all are fundamental. Schütz might with reason point to the passage of Pliny32 (Natural History 36. 72f.), which Buchner also took as a starting point and which definitely does not speak of a sundial, but for which a rather more suitable fit is simply a ‘meridian’. So the device really would have made the waxing and waning of the dierum ac noctium magnitudines (the extent of days and nights) visible through the movement of the midday shadow’s length from the longest day (shortest night) to the shortest day (the longest night)—that does not imply any sort of sun clock (to determine the time of day), whose hour lines have obviously not yet been uncovered. Apart from this, one can doubt whether there was any sense in this kind of huge sundial33 (which was quite simply only visible from a bird’s-eye perspective). The device could not have been a ‘solarium’, it was instead a meridian, an assumption Buchner obviously accepts for the ‘Vespasianic’ level.34 But with such evidence it is still possible to make a connection between the equinox and the Ara Pacis, especially if this connection indicates the date rather than the hour of day. The question remains was the line of the equinox marked out? and how was this done? After all one might suppose that regular paving already made clear the straight line of the shadow’s path on the corresponding day. Now to Buchner’s most indisputable mistake (and, irritatingly, one accepted without comment in the research): under the Julian calendar, Augustus’ official birth-date, 23 September, definitely did not fall on the equinox—so long as one does not take as a starting point the coincidence whereby in the year 13  it did indeed fall erroneously on the 23rd thanks to faulty calculations (which Augustus rectified during his lifetime).35 According to the intention and convention of the Julian calendar, the autumn equinox must fall on 25 or 26

30

31 1997: 93. Buchner 1982: 29. 33 Schütz 1990: 4532–5. Schaldach 1997: 80. 34 So Schaldach 1997: 84. See Buchner 1996: 36 and fig. 22 (392) with a new version of his sundial. 35 Schütz 1990: 446ff. 32

356

The Peace of Augustus

September.36 And that is no mere detail: Schütz could refer to Geminos, who makes it clear that it was quite well known (as simple observation shows) that it was precisely at the equinoxes that the length of the shadows change most clearly and most quickly.37 It would indeed have been clearly visible, not in the year 13 (when erroneously the ‘Buchner equinox’ worked) but already a few years later, that on the natalis of the princeps the top of the shadow still lay in the sign of Virgo and at least three degrees away from the equinox. The error in the rhythm of the cycle (that fell under Lepidus’ competence as pontifex maximus) was only eradicated in the year  4 with Augustus himself as the pontifex maximus,38 and thus the birthday of Augustus was set away further from the equinox in such a way that it was to be seen on the Field of Mars. If the intention in this arrangement was to make a reference to the movement of the sun (= movement of the shadow) at the equinox on the Ara Pacis (and this I take to be Buchner’s central argument), then it could not have been a reference to the birthday of the princeps as well. Unless the intention was to demonstrate that this birthday did indeed fall very near the equinox—the regulae would however have made clear precisely the difference, and two times in the year at least six days would have shared the dignity of this proximity to the equinox. Yet it is clear that if the ara was orientated exactly along the line of the equinox, on September 23 the tip of the gnomon’s shadow definitely did not make its way onto this plumb line, but instead diverged from it visibly. In view of this far from esoteric connection it is hard to understand why simply all the research as a rule continues to speak of Augustus’ birthday on the equinox, and that Buchner himself keeps placing the equinox on 23 September and still designates the arrangement on the Field of Mars as a ‘birthday arrangement’.39

36 One must admit after all that relevant Handbooks are not especially lucid on this question. Ginzel (1911: 282) following the sources mentions the 26th (but also the 24th and 25th) as the date of the equinox, while on p. 285, Ginzel’s exact calculation for the year 45  gives the 25th after noon. On p. 519, there is an overall tabulation of the movement of the Julianic year from which one can infer that the Julianic autumn equinox, as astronomically exact for 100 , happened at about midnight on the 26th; however, 200 years later it came to occur about midday on the 24th. But what can be considered a normal and conventional date can be gathered from the evidence in Abry (1988: 107) for her horoscope of Augustus; according to Ovid, Columella, and the fasti Venusini, for the Julianic 23 September, there must have been a shortfall of at least three degrees up till the equinox, which suggests rather the 26th as the commoner date for the autumn equinox. 37 Geminos (ed. Manitius) VI, 35; see also, Cleomedes I 4, 30–44, p. 19 Todd. 38 Samuel 1972: 156f. 39 Buchner 1996: 35: ‘It remains true that the equinox line, the line of 23 September (sic), of the birthday of Augustus, points to the middle of the Ara Pacis, while the h. A. and the Ara Pacis together constitute a birthday complex’.

The Peace of Augustus

357

Fig. 14.2 Augustus’ horoscope in Rome on 23 September 63 , c.5”15 (System B).

But they knew better in the year 9 : indeed the decision of the provincial assembly of Asia in this year (OGIS 485) speaks explicitly of the proximity of Augustus’ birthday to the date of the equinox—and this was set up as an occasion (coincidental or divine) to devise this honour (namely that henceforth in Asia the year would begin with this auspicious birthday). With the birth of Augustus there begins the time of life— if this distinction stresses the analogy with the ‘Birth of the Year’, and if the coincidence that Augustus’ birthday ‘nearly’ coincides with the equinox, which was already in use as a marker that dictated the calendar, made the honour possible,40 then it is to be stressed that in this case it was no problem that it was just an approximate coincidence. Greater 40

See the commentary of Laffi 1967: 49ff.; Kienast 19993: 247f.

358

The Peace of Augustus

Fig. 14.3 Augustus’ horoscope in Rome on 23 September 63 , c.5”30 (Hipparchic system).

exactitude was not necessary for a civic calendar—it was sufficient that the natural ordering of time (the proximity to the year point) made the honour possible, or even suggested it. Whether September 23 fell exactly on the equinox was of no matter here,41 when civic, not astronomical time was at stake. Both events from the year 9  refer to a fundamental concept that made visible the ‘natural’ ordering of time (whose guarantor and protector was now the pontifex maximus Augustus, as Caesar’s successor) as the firm basis of civic time. The Julian calendar was in fact introduced into the whole of Asia at this occasion, which honoured 41 The beginning of the Roman political year on 1 January was still further away from the date of the winter solstice.

The Peace of Augustus

359

Augustus. But for the question of the Ara Pacis and the equinox in Rome the only thing that is important is that an inscription from the year of dedication of the arrangement on the Field of Mars establishes in words that Augustus’ birthday simply did not fall on the equinox. Now if one had wanted to promote in Rome the coinciding of Augustus’ birthday with a cosmic-providential pivotal moment of time (as of the world), then an arrangement that demonstrated with exact precision the astronomical imprecision of the coincidence would be a very unsuitable means (at all events, if one conceives of the connections the way Buchner does). In any case, a certain vagueness appears in the literature concerned with the details suggested here about calendar and astronomy well before Buchner,42 but here such details are what is decisive once and for all. And that the ‘birthday device’ in Buchner’s version is quite unlikely is substantiated by a further detail that has gone unnoticed up to this point. According to the general, almost canonical view Augustus is supposed to have been a ‘Libra’ (compare Schmid 2005: 19–36)—only it is not at all so straightforward, since it depended on which zodiac was taken as a basis for the astrological calculation. It must be stressed that in Augustus’ time there were fundamentally two possibilities:43 besides the zodiac that we know of today44 there was a zodiac that, following Otto Neugebauer, is called ‘System B’45 (it is also Mommsen’s ‘Bauernkalender’). Now quite obviously Caesar as reformer of the calendar made use of this,46 which was also the reason why (together with the authority of Neugebauer, whose version points in this direction) it is widely assumed that, especially after Caesar and in Rome, it was this one that became the canonical zodiac. Now in contrast to the way that is now familiar to us, this zodiac simply locates the so-called ‘Year Points’ not at the exact start, but

42 Typical is the real chaos that J. Bayet created, in a paper that was by no means irrelevant, when he observed that according to Eudoxus, who is dated to the second century , the equinox ought to fall on 19 September, but Ptolemy had fixed it exactly on the 23rd (the source is J. Carcopino). Bayet had evidently (among other things) overlooked the fact that a different zodiac-system was being used and had therefore confused the equinox with the beginning of Libra—on which, see below. 43 Abry 1988: 107 believes it to be certain that ‘for the contemporaries of Augustus the sun (i.e. on Augustus’ birthday) was already at the 4th degree of Libra’; similarly, Brind’Amour 1983: 73f. 44 Marked out by Columella as the Hipparchi ratio, their greater subtilitas (subtlety) is stressed (9. 14. 12). 45 Neugebauer 1975; 594–600; Neugebauer and van Hoesen 1959: 4 and 15. 46 Rehm 1929: 1156ff.; Rehm 1922: 323f.; Ginzel 1911: 281–4; 420f. See Appendix I, in Schmid 2005: 335–9.

360

The Peace of Augustus

precisely on the eighth degree of the cardinal (or ‘tropic’) signs. This means that the solstices must at all times mark the eighth degree of Cancer (and Capricorn, respectively) and the equinoxes the eighth degree of Libra (and Aries)—whereas for us the convention (also in use in antiquity) holds that these signs should also begin with these points of the year (so that the equinox is supposed to mark exactly the first degree of Libra/Aries). The fasti Venusini report (CIL I, s. 221) the correct Julian solstice to be on the 26th of June, but at the same time it is noted that the sun entered the sign of Cancer on 16 June. So it is clear that here the zodiac in use was ‘System B’—according to this the sun would already be in Libra on 23 September (since about 18/19) at around 4 degrees (but with this still around 4 degrees away from the equinox, which here must lie at 8 degrees of Libra).47 If one simply compares the meridian according to Buchner’s excavation with the arrangement of the signs of the zodiac48 in light of this question, it is shown beyond doubt that this arrangement cannot have used the zodiac ‘System B’ (supposedly the absolutely canonical one at that time). For according to this convention at the equinox exactly the eighth degree of Libra and the eighth degree of Aries had to lie opposite one another. Namely these degrees mark two dates (in spring and autumn) on which the increasing (climbing up the line of midday) and the decreasing shadow have precisely the same length. But then the signs of the zodiac could not lie parallel to one another (as on the Field of Mars), but had to be organized so they were clearly shifted opposite one another—according to Fig. 14.4 for the section dug up by Buchner (compare Buchner’s find according to Fig. 14.5). So, it seems certain that the huge gnomon showed the birthday of Augustus according to the Hipparchi ratio in Virgo, around the 26/27th degree. If Augustus counted as a Libra, or wished to be so counted, this mechanism would have been completely unsuited to celebrating the nativitas Augusti! But there are thoroughly good reasons for accepting that Augustus was considered a Libra (that is, that for him a horoscope according to ‘System B’ was probably used);49 they have been put together by J.-H. Abry.50 To be noted is a passage from Manilius (4. 546–52) concerning someone born under Libra: ‘when the autumn shearings begin rising, a fortunate one is born under the balance of Libra . . . ’.51

47

See on that Figs. 14.2 and 14.3. Buchner 1982: 76f.; Buchner 1996: 392 with fig. 22. See appendix II, in Schmid 2005: 405–8, with the different possibilities. 50 1988: 108–11. 51 ‘As judge, he will set up the scales of life and death, and place his yoke upon the lands and make laws. Cities and Kingdoms will tremble and be ruled by the command of one 48 49

The Peace of Augustus

361

Fig. 14.4 For discussion of Fig 14.4 see: page 360.

Similarly, she notes an agate stone52 with a man’s head, behind whose throat scales are to be seen clearly (the bowls are clearly visible on both man’. (One notes the elegant bringing together of the three references for Libra: chelae (claws); libra (balance); iugum (yoke).) For the interpretation of this passage as referring to Augustus, see (in addition to Abry) Goold 1977: 266; Gundel 1926: 133f. (referring to Manilius 4. 205ff. that Servius Tullius ought already to have been born under Libra—against this, Hübner 1982: 569: suggesting the Jurist Servius Sulpicius); Brind’Amour 1983: 66, who sees in the ‘rising shears’ a hint of the Libra-ascendant in Augustus’ horoscope. 52

Fossing 1929: no. 1197.

362

The Peace of Augustus

Fig. 14.5 Excavated fragment of the animal-circuit pavement, after E. Buchner.

sides), while underneath the throat one can recognize the sign of Capricorn.53 For this, I would like to refer to a critique by Franz Boll of CCAG Bd. 8. 3:54 Boll brought it to attention that according to an epitome of Thrasyllos’ lost work, he was using the ‘System B’ zodiac.55 Since Thrasyllus is supposed to have come to Rome with Tiberius in the year 2 56 and, as the leading astrologer of his time, would probably have been the key adviser in Augustus’ circle57 it is obvious that at least the later Augustus was deemed a Libra—so presumably the edict of the year  1158 made the ‘Libra version’ of Augustus’ horoscope official.59 As the birth of Augustus was near sunrise (the ‘horoskopos’) the question of the

53 Fossing 1929: no. 177 had already taken this to refer to Augustus. Further reference was made to a series of gems and glass-pastes (see Zwierlein-Diehl 1979: no. 811) amongst which there were regularly three of the tropical signs, namely Aries, Cancer, and Capricorn, represented together with a man’s head, which consequently should stand for the sign Libra, as one of the four hinges of the world. 54 In Wochenschrift für Klassische Philologie 1913: no. 5, 119–27. 55 See CCAG 89. 3 p. 99.6. Boll also pointed out how Manilius is not in any way decisive between the two zodiacal systems; III. 257 and III. 680 indicate the eighth grade as the year’s turning-point; while his observations on the solstitial colure (I. 622) and on the rising times (III. 396ff.) both imply the ‘Hipparchic’ system. 56 Cramer 1954: 92ff. 57 Cf Suetonius, Divus Augustus 98.4, according to which Thrasyllus seems to have been close to Augustus, at any rate shortly before his death. 58 Dio Cassius 56. 25. 5. 59 Strictly speaking, that does not exclude the possibility of one astrologer Theogenes in Apollonia worshipping ‘a Virgo’ instead of ‘a Libra’.

The Peace of Augustus

363

relevant sun-sign affected the ‘horoskopos’ also; the rising sign being crucial in astrological interpretation. If we can thus give precedence to the Libra version of the Augustus horoscope (the Capricorn-question is not affected by it)—in any case for all the uncertainty, indications of a connection ‘Augustus-Virgo’ seem to be lacking—then there must have been weighty reasons why, in the arrangement by Novius Facundus on the Field of Mars, a different zodiac was chosen from the one that visualized the nativitas of Augustus. Nor is it difficult to suggest the reason why: if the wish was for this arrangement (precisely as meridian) to make ‘time’ visible as a device of cosmic symmetry which allowed the oscillation of the length of days around the middle of the equinoxes to be perceived, then the ‘System-B zodiac’ was conceptually unsuitable for this, precisely because it did not allow the symmetry of the relationships to emerge visually: in this device it is bisected precisely (hence the solstice signs Cancer and Capricorn would have had to go ‘around the corner’ at both ends of the midday line). But here it ought to have come down precisely to this symmetry (or to the ‘cosmicity’) of time. So if this device was to show that the Ara Pacis lay on the straight line that the top of the gnomon’s shadow described at the equinox60—then it must have been the equinox itself that was supposed to denote, show or bring home to observers something about the peace of Augustus.61 Whereas it should be noted here that this very phenomenon of the dead-straight shadow line at the equinoxes (in the path of whose extension the Ara Pacis perhaps lay) worked as a proof that, according to Ptolemy, the earth stood in the centre of heaven’s sphere.62 In a ground-breaking work on ideas of peace in antiquity, Harald Fuchs has, among other things, pointed out a concept (whose origins he took to be from the sophists)63 according to which the equal proportions of day and night in the yearly cycle could exemplify the concept of a norm in physis (nature), which reconciled conflict in an isotes free 60 The so-called ‘Dovetail’ figure in the original, and widely disseminated, presentation of Buchners’ thesis makes it clear that it is only at the equinoxes that the shadow with the tip describes a straight line. The further the sun gets away from the equinox, the more the path of the shadow’s tip bends in relation to the equinoctial middle either in a convex or else concave manner. Cf. Figs. 14.6 and 14.4. 61 The association of Pax with Augustus required no further elaboration, if it was already (RGDA 12) included in his nomenclature. 62 Ptolemy, Syntaxis 1. 5. If Earth was not in the middle of the celestial spheres (but shifted towards the north or south against one of the poles), then: ‘in the equinoctial days, the shadows of the gnomons thrown by the rising (sun) together with those thrown by the setting sun no longer fall, on the planes parallel with the horizon, for sensory perception along a straight line’ (transl. from Manitius). 63 Fuchs 1926: 109f.

364

The Peace of Augustus

from envy.64 As a model for Augustine, Fuchs reconstructed a Hellenistic conception of the cosmic state of peace,65 which can also be identified in Gregory Nazianzen,66 and which was illustrated not least by the phenomenon of the equinoxes. Philo of Alexandria was also familiar with the concept of the self-pacifying cosmos, whose ‘isemeric’ isotes binds together the hybrid excess as ‘isonomia’ and which was also supposed to be the ‘mother of equity/justice’ as equinoctial parity.67 Such parity, as the nomos akinetos of nature, itself clearly exemplifies the parity (the basis of justice) of the political sphere.68 Now according to Fuchs a kind of all-embracing conception of the pax omnium rerum in Latin literature is supposed to have been passed on by Varro (and indeed on the basis of a Stoic or Stoicizing source).69 This theory about Varro’s Logistoricus pius de pace has not been accepted unanimously,70 but the basic motif of a cosmic (as ‘physical’) foundation for peace and order in the principle of immanent equipoise in the struggle of day and night (marking the seasons by the shift in the length of days) also left further traces behind. It is a matter of a complex of associations that can be traced back— together with the fundamental significance of the idea of isorrhopia (equipoise)—to the beginnings of Greek thought.71 One can at least suggest that the well-known ‘fragment’ of Anaximander (DK 12 A 9), according to which being is involved in mutual exculpation (with tisis and dike, vengeance and justice), was already inspired by the recurring phenomenon of the ‘equalizing’ of opposites

64

Fuchs 1926: 110, on Euripides, Phoenissae 541ff. (See also Sophocles, Ajax 669ff.) Fuchs 1926: 95: ‘Peace was regularly conceived here as an orderly state, in which relations were in balance’. A Stoic idea of ‘cosmic peace’, which influenced the new Roman understanding of peace, is also adopted in Thome (2000: 96). 66 Fuchs 1926: 97f. The basic idea here: moderating the extremes of the seasons through middling equalization and taking the same share, maintaining the day and night of the year. 67 Philo, De specialibus legibus 4. 232. On the equinox, which fights pleonexia of light or of darkness: see Cleomenes 1. 4. 235 p. 26 Todd. On this passage (= 70 Ziegler), see also Reinhardt 1926: 123ff. 68 Philo, De specialibus legibus 232 (cited by Fuchs 1926: 124f., who also emphasized that Philo is explicitly citing the older researchers on nature). See on ‘Kosmifizierung’ of the nomos akinetos also Pseudo-Aristotle, De mundo 400b ff. On Philo’s cosmic interpretation of Augustus, see Peterson 1935: 28f.: on his Legatio ad Gaium 147, where, following Plato’s Timaeus, the action of the demiurge in creating order and disorder can be compared with the politically analogous action of Augustus. Peterson points out that here the ‘political reinterpretation’ of the demiurge’s action is based on the similarity of meaning between taxis and politeia. 69 Fuchs 1926: 142ff. 70 Somewhat sceptical, Weinstock 1960: 46. See the overview in Zecchini 1985: 109f., with n. 4. 71 Abundant material on the subject in Vlastos 1995 (originally 1947). 65

The Peace of Augustus

365

at the equinoxes.72 The idea of an ideal ‘balance’ which controls the ‘extremes’ of the seasons in a harmonizing middle, is already to be found in Homer: on Scheria among the pious Phaiakians fruits grow not just once a year (and then all at the same time), but constantly and without interruption throughout the whole year (Odyssey 7. 117–26).73 This motif of the blessed Eukrasia is once again taken up in Iambulos’ Utopia: on the ‘islands of the sun’ ‘equinox’ prevailed the whole year over, at the equator (Diodorus Siculus 2. 56. 7).74 The Isorrhopia (which is realized temporally at the point of equal day and night) counts above all as an attribute of perfection in itself, of an exemplary completeness whose reality is more complete and more ‘dense’ (because impartially applicable).75 So it can also designate the akme of a development76 as well as establish the supremacy of a people, who live in the decisive ‘Omphalos position’ of the ‘Middle climate’.77 Whoever was located ‘in the middle’ of the world was providing the telos, which gave history its 72 Lloyd 1966: 212 n. 4: ‘the cycle of the seasons probably provides the best example of an interaction of the type which Anaximander’s fragment appears to describe’; Vlastos 1995: 81 with n. 154; also, 60 with evidence on equinox-isomoiria. On the question whether Anaximander could accurately estimate the equinox, see the sceptical assessment by Dicks 1966: 31ff. 73 Svenbro (2000: 347 n. 39) observes, following a hint by Vernant, that there is a suggestion in the description of Elysium (Odyssey 4. 563–8) of an ‘équinox du printemps éternel, dont tout excès saisonnier est éliminé’. Pindar, by the way, picked this up: ἐσθλοί (the Good) will lead a life without effort in the afterlife: ἴσαις δὲ νύκτεσσιν αἰεί, / ἴσον ἐν ἁμέραις ἅλιον ἔχοντες (Olympian 2.61ff.). According to Wheatley (1971: 428), in ancient China in the Emperor’s palace, there was the place of the world-axis (established by specialists) where, among other things, the four seasons mixed together and yin and yang harmonized. At the same time, it was thought a matter of dogma that there was in that place a gnomon that at the summer-solstice threw no shadow at mid-day (just as at the equinox)—a belief that applied to Jerusalem also (see Wheatley 1971 for parallels from the Indian and Iranian areas). 74 According to Vlastos 1995: 60, there is here an attempt to establish a causal nexus between astronomical theory and climatic isomoiria. The opinion that the climate must be moderate at the equator was discussed by Cleomedes (1. 4. 90–131, pp. 21–3 Todd) as the opinion of Poseidonios—on this, Abel 1974: 1076f. 75 See Vlastos 1995: passim, in advance on ‘cosmic justice’, which shows itself in the balance of opposites; Ryffel 1949: 13, 182 n. 342 (on the image of Libra in association with the balanced (mixed) constitution). On the association of ‘symmetry’ and ‘opportunity’ (kairos) (as the moment in which the symmetry is realized) with the image of Libra by Lysippos (after Polykleitos), see Knell 1986: 179f. 76 Schmid 1998: 62. That ‘above all in the Augustan period “symmetry” was not just used as formal device’, ‘but was particularly emphasised to demonstrate certain material and ideological or even cultural-political ideas and demands’: Knell 1986: 170f. 77 See Aristotle, Politics 1327b; Plato, Timaeus 24c; Epinomis 987d; Overview in Hübner 1984: 229 n. 299, where rightly Virgil, Georgics 2. 136ff. (the laudes Italiae) is added to this association. On eucrasia: Euripides, fgt. incertum 981 (Nauck), and also a fragment of Aristophanes from Athenaeus 9. 14 (Athens regularly enjoyed the fruits of all the seasons).

366

The Peace of Augustus

unity, a spatial counterpart. He turned the pivot of time into a pivot of space. The concept of a centre to the world made a convergence between cosmos and history possible: cosmos was a space whose movement around a motionless centre was time. It should be pointed out that for its part the sign of Libra—not yet long in use by Augustus’ era78—is in fact of ‘calendrical’ origin.79 So it ‘signified’ a constellation at best as a secondary consideration: first and foremost, it was the apt symbol for the equinox that balances and evens out.80 Libra is the iustum sidus81 and endows freedom, for Illic perpetua iunguntur pace diebus / obscurae noctes; aequo stat foedere tempus (Manilius 3. 309f.) The equinox also ‘creates’ the universal norm of a ‘mundane’ time in the so-called ‘equinoctal hours’ (nowadays this is the normal equal hours for day and night, independent of the length of day).82 And this is what the equinox stands for—as a guarantee, as it were—for the arrangement of differences, for the connection between the mundus and its homonoia, homologia, harmonia, isomoiria, which shows itself in the krasis of all differences (Pseudo-Aristotle, De mundo 396f.).83 For the central norm is also both the starting point and

78 See Gundel 1926: 117ff. The first Latin evidence is Varro, De lingua Latina 7. 14; for the possibility that Varro was the first Roman to have replaced the older name Chelae with Libra, Le Boeuffle 1983: xxvi n. 4. 79 Boll, Bezold, and Gundel 1926: 52; Gundel 1926: 119 (who also mentioned a possible origin from the ‘Hypsoma-Lehre’: Saturn is considered to be ‘elevated’ in Libra, with which the judge’s scales can be associated). 80 Varro, De lingua Latina 7. 14: Signa quod aliquid significent, ut libra aequinoctium (‘They are called signa because they signify, as Libra does the equinox’). 81 Manilius 3. 433: ad iustae sidera Librae (to the stars of the just Balance); 3. 305 iusta Libra. On iustus in the old Roman sense, Klingner 1965: 625: ‘He who adheres to ius and iura, to the rules that exist for the conducting of conflicting claims, to the ceremonials, in which such a conflict can be played out in an orderly way . . .’. Seen like this, Nature herself was iusta, for the Sun, itself full of religio, just made the strict observance of those ‘Rules and Ceremonies, which made life moral instead of immoral’ (Klingner 1965) vividly clear. The balance is the sign, just as the instrument, of justice; it was under this sign that Servius Tullius, who is said to have invented the scales, was also born (Manilius 4. 209ff.—see Gundel 1926). According to Detienne (1967: 37ff.), the Balance belongs to the mythical King (Minos) and to Zeus. 82 It might be of interest here that the calendar of the French Revolution, its New Year and the New Era of the French (ère des français) began with the autumn equinox 1792 (the day of the proclamation of the Republic, which coincided with the equinox). Thus the entry into the sign of Libra (for Paris), was calculated to the last second and from then on every year started at midnight exactly at the time of the entry into the sign of Libra (Sun at zero degrees, zero minutes, zero seconds) as it fell for the Paris Observatory. It was specifically decided at the Convention nationale that it was all about: ‘la co-incidence de l’année civile avec les mouvements célestes’ (see Maier 1991, 100ff.). 83 See Moraux (1984: 23ff.), who regards this as influenced by Heracleitus.

The Peace of Augustus

367

the point of convergence for all differences, which appear arranged symmetrically or ‘in pairs’ between the equinoxes.84 Indeed in Plato (Timaeus 36b, with the figure of the lying Χ) the (double) location at which the ‘periods’ of self-sameness and alterity diverge and converge was the equinox (as the point of intersection of ecliptic and equator, course of the sun and middle of the world) as the navel (omphalos) of the world that was the omphalos of intelligibility at the same time. At the equinoctial points the zodiac was suspended in the middle of the mundus-sphere: per mediam, mundi praecordia, partem/ disposita (Manilius 3. 61f.). These points were a ‘symbolic extension’ of that fulcrum [Mittelschwebe] of which Aratus sang (compare Phaenomena 22f.), and Germanicus could write that the equinoctial signs held the stars in the vault of heaven in balance.85 Now Lucan (Pharsalia 1. 45ff.) also wanted to have the deified Nero given a central position in heaven so that he would not tip the cosmos to one side; the passage Librati pondera caeli / orbe tene medio (1. 57f.) is indeed fairly confusing from an astronomical viewpoint (because in the process Nero was supposed to look down on Rome perpendicularly).86 So the suggestion made by S. Eriksson, that the equator of the heavens must be meant here as the orbis aequinoctalis,87 seems to be a plausible explanation. The allusion to Manilius and Germanicus are unmistakable,88 the vocabulary of librare, libratus hints at Libra here all by itself, and hence the association with orbis

84 Varro, De lingua Latina 9. 23–9 on the principle of grammatical analogy, which he sees as grounded in the Cosmos (see 9. 23: si enim usquequaque non esset analogia, tum sequebatur, ut in verbis quoque non esset (‘for if the analogy had not existed somewhere, then it follows that it would not have existed in the words either’). He takes the symmetrical structure of the world (the tropic circle and the polar circle with the equator as the centre), as evidence for the reality of the linguistic order. He also speaks of a similitudo gemina (26) (twin similitude) which he wants to detect in the movement of the tides, the pairing of the sexes, and the limbs of human beings (29). 85 Aratus, Phaenomena 499f.: Bis redit haec facies, librat quae sidera mundi / cum ver fecundum surgit cum deficit aestas. See Eriksson 1956: 93. Bayet (1939: 167) remarked quite rightly that Manilius’ vocabulary, when he comes to speak about the equinox, takes on a striking insistence and clarity. It was a favourite theme in poetry (Gundel 1926: 124); for the poetics of Libra see Hübner 1972. 86 See Arnaud 1987. 87 Eriksson 1956: 88–95, esp. 91. According to Arnaud 1987: 185, one should understand here an ‘équateur de l’oecumène’, which passes through Rome. 88 Eriksson 1956: 93: amongst others, Manilius 4. 457ff. (the Augustus passage: felix aequato genitus su pondere Librae; ‘blessed is he that is born under the Balance’s equilibrium’). In addition to the evidence offered by Eriksson see Manilius 1. 173 (librato pondere); 2. 921 (suo libratum examine mundum). Of which the latter passage is not referring to the Balance, but to Venus.

368

The Peace of Augustus

medius is also obvious.89 And the hint at pax can follow here without explanation: pax arises from the ‘central positioning’ of Nero (pax missa per orbem . . . : Pharsalia 1. 61). The ‘construction of the centre’ is a subject on its own, and there is sufficient anthropological documentation for it.90 Of special interest is the phenomenon of an ‘axiality’, which was constructed through ‘cardinal’ orientation of space in an almost ritualistic fashion.91 Hence as a rule the capital (as the caput mundi) and within it the royal residence in particular, was meant to constitute the ‘quality of the centre’ of the umbilicus mundi. So conceived the omphalos was not only the point at which the heavens were nearest (as a kind of ‘linking place’), but also the central axis of the world, the cardo, must run through it—and this was the pivot92 of ‘reality’.93 Obviously such a centre of the world was considered to be transportable in principle—it could even be duplicated;94 in this way the palace in the centre (and as centre) of the ‘residence’ corresponds not just to the pole star eventually,95 but is also a kind of symbolic equator—for this was where the omphalos was to be found.96 This ‘mythical equator’ is just the place that Nero is supposed to have occupied—beyond that, its ‘qualities’ could seemingly be transferred to the king; according to an ancient Chinese source the latter is supposed to have reigned as the pivot of the world, and from such centrality he guaranteed the ‘peace of all countries’.97 Hence the characteristics of the ‘centre’ (the main city, capital, or residence) are projected onto the king himself, who is supposed to embody the ‘whole of the world’ as a city or place, according to the principle of pars pro toto.98 Fundamentally the spherical model view of the world in Greek cosmology and astronomy was a comparable construction of centrality or

89 Eriksson by the way wanted to see in the round a replica of Virgil’s catasterism of Octavian (Georgics 1. 24ff.) ‘in the Balance’. 90 See Wheatley 1971; Geertz 1977. 91 92 Wheatley 1971: 423ff. Ibid. 428ff. 93 Ibid. 444: ‘the capital, that point of absolute reality about which the world revolved’. 94 Ibid. 417f.: as long as it was ‘an attribute of existential rather than geometrical space’. 95 Ibid. 428: the Kaaba at Mecca should also lie ‘under the Pole star’. 96 As to the middle of the world in Delphi, Varro (De lingua Latina 7. 17) objected: this could not really be the centre of the globe. 97 Wheatley 1971: 430, see also 434: the king can incorporate centrality in his person, because he carries weight as the guarantor of axial transmission. He is (see Schmid 2005, ch. 2) the guarantor of a ritual construction coming from the centre as a point of reference for unity and cohesion. See Grottanelli 1987: 313: ‘often his abode and his person are symbolically indicated as microcosm or as the center of the cosmos’. See Geertz 1977: 157 on Queen Elizabeth as ‘center of the center’. 98 Wheatley 1971: 431ff.

The Peace of Augustus

369

the centre, insofar as it enabled the ‘fulcrum’ of human habitat (medio suspensa manet: Manil. 1. 180)99 to become rationally and geometrically plausible. In this ‘worldview’ it is the very principle of the symmetry of the mundane fundaments that is of major importance,100 and the ‘spherical basis’ of this mundane symmetry, which obviously coalesced around the equinox in the rising and falling pendular movement of light, was entirely evident.101 This became visible in the phenomenon of the strictly proportional waxing and waning of the length of the day in two respective zodiacal months: thus the day increased under Capricorn by the same length as in Gemini, just as Aquarius and Taurus corresponded, and Pisces and Aries. In the waxing, on the other hand, Cancer and Sagittarius, Leo and Scorpio, Virgo and Libra correspond to one another.102 The same proportionality of a divine geometry manifested itself (precisely for astrologers) in the times of rising and setting of the signs: the signs of Capricorn and Gemini, Aquarius and Taurus rise and descend at the ascendant and descendant respectively in the same span of time.103 The shadows of a meridian enabled one to perceive this heavenly consensus of indestructible harmony in the strict rhythm of the growth and decline of their lengths. Fig. 14.7 shows the arrangement of these correspondences: horizontally the equinoctial axis (here 0 Aries—0 Libra), upon which the time values are ‘mirrored’,104 vertically the axis of the solstice, which separates the rising and falling signs. Now when one places the lines of correspondence of this figure straight, one gets the structure of the so-called ‘Sun Dial’ (see Fig. 14.6): the signs (in relation to their order) to the left and right of the line of the solstice (= midday), each have the same shadow- or day-lengths as they rise and fall. It is these very correspondences that the device excavated by Buchner made visible (as a meridian). If in this way the simple interplay of the circular kinesis of the sun’s course in the shadow of a gnomon fabricated the spherical harmonia 99 See also Pseudo-Aristotle, De mundo 391b13 on the life-bearing middle of the world (the Earth), as ‘Mother and Hearth’. 100 For symmetry in the Platonic Cosmos see Schmid 2005: 133 n. 57: see also, generally on symmetry in antiquity, Ryffel 1949: 150 (in the political context of the ideal balance); Hommel 1988: 44f.; Knell 1986. See also Hyginus. 1. 6. 1–3, pp. 8ff. Le Boeuffle. 101 For example, Cleomedes. 1. 3. 2—p. 14 Todd; p. 16 Todd and passim; Hyginus 4. 3. 3—p. 120 Le Boeuffle. 102 Manilius 3. 468–82. For illustration, see Goold 1977: Introduction lxxx (fig. 22) and lxxv (table 2). 103 Manilius 3. 395ff. 104 It is moreover clear that the astrological doctrine of the mirror-points (antiskia) refers to these symmetries, which, according to Boll, Bezold, and Gundel (1926: 102), should go back to Hipparchos.

Fig. 14.6 For discussion of Fig. 14.6 see: page 369.

Fig. 14.7 For discussion Fig. 14.7 see: page 369.

The Peace of Augustus

371

mundi as a paradigm of ‘order’, then ‘centrality’ was evoked here at levels that were both thoroughly rational and at the same time ‘solar’. But the world as proportioned in this way was arranged ‘geocentrically’ around a centre that (rather paradoxically) contradicted the ex-centric hierarchy of a world-model that placed ontological superiority into the spheres. If here the equinoctial line (associated with the Ara Pacis) marked the ‘middle’, a contradiction emerges, in which the cosmological dogma frayed at the edges: perhaps it suddenly became obvious to the author of the De mundo that the true centre that guaranteed ton holon harmonian kai soteria, could in no way be the ‘lesser’ earth (in the sub-lunar field), but must be sought above, in the zone of heavenly brightness.105 Precisely because the ‘actuality’ of mundane substance only befitted the bright shell, the centre as the focal point of reality was to be extrapolated: the centre had to be hypostasized at the periphery of heaven first (because it was what surrounded the sphere of heaven) in order for it to be ‘reified’ then, as the ‘capital’ place. The true omphalos was out there, because only there (in the field of superior intelligibility) could it be perceived—indeed on earth it was everywhere and nowhere, that is, without a place. Hence the spot where the equator and the zodiac intersect was most suitable to represent the ‘omphalic’ quality: where both spheres of mundane circulation and periodicity converge (the sign of Libra having semantically absorbed the equinoctial quality of the spring equinox). On earth the centre of the world was the projection of the heavenly centre, and since the centre of the earth was unaccessible, ‘equinoctiality’ was the given, and indeed the most rational concept for evoking the ‘cosmic centre’; for the same reason Mithras too had to have his ‘seat’ right there at the equinoxes, which are actually not ‘places’ at all.106 The centre around which the length of the days in the year revolved was the navel of time that could be conceived geometrically as a mundane structure. It was the same structure that was arguably meant with the lyre of Apollo in the pseudo-Heraclitean Skythinos-fragment:107 the son of Zeus, who holds the beginning and end of everything, put together (harmozetai) this lyre; he used the glittering light of the sun for a plectrum.108 This motif: Apollo (here potentially as the solar Apollo) harmonically intonizing the ‘togetherness’ (unity) of the world and the sun109 (acting as

105

106 De mundo 399b. 33ff. Porphyry, De antro nympharum 24. Herakleitos, DK 22 C, 3. 1, after Plutarch, De Pythiae oraculis 402A. Adapted by Kleanthes, SVF 1. 502. 108 echei de lampron plektron heliou phaos. 109 It is possible that Varro too expressed this point in writing: Boyancé 1936: 100, 105, with n. 5 on Varro’s onos lyras. 107

372

The Peace of Augustus

the ‘plectron’, so to speak, of appearance), is to be interpreted most meaningfully in view of the point of the sun’s rhythmical movement around the year-points.110 P. Boyancé referred to an Orphic hymn with a related motif: Apollo mixes the seasons together harmonically (isemerically).111 The meridian on the Field of Mars could be read as this lyre of Apollo, as a mundane paradigm to boulomeno horan. In this lyre the equinoctial line was the mese, the middle string, on which the tuning of everything depended,112 it was granting a harmony that vouchsafed life, and whose patron was Apollo.113 And now the very presence of Apollo as an element of the visual programme of the Ara Pacis Augustae has been aptly emphasized.114 Through the ‘tendrils of Apollo’, with their swans arranged in symmetrical order on the orbiting relief, there descends (according to H. P. L’Orange) ‘the divine harmony of Apollo on earthly nature: therefore the regularity and the clear symmetry of their appearance’.115 The celebration of a peace, which was set up as a mundane foedus temporum by Apollo, the lord who held the centre and the unity of all that was visible, being also conceivable as a ‘physical’ principle, would have entirely fitted the mythico-rational ambivalence that the Augustan imagination also demonstrated elsewhere.116 And the

110 Boyancé 1936, 90ff. His attempt to understand the middle position of the sun in the planetary spheres is misconceived (99). On the harmony of the spheres with the sun as mese, see also Burkert 1962: 297 n. 122 (which would have been inspired by Eratosthenes); on his epic poem ‘Hermes’ (about the harmony of the spheres, which should reflect the harmony of Hermes’ lyre) see also Hardie 1986: 10. 111 Boyancé 1936: 91: Hymn XXXIV, beforehand, 21: mixas cheimenos tereos t’ ison amphoteroisin. See also De Grummond 1990: 672 (Apollo as the master of the order of the seasons). 112 Boyancé 1936: 98. 113 See on the earlier form of comparison, dear to the Pythagoreans, of the structure of the universe, the state, the family, and the human race with the structure of an instrument such as the lyre Moraux 1984: 667, 678ff., with evidence of the musical analogy between world-order and politics, a subject that was also favoured by authors who ‘owed so much to the Platonic-Aristotelian tradition’. See for example Archytas, De legibus 35. 21–30: the effect of law is similar to that of the sun in the world; similarly, the eukrasia ton horon resembles eunomia (and ibid.: referring to the ‘kitharödischen nomoi’). 114 See for example Pollini 1993: 196ff.; Castriota 1995: 60ff., 90ff.; and others. On the geometrical limiting of the proliferating tendrils as an Apollonian feature, see also Sauron 1988: 34ff. 115 L’Orange 1991: 259 (originally 1962). See Castriota (1995: 66), who speaks about the swans’ ‘paracletic’ and ‘theonymic’ character, which, however, he wants to apply to Dionysus and Venus as well. 116 Brendel 1979: 189 (referring to the Gemma Augustea): ‘In Rome apparently the allegories of this description were an Augustan innovation; their possible antecedents, for instance in the aulic art of Ptolemaic Egypt, are as yet insufficiently known. Their original

The Peace of Augustus

373

associative connection of the Ara Pacis with the equinox (which is in fact implicit in Buchner’s interpretation) as a reference point of symmetry and analogy could be a key to the markedly symmetrical stylistic elements of this monument, whose exactitude in their connection with the ‘lush abundance’ (in the tendril- and fecundity-symbolism) was ‘irritating’ to P. Zanker, so that he was moved to comment on ‘a yearning of the Augustan period for order and law that seems almost neurotic’.117 So above all a connection between freedom, centrality, and abundance under the aegis of Apollo, the cosmic Kitharode, would thus reveal itself—a connection in which the motif of a ‘cosmic’ peace, which is also associated with ‘natural’ consequences, is substantiated in multiple ways.118 And in this regard the rather unusual extension of the so-called floral zone of the Ara Pacis119 certainly indicates that the Peace of this altar was expressed in dimensions that we would designate ‘natural’— indeed the parallel between the (historically tangible) procession-scene and the zone of vegetation120 (which always emerges symmetrically, each from its root) points to an analogy in disparity: of the human, vegetative, animal, and divine domains. Thus it stands as an invitation to see in view of the unifying frame of the cosmos a possible play of analogies that was related to a third, whose physico-divine status contained an element of dialectical agreements. Hence among other things a connection was to be made between history (which in fact did not in itself point towards monarchy) and a mythical place (the only one that fitted a monarch as theme was the Roman institutions under the Empire; their general tendency, to transpose the actual political body into an image of semi-mythical order functioning as the terrestrial analogue of the cosmic organism’ (cited in Hardie 1986: 2). 117 Zanker 19902: 185. (On the ‘neurotic’ symmetry, see the reductive comment in Galinsky 1992: 466.) On the accuracy and consciousness of the symmetrical design of the tendril-work, see Knell 1986: 173, after which this concept of symmetry became ‘the symbol of the Pax Augusta, of a happily pacified, well-disciplined world’; ‘the visual language of the tendrils represents a microcosm whose middle emerges from a root’ (see on the timecentering kairos above, n. 75); ‘symmetry becomes here the deliberately applied instrument for the official expression of the ideology of the Principate’. 118 Fuchs 1926: 169ff., on the old idea of Peace as the time of blessing and natural abundance; ibid. 189 n. 1, with evidence about the pax of winds, rivers, and other such forces of nature; Castriota 1995: 124–44, for the association of an aurea aetas (Golden Age) of peace with the concept of cosmic renovatio. 119 Galinsky 1992: 463: the floral zone is larger than the figurative one and is considered ‘the largest known application of the motif in antiquity’. 120 See Castriota 1995: 164; Sauron (1988: 37f.) sees in the tendrils the ‘majesté du monde végétal’, which must bow to strict geometry, following a ‘desire for the order, the hierarchy and the clarity of the radiant Apollo—and this Apollonian geometry is reflected in the folds of the garments in the figurative scenes’.

374

The Peace of Augustus

the apex societatis).121 This ‘myth-historical’ feature is thoroughly realized on the Ara Pacis122—but here the nod towards a ‘cosmic background’, to the ‘natural’ foundation of the pax perpetua would have been a significant addition. For in light of this it became clearer what was intended by the new order, in which A. Borbein wished to recognize the timeless ‘presence of the programme realised’.123 Thus the Demiurge (he had with divine profundity folded the world, so to speak, in the creases of time) was also the avowed paradigm of the new Augustan order, which here testified to the more essential cosmic quality of the world. If the equinoctial ‘central symmetry’ of the mundus was to be meaningfully presented in the astrological symbol of Libra, then it is not without interest that in astrology according to the so-called ‘oikoidoctrine’124 the planet Venus was assigned to Libra.125 If we may thus designate Venus126 as the mistress of the autumn equinox, then the room for associations expands: the seated figure of a lady described and interpreted over and over, on the east side of the ara (the ‘Tellus’, also Italia, Pax, or Ceres: see Fig. 14.8),127 with the two putti and the symmetrically arranged ‘veiled girls’ (‘Schleiermädchen’), has been interpreted by Karl Galinsky to signify Venus.128 In connection with this it might be telling (yet has been overlooked, despite H. Nissen pointing it out)129 that in the year 46  Caesar set the natalis of the temple of Venus Genetrix on the very date of the autumn equinox, according to the 121

No doubt the Altar of Peace is a deeply monarchic monument, but it is so in a subtle, one might well say, highly complex way. Formally the princeps was admittedly highlighted by size, presented as citizen-priest among the other citizens in a popular procession, which launched the Peace-cult, and which in an honorific way bore his name. 122 See Borbein 1975: 244; Galinsky (1992: 462) points at similar links between history and myth in Aphrodisias. The link refers to the common occurrence of recognizable historical (contemporary) figures (hence the question as to whether the foundation procession of 4 July 13  was depicted here perhaps misses the point) and mythical reflections, such as Romulus or Aeneas. On the similarity here of Aeneas-Augustus: Castriota 1995: 157 (the ideology of mythic prefiguration), who (p. 167) also asserts that in no other comparable Hellenistic monument is the incumbent monarch so unequivocally portrayed in an ‘historical’ presence as the focus of mythical comparison. 123 124 Borbein 1975: 265. On which see Bouché-Leclercq 1899: 182ff. 125 At the time of Augustus, these assignments were well known (see Schmid 2005: 228 n. 221 on the thema mundi, to which they relate and which was also mentioned by Thrasyllus according to CCAG 8. 3. 100 no. 27ff.); see Abry 1996: 132f. 126 Venus as star seems already to have been linked with the goddess by Caesar: see Schilling 1954: 320, on coins with Venus and the star. 127 Spaeth (1994: 66f.) provides a useful overview of earlier interpretations. 128 1992; 1996: 148f. in the context of his thesis of the ‘polysemic’ readability of Augustan images, according to which a one-sided interpretation in view of the ‘intentional multiplicity of meanings’ would not be fair. 129 Nissen 1906: 335f.

The Peace of Augustus

375

Fig. 14.8 Seated Venus/Tellus/Italia, with veiled maidens, from the Ara Pacis. The so-called ‘Panel of Tellus’.

Julian calendar.130 Caesar himself, who at the time was actually busy introducing the calendar according to which for the first time such a date could be given a fixed astronomical relationship, can hardly have chosen it by chance or without knowing. And already Jean Bayet has pointed out a coin of M. Cordius Rufus,131 which according to M. Crawford dates from the year 46 132 and which on the obverse (the front shows the Dioscuri with a star) portrays Venus with a set of scales in her hand (see Fig. 14.9).133 Manilius usually suggests the associability of Venus with the ‘equilibrating principle’, according to which (II 918–26) Venus allows humans to rule through the countenance she turns towards the world: in the centre of heaven134 between rise and fall (suo libratum examine It is clear, though, that in the calendar of the year 46  26 Sept. did not fall on the equinox. On the dating question, see now Ramsey and Licht 1997: 97, 183f. 131 Bayet 1939: 159; he dated it to 49  (no sources given). 132 Crawford 1974: no. 463/1a: on the assignment, see 474ff. where it is considered that the Venus-type as displayed might be derived from the cult-image of Venus Genetrix. 133 On this motif (Venus with balance, with further references, as ‘the iconographic type of the Caesarian Venus’) see Abry 1996: 133f.: ‘an astrological interpretation of the Caesarian Venus’ (by S. Reinach in 1917), would be ‘unanimously rubbished by modern commentators’; it is seen as an allusion to Caesar’s iustitia (for the relevant evidence, see below n. 37). 134 Manilius wants to assign Venus as an attribute of the middle of the heavens (medium coeli). 130

376

The Peace of Augustus

Fig. 14.9 Denarius issued by moneyer Cordius Rufus, 42  (RRC 463/1a (see pl. LIV). Obv., showing Venus, carrying Cupid on her back; for discussion, see vol. 1, pp. 473–5.

mundum). Here Venus signifies a ‘middling’ symmetry: in the culmination of the day’s arc she balances the Orient and the Occident—that the world (and with it humankind) receives divine patronage must be connected to this (proportional) ‘balance’ (which is precisely what the image of the horoscope conveys as ‘structure’); the sign of Libra is alluded to by association. Because Venus could be associated with Pax (by means of the tertium of the equinoctial harmony symbolized by Libra) it allowed, probably in Caesar’s footsteps, new forms of expression, even of the ‘Julian’ components as well (which are present with Aeneas) in the peace complex on the Field of Mars:135 here Venus also became a cosmic component of the ‘Julian world-mission’ as the genetrix both of Julius and of Rome.136 In the ‘polysemy’ which according to Galinsky137 is precisely what makes the Augustan imagery accessible in several ways, a harmonious association of concepts such as ‘centre’, ‘reconciliation of contradiction and 135

The Campus Martius indicates that the god Mars (because war is afflicted with death) must remain outside the pomerium (Magdelain 1968: 62f.; Rüpke 1990: 36ff.); so Augustus with the temple of Mars Ultor brought war into the peace-zone, so to speak. He also afforded a new accent (Rüpke: 1990) to the Campus Martius with (of all things) the Altar of Augustan Peace. War and peace came together in a field of new overarching (Apollonian) unity. 136 See the well-known inscription of the Ephesians, which honours Caesar as worldsaviour and as son of Ares and Aphrodite (Dittenberger, Sylloge 760). On the astral associations of the ‘transmitters of culture’ there are anthropological parallels. So Gundel (1922: 180) reports myths of the marriage between mortals and stars. The home-comer from the star-world brings weapons, cultural improvements, or secrets from the land of heaven. So might an astrally interpreted Anchises saga (with a glance at the starry Caesar) refer to a star-born Julian clan. Thus Rome by itself can be considered through the story of Aeneas as the star-born folk of the Aeneads (and thus the Julian ancestor becomes the mediator of the indestructible quality of descent and destiny). 137 Galinsky 1992: where (472) the connection between semantic and political ‘multivalence’ is emphasized, with a reference to J. Béranger (1953), who stressed the respective fuzziness of political concepts.

The Peace of Augustus

377

Fig. 14.10 Augustus’ horoscope in Rome on 22 September 63 , c.5”35 (System B).

division’, ‘Apollo’, ‘symmetry’, ‘cosmos’, ‘Venus’ (Julian and planetary), ‘Peace’, and also Tellus were brought together without compulsion through the ‘equinoctial association’—there remains ‘Italia’ (as far as the possible meanings of the so-called ‘Tellus-figure’ are concerned). And according to Manilius (IV 768–77) Italy—as the only place in the world—is supposed to be governed by the sign of Libra: Quod potius

378

The Peace of Augustus

colat Italiam, si seligat, astrum / quam quod cuncta regit, quod rerum pondera novit, / designat summas et iniquum seperat aequo, / tempora quo pendent, coeunt quo noxque diesque? / Hesperiam sua Libra tenet, qua condita Roma / orbis et imperium retinet discrimina rerum / lancibus et positas gentes tollitque premitque, / qua genitus Caesar melius nunc condidit urbem / et propriis frenat pendentem nutibus orbem.138 As there was no blueprint for such an attribution, it has been suggested that here Manilius alluded ‘to the horoscopes of the city of Rome and of Augustus’.139 But then the same question arises for that horoscope of Rome, in which, according to Tarutius Firmanus not only the moon was supposed to be in Libra, but also the sun.140 If one can assume that Tarutius Firmanus perhaps even took the equinox as the birth date of Rome,141 then we must ask what was so attractive about this association of Rome and Libra (even for Tarutius and Varro). And we may simply suggest here a ‘translation’ of the concept of a blessed eukrasia into astrological terms. ‘Libra’ was the astrological version of the idea of a civitas populi Romani that the heavenly spirit egregia temperataque regione conlocavit, uti orbis terrarum imperii potiretur142 (i.e. that Rome was placed in a ‘temperate’ region in order to rule the world). Through its ‘nativity’ Rome was destined for a hegemonial ‘omphalos position’. These qualities of the Romans did not, however, inhere in them

138

See also the harmonious mixture at Manilius 4. 719ff.: Martia Romanis urbis pater induit ora / Gravidumque Venus miscens bene temperat artus (‘the Father of the city gives the Romans Martial features; and Venus sharing with the War-god fashions their limbs in proportion’). Likewise, for the role of Rome in uplifting and lowering to do justice see Virgil, Aeneid. 6. 852: pacique imponere morem, parcere subiectis et debellare superbos. Housman wanted (1913: 112) to recognize Tiberius as the Balance-born Caesar (because he had the moon in the Balance), but the interpretation with Augustus seems to me substantially more plausible. 139 Gundel 1926: 130. 140 See Schmid 2005: 201f. According to Abry 1996: 134f. the association Rome-Libra means that Rome should be seen as the incarnation of Justice. 141 Grafton and Swerdlow 1985: 459. 142 Vitruvius 6. 1. 11; see Vitruvius 6. 1. 10: ‘The Roman people has boundaries in the middle of the world’. See also Pliny, Natural History 2. 190 (only in the climate-zones of the media mixtura can great empires arise); 3. 39 (Rome appointed as the Fatherland of all peoples); Virgil, Georgics 2. 146ff. The Poseidonian origin of the passage of Vitruvius about Rome’s imperial position at the omphalos has been contested by Ferrary (1988, 382–94); he prefers to ascribe the idea to Vitruvius himself or to Varro. Schmidt (1980: 26ff., 54ff.) is committed to the Poseidonian origin. Poseidonius produced the ‘people of the Middle’ by combining Platonic theory of hierarchies with Aristotelian construction of centrality; see also Schmidt 1980: 60ff. on Strabo 6. 4. 1, where Italy is constructed as the Middle, through which ‘a preference for Italy understood teleologically, and a special disposition for political hegemony in practice is derived from this’ (p. 65).

The Peace of Augustus

379

genetically, but were associated with the climate of their habitat.143 The Romans were a ‘Volk ohne Eigenschaften’, a kind of ‘epitome’ of human potential that balanced the extremes of character (southern craftiness and dumb northern fighting-spirit).144 Hence Rome was also the ‘symbolic equator’ in which the cosmos as an arrangement that in itself was symmetrically balanced and pacified now turned into ‘history’ and into a ‘genetic’ (and providential) quality of a society. Rome-Italia was meant to ‘pacify’ the world, because the world itself was the oscillation around an orderly converging centrality. The undisturbable symmetry of the kinesis of this world was in itself the epiphany of a ratio mundi: the new regime professed the Apollonic brightness, embodied through the ‘one man’ who in a deeply meaningful way was designated by his horoscope according to ‘System B’145 through the iustum sidus of Libra. In this man cosmic providence and a historical mission146 of the Italian soil, which had always been pregnant with imperium, achieved final realization.147 An arrangement that in such a way linked together the peace of Augustus with the eternal ordering of time through the harmonic equilibrium—through a ‘heavenly peace’, in the ancient Chinese style— would also explain the absence of Pax (which for de Grummond was the basis of her explanation), if one hesitates to interpret ‘polysemically’ the 143 Rome is bound up with the heavens: Manilius 4. 694–756. See also, Virgil, Georgics 2. 460, on Italy as iustissima tellus (the most just of lands). It is important to note that according to ancient ideas it ‘was above all climatic influences that determined the cultural development of a people’ (Glei 1991: 341); see Cairns 1989: 125ff. on the ethno-geographic tradition, esp. 125 with the remark that this tradition recognized ‘the determinant role of terrain over race’. Thus ‘genetics’ is decidedly orientated towards the mundane. 144 On the double nature of Italy as potens armis et ubere glaebae (powerful in weapons and in the soil’s riches; Virgil, Aeneid 1. 531) e.g. Glei 1991: 74; in mythical terms, this means the double lineage of Mars and Venus (whose agreement the Augustan monuments readily emphasized) where you can add that the corresponding planets controlled the two equinoxes (Mars being oikodespotes of the sign Aries and so of the spring equinox). 145 See Fig. 14.10, the horoscope of 22 September, in the ‘Libra-Version’. (The most likely candidate of possible versions used by Augustus: Schmid 2005: 19–30.) 146 On Poseidonios’ supposed description of the historian as the servant of providence see the points of view (pro and contra) in Schmidt 1980: 73f. n. 38; likewise, on Poseidonios as the herald of a cosmic historiography, Malitz 1983: 416. On the Virgilian interpretation (after Georgics) Glei 1991: 72: ‘The divine selection of the Virgilian homeland of Italy offers an incomparable chance of salvation’. On the Virgilian ‘ideology of Italy’ (which finds its monumental expression in the Tellus-relief of the Ara Pacis) see Glei 1991: 65, 268ff. 147 Virgil, Aeneid 4. 229: Italy is gravida imperiis (pregnant with empires). On the arrival of Aeneas at Pallantium, the simple country spot where Roman power shot heavenwards (quae nunc—so Virgil looking back—Romana potentia caelo aequavit: Aeneid 8. 96–101) and with which, according to Binder 1971: 40, Roman history started; further discussion in Schmid 2005: 373ff.

380

The Peace of Augustus

sitting figure with the putti as Pax.148 De Grummond laid particular stress on the flanking ‘Schleiermädchen’, usually taken to be aurae,149 which she now wants to see as horae.150 This explanation has been partly accepted;151 and certainly the personification of the seasons is not unfamiliar in the Augustan period.152 If, according to Wilhelm Gundel, Venus was also taken to be ‘double-faced’153 (she was both the morning and the evening star), perhaps the two guises might also signify the hours of morning and of the evening (the one on the left on the swan seems to be rising, the one on the right on the ketos is portrayed as falling: for the moment balance reigns, perhaps as at dawn).154 Indeed the association suggested here is compatible with aurae-interpretation: it would emphasize the spatial, not the temporal aspect of ‘blessed eukrasia’—both aspects are connected by ‘cosmic’ analogy.155 Therefore the ‘horae-interpretation’ would be conspicuous because here a (figure of) Eirene comes even closer to Hesiod’s interpretation (Eirene as Hora is the daughter of Zeus and of Themis, together with Dike and Eunomia and the Moirai: Theogony 901–4). That would be an apt illustration of the fundamental paradigm shift that had reversed a development in the course of which Eunomia, to the extent that she belonged to the Greek polis and its selfconsciousness, had disposed of her fatal sisters.156 That it was precisely the sign of Libra that could give occasion to reflect on how the world was ordered is proven also in late antiquity by the De ratione Librae of Ausonius.157 One learns there that in the

148 De Grummond 1990: see e.g. 663: ‘Pax thus appears here as a goddess for all seasons in a rich cosmic milieu’. 149 See Spaeth 1994: 67 n. 14; it might be remarked that de Grummond (672ff.) and Spaeth (88) think of astrological connotations here, referring to the imitating ‘Carthagorelief ’ (Galinsky 1996: 150f.). Tentative interpretations stretch from Moon-Diana to stella Veneris. 150 De Grummond 1990: 670. 151 Galinsky 1992: 459 (who wants to link them with Venus); Castriota 1995: 19f., who associates this with Apollo as ‘lord of the solar year’; ibid. 66f., where the possibility is considered of an older concept of horae that would subsequently have been modified into aurae. 152 See Simon 1986: 139f.; Long 1992: 487f. 153 Gundel 1922: 33, following Dorotheus of Sidon (CCAG II. 82). 154 I wonder whether the interpretation as nyx and hemera would not be possible; notice the poppy-capsules on the side of the figure descending into the sea on the ketos; Virgil, Aeneid 2. 250: vertitur interea caelum et ruit Oceano nox (meanwhile, the sky is turned and night rushes into the sea). For nyx and hemera as personifications, see Walde 2000: 1076f. s.v. Nyx; Bernert 1937: 1663–72 s.v. Nyx; Jessen 1913: 230ff. s.v. Hemera. 155 See Beck 1998: 338 n. 15: ‘In the heavens, space and time (annual time, that is) are precisely interchangeable: “when” is “where” and “where” is “when” ’. 156 157 On this, see Ostwald 1969: 63f. See Schmidt 1991.

The Peace of Augustus

381

so-called ‘vault allegory’ (‘Gewölbegleichnis’) Libra could be taken to be the crowning keystone of the universe.158 For it was the lapis fornicationis,159 which according to the De mundo of Pseudo-Aristotle160 was also to be designated as the omphalos lithos (centre-stone) which could have been part of the explanation of the image of a balanced cosmos that was ‘poised’ and supported itself: it supports everything in the crowning central position by means of distributing its weight to both sides. If the ‘equinoctial centre’ is a plausible allusion to a Peace of Augustus resulting from mundane providence, then the ‘ancient Chinese style’ of this reference to sociomorphic cosmicity (to be deduced here as a suggestion of an image with various meanings) is striking enough. But in ancient China the reference to ‘axial centrality’ was an element of a determined ‘monarchical field’—thus P. Wheatley quotes an ancient Chinese poem to illuminate an ‘astrological mode of thought’: The capital of Shang was a city of cosmic order The pivot of the four quarters Glorious was its renown Purifying its divine power Manifested in longevity and tranquillity And the protection of us who come after.161

If the ‘Peace’ of Augustus along with the imperiosa civitas also put the world in order—or, more exactly, if it put the capital society back to its predetermined course as a just ‘cosmicity’ of the world, recognizably in the Apolline centre—then it was now for the first time realized politically as a project of fatally mundane necessity. Such allusive imagery may be taken as revealing the Augustan period’s (or the Augustan order’s) need to see itself as a kind of ‘real-life theodicy’ (‘realexistierende Theodizee’). A divinely ordered mundus was useful orientation: the furthest past of a Saturnian piety162 (and so the priscus Romanus moulded by it) was everpresent as an ideal—just as, according to Aratus (Phaenomena 135), the virgin Dike was long gone from the world, though still present and visible 158

Ibid. 428ff. Seneca, Epistles 118. 16: unus lapis facit fornicem, ille qui latera inclinata cuneavit et interventu suo vinxit (one stone creates the arch, the one that wedged up the inclined surfaces and dominated by its intrusion). 160 161 De mundo 399b 30ff. Wheatley 1971: 450. 162 On the contemporary life of the countryside ‘as an image of the bygone Golden Age’ in Virgil, see for example Kubusch 1986: 102—following Varro’s idea (De re rustica 3. 1. 4–5) that the farmers are remainders ex stirpe Saturni regis (from the roots of King Saturn), which Virgil through the interpretation of Italy as saturnia tellus (the Saturnian land) localized, so to speak. On the Augustan Golden Age as the model of the aetas Saturni, see Kubusch 1986: 137f. 159

382

The Peace of Augustus

among the stars: hence the heavens mirror the reality of the time of ‘golden justice’. The mundane order was without fault: there was only the profanum vulgus to be despised, to be hated for its folly and greed as the ugly alternative to cosmic imperium Iovis having once more defeated the giants.163 Horace as the herald of a new, noble authority (in whose name it was once more supposed to be dulce et decorum pro patria mori)164 belonged to the Muses. And the latter proclaimed in Hesiodic fashion to altus Caesar (as they did to Jupiter and to the poet) that cosmic ordering of the whole: qui terram inertem, qui mare temperat ventosum et urbis regnaqua tristia, divosque mortalisque turmas imperio regit unus aequo.165

Friedrich Solmsen rightly remarked on the victory of the gods over the giants praised in the Roman Odes that Horace was careful ‘to establish connections between the correct political order and the heavenly ordering of the universe’.166 Hence it is understood that Augustus too, if he saw himself (or would have wished himself to be seen) as the royal priest of a new ritual of mundane power that imparted meaning, could not simply re-conceive Rome as the capital, ceremonial centre.167 He had to fit in with what existed, and perhaps the ‘intentional multiplicity of meanings’ in the imagery of the Ara Pacis would be, following Galinsky, an element of the Augustan programme, which ‘solicits the viewers’ participation rather than suffocates it’.168 So the play of meanings evoked

163

164 Horace, Carmina 3. 1. 6. Ibid. 3. 2. 13. Ibid. 3. 4. 45–8. See also 3. 5. 1ff. on Augustus as an earthly representative of the heavenly Iuppiter tonans caelo; 3. 6. 5 on Rome’s divine dominion: dis te minorem quod geris imperas. 166 Solmsen 1972: 154. For the rest, it should be emphasized that only such a Hesiodic understanding of the Muses would have made it at all possible to have an equation of the poetic with the political, by which Pindar as well here naturally offered a starting-point; see just the first Pythian, in which Apollonian Melos becomes an anti-Typhonic order and in the structure of the landscape the associated elements of a temple are associated (eagle, columns, metope): the landscape itself, with towering Etna, is ‘kosmos’ in one sense, which the polis mirrors as the (humanly conceived) architecture of the temple. Horace’s archaic intonation also has its political sense: he evokes the possibility of an approach of the divinely (and musically) arranged world towards the world of human beings, an approach that was not yet hampered by a radical conception of nature. 167 For that reason, a tendency highlighted occasionally by Nissen (1906–10) towards the cardinal orientation of Augustan temple-alignments (according to the solstices and equinoxes: see 285, 302, 346, 240) is probably not an indication of an effective and targeted element: see the assessment of this by Gros 1976: 147ff. 168 Galinsky 1996: 149. 165

The Peace of Augustus

383

was intentionally associative, indeed the ‘meaning’ (perhaps of a luminosity of the world pacified by the gods, which was ‘given back’ to the society through the victory of a providentially pious mediator) was not simply imposed. The comprehensive message was a matter of subtle and composedly tasteful hints,169 of an ingeniously constructed pattern of floating associations on the Field of Mars—almost a game, which moreover had been designed on various levels of symbolism at the same time (a certain ‘Engführung’, though, was not to be missed).

A FT E R W O R D The main argument of this very slightly revised chapter from my dissertation (Schmid 2005) was part of a more extensive picture, maintaining the conception of a general paradigm shift in antiquity (beginning in the fourth century ) that resulted in a systematic theology of the cosmos (providing also the Greek background of astrology) and in the triumph of monarchy in a field of decidedly anti-monarchic societies. I have extended this argument in more recent works, where I claim a Greek ‘modernity’ having been outweighed gradually by an ‘anti-modernity’ that included new pieties, theology, teleology, monarchy, essentialism, ‘metaphysical’ downgrading of appearance, and the like, all being finally absorbed in the rise of Christianity (Schmid 2020; 2016, 367–438; 2011). The title of the book as originally conceived, failing to impress the publisher, included ‘an element of political theology of the early principate’. It was meant to be about the problems of creating a monarchy within a society that lacked important ‘monarchical’ premises (as, most obviously, the mere acceptability of a ‘rex’). The chapter on the Ara Pacis wanted to show the connected elements on the Field of Mars as parts of a meaningful story (accepting not the conclusions of Edmund Buchner but some of his basic premises). My interpretation of the whole ‘arrangement’ did not provoke a great response: I had some correspondence on the subject mainly with specialists in the astronomical issues, some of them obviously having tried to communicate with Buchner without receiving an answer (e.g. Frans Maes; see Maes 2005). But I was mostly concerned with the question of monarchy: with the almost unique case of having to ‘invent’ monarchy from the inside of a society being distinctly averse to it. Kingship as I conceived it (see Schmid 2005, 65–91) was in need of a cosmology, and it needed to be a cosmology that was connected to the imagery of a society’s origins and aims. What was at stake, then, for an inventor

169 An almost ironizing distance is also evident in the small animal-scenes of the floral zone and also in other almost imperceptible asymmetries: ‘Nature is ordered but not excessively so’ (Galinsky 1996: 152).

384

The Peace of Augustus

of monarchy (and supposing that Augustus was such an inventor) must have implied the construction of a cosmology with mythic qualities, but without any ritual foundation in society. In the first part of the book I have tried to show that horoscope-astrology (a rather recent invention at that time; the ‘Invention of Nativity’ is the title of a work in progress located at the Martin Luther University of Halle-Wittenberg to be finished by 2022, funded by the DFG) was probably a main instrument for this ‘construction’, using cosmology as a means for fatally and providentially distinguishing individuals, and obviously capable also to distinguish some of them as ‘born’ kings. The alleged correlation of a religious monument of Pax with a meridian as marker of a cosmic visibility of time, which made time itself visible as an ‘Apollonic’ force of order, justice, and harmony, might then appear as a significant complement to the construction of a socially relevant cosmology—a cosmology, though, that had no given roots in Roman tradition. The issue of a new royal mythology has obviously not been settled by the first princeps and his followers, the experimenting with new gods (usually solar ones) by Roman emperors ending with the final paradigm of the Christ, who with his sacred body would eclipse or replace the cosmos as the monogenes, the ‘single born’ son of God in the final sentence of Plato’s crucial dialogue for cosmo-theology, the Timaeus.

B I B L I OG R A P H Y Abel, Karlhans (1974). Zone. Das Problem der Biosphäre im geographischen Denken der Antike. Munich. (From RE Suppl. XIV.) Abry, J.-H. (1988). ‘Auguste, la Balance et le Capricorne’. REL 66: 103–21. Abry, J.-H. (1996). ‘L’horoscope de Rome (Cicero, Div. 98-9)’, in Bakhouche, B. et al. (eds.), Montpellier. Arnaud, P. (1987). ‘L’apothéose de Néron-Kosmokrator et la cosmologie de Lucain’. REL 65: 167–93. Barton, T. (1994). Power and Knowledge. Astrology, Physiognomy and Medicine under the Roman Empire. Ann Arbor. Bayet, J. (1939). ‘L’immortalité astrale d’Auguste’. REL 17: 141–71. Beck, R. (1984). ‘Mithraism since Franz Cumont’. ANRW II,7.4: 2002–115. Beck, R. (1998). ‘Qui mortalitatis causa convenerunt. The meeting of the Virunum Mithraists on June 26,  184’. Phoenix 52: 335–44. Béranger, J. (1953). Recherches sur l’idéologie du principat. Basle. Bernert (1937). RE 17: 1663–72, s.v. Nyx. Binder, G. (1971). Aeneas und Augustus. Interpretationen zum 8. Buch der Aeneis. Meisenheim. Bleicken J. (1998). Augustus: eine Biographie. Berlin. Boll, F. (1950). Kleine Schriften zur Sternkunde des Altertums, ed. V. Stegemann. Leipzig. Boll, F., C. Bezold, and W. Gundel (1926). Sternglaube und Sterndeutung: Die Geschichte und das Wesen der Astrologie. Berlin.

The Peace of Augustus

385

Borbein, A. H. (1975). ‘Die Ara Pacis Augustae. Geschichtliche Wirklichkeit und Programm’. JDAI 90: 242–66. Bouché-Leclercq, A. (1899). L’Astrologie grecque. Paris. Boyancé, P. (1936). Études sur le songe de Scipion. Limoges. Boyancé, P. (1966). ‘L’Apollon solaire’, in Mélanges J. Carcopino. Paris, 149–70. Brendel, O. J. (1979). Prolegomena to the Study of Roman Art. New Haven. Brind’amour, P. (1983). Le calendrier romain. Ottawa. Buchner, E. (1982). Die Sonnenuhr des Augustus. Mainz. Buchner, E. (1996). ‘Ein Kanal für Obelisken. Neues vom Mausoleum des Augustus’. Antike Welt 27: 161–8. Burkert W. (1962). Weisheit und Wissenschaft. Studien zu Pythagoras, Philolaos und Platon. Nuremberg. Cairns, F. (1989). Virgil’s Augustan Epic. Cambridge. Castriota, D. (1995). The Ara Pacis Augustae and the Imagery of Abundance in Later Greek and Roman Art. Princeton. Cramer, F. H. (1954). Astrology in Roman Law and Politics. Philadelphia. Crawford, M. H. (1974). Roman Republican Coinage. Cambridge. De Grummond, N. T. (1990). ‘Pax Augusta and the Horae on the Ara Pacis Augustae’. AJA 94: 663–77. Dicks, D. R. (1966). ‘Solstices, equinoxes and the Presocratics’. JHS 86: 26–40. Eck, W. (1998). Augustus und seine Zeit. Munich. Eriksson, S. (1956). Wochentagsgötter, Mond und Tierkreis. Göteborg. Ferrary, J.-L. (1988). Philhellénisme et impérialisme: aspects idéologiques de la conquête romaine du monde hellénistique. Rome. Fossing, P. S. (1929). Catalogue of the Engraved Gems and Cameos in Thorvaldsen Museum. Copenhagen and London. Fuchs, H. (1926). Augustin und der antike Friedensgedanke. Berlin. Galinsky, K. (1992). ‘Venus, polysemy and the Ara Pacis Augustae’. AJA 96: 457–75. Galinsky, K. (1996). Augustan Culture. Princeton. Geertz, C. (1977). ‘Centers, kings and charisma: reflections on the symbolics of power’, in J. Ben-David and T. N. Clark (eds.), Essays in Honour of Edward Shils. Chicago. Ginzel, F. K. (1911). Handbuch der mathtechnischen Chronologie II. Leipzig. Glei, R. F. (1991). Der Vater der Dinge. Interpretationen zur politischen, literarischen und kulturellen Dimension des Krieges bei Vergil. Trier. Goold, G. P. (1977). Manilius, Astronomica. Cambridge and London. Gordon, R. (1996). Image and Value in the Graeco-Roman World. Studies in Mithraism and Religious Art. Aldershot. Gradel, I. (2002). Emperor Worship and Roman Religion. New York. Grafton, A. T. and N. M. Swerdlow (1985). ‘Technical chronology and astrological history in Varro, Censorinus and others’. CQ 35: 454–65. Gros, P. (1976). Aurea Templa. Recherches sur l’architecture religieuse de Rome à l’époque d’Auguste. Rome. Grottanelli, C. (1987). Art. ‘Kingship’, in M. Eliade (ed.), The Encyclopedia of Religion. New York, 8. 312–17.

386

The Peace of Augustus

Gruen, E. S. (1985). ‘Augustus and the ideology of war and peace’, in R. Winkes (ed.), The Age of Augustus. Providence, 51–71. Gundel, W. (1922). Sterne und Sternbilder im Glauben des Altertums und der Neuzeit. Leipzig. Gundel, W. (1926). RE 13.1, 116–37, s.v. Libra. Halsberghe, G. H. (1972). The Cult of Sol Invictus. Leiden. Hardie, P. R. (1986). Virgil’s Aeneid. Cosmos and Imperium. Oxford. Hommel, H. (1988). Symbola. Kleine Schriften zur Literatur- und Kulturgeschichte der Antike. Hildesheim. Hübner, W. (1972). Hübner, W. (1977). ‘Das Sternbild der Waage bei den römischen Dichtern’. Antike und Abendland 23: 50–63. Hübner, W. (1982). Die Eigenschaften der Tierkreiszeichen in der Antike. Ihre Darstellung und Verwendung unter besonderer Berücksichtigung des Manilius. Wiesbaden. Hübner, W. (1984). ‘Manilius als Astrologe und Dichter’, in ANRW II 32.1, 126–320. Jessen (1913). RE 8, 230ff., s.v. Hemera. Kienast, D. (19993). Augustus, Prinzeps und Monarch. Darmstadt. Klingner, F. (1965). Römische Geisteswelt. Munich. Knell, H. (1986). ‘Symmetrie in der Antike. Form, Begriff und Bedeutung’, in Kunst, Natur und Wissenschaft. Ausstellungskatalog. Darmstadt. Koch, C. (1949). Religio. Studien zu Kult und Glauben der Römer. Nuremberg. Kubusch, K. (1986). Aurea Saecula. Mythos und Geschichte. Frankfurt am M. Laffi, U. (1967). ‘Le iscrizioni relative all’introduzione nel 9 A.C. del nuovo calendario della provincia d’Asia’. Studi classici e orientali 16: 5–98. Le Boeuffle, A., ed. (1983). Hygin, l’Astronomie. Paris. Lloyd, G. E. R. (1966). Polarity and Analogy. Two Types of Argumentation in Early Greek Thought. Cambridge (repr. Bristol, 1987). Long, Ch. (1992). ‘The Pompeii calendar medallions’. AJA 96: 477–501. L’Orange, H. P. (1991). Ara pacis Augustae. Der florale Bereich, in G. Binder (ed.) Saeculum Augustum III. Darmstadt. Maes, F. W. (2005). ‘Die Sonnenuhr des Kaisers Augustus. Aufstieg und Niedergang einer Hypothese’, in Deutsche Gesellschaft für Chronometrie, Jahresschrift. Stuttgart, 168–84. Magdelain, A. (1968). Recherches sur l’‘imperium’. Paris. Maier, H. (1991). Die christliche Zeitrechnung. Freiburg. Malitz J. (1983). Die Historien des Poseidonios (= Zetemata 79). Munich. Merkelbach, R. (1984). Mithras. Königstein. Moraux, P. (1984). Der Aristotelismus bei den Griechen. Vol. 2: Der Aristotelismus im 1. und 2. Jh. n. Chr. Berlin—New York. Neugebauer, O. (1975). A History of Ancient Mathematical Astronomy. Vol. 2. Berlin. Neugebauer, O. and H. B. Van Hoesen (1959). Greek Horoscopes. Philadelphia. Nicolet, Cl. (1991). Space, Geography and Politics in the Early Roman Empire. Ann Arbor.

The Peace of Augustus

387

Nissen, H. (1906). Orientation. Studien zur Geschichte der Religion. Berlin. Ostwald, M. (1969). Nomos and the Beginnings of Athenian Democracy. Oxford. Peterson, E. (1935). Der Monotheismus als politisches Problem. Ein Beitrag zur Geschichte der politischen Theologie im Imperium Romanum. Leipzig. Pollini, J. (1993). ‘The acanthus of the Ara Pacis as an Apolline and Dionysiac symbol of Anamorphosis, Anakyklosis and Numen Mixtum’, in M. Kubelik and M. Schwarz (eds.), Von der Bauforschung zur zur Denkmalpflege. Vienna. Ramsey, J. T. and L. A. Licht (1997). The Comet of 44  and Caesar’s Funeral Games. Atlanta. Rehm, A. (1922). RE XI.2. s.v. Kykloi, 2321–28. Rehm, A. (1927). ‘Der römische Bauernkalender und der Kalender Caesars’, in Epithymbion. Reichenberg. Rehm, A. (1929). RE III.A.1 s.v. Sosigenes (6), 1153–57. Reinhardt, K. (1926). Kosmos und Sympathie. Munich. Richard, J.-C. (1963). ‘Pax, Concordia et la religion officielle de Janus à la fin de la république romaine’. MEFRA 75: 303–86. Rüpke, J. (1990). Domi militiae. Die religiöse Konstruktion des Krieges in Rom. Stuttgart. Ryffel, H. (1949). METABOLE POLITEION. Der Wandel der Staatsverfassungen. Berne. Samuel, A. E. (1972). Greek and Roman Chronology. Munich. Sauron, G. (1988). ‘Le message esthétique des rinceaux de l’Ara Pacis’. RA 1988: 3–40. Schaldach, K. (1997). Römische Sonnenuhren. Eine Einführung in die Gnomonik. Frankfurt. Schilling, R. (1954). La religion romaine de Venus. Paris. Schmid, A. (2005). Augustus und die Macht der Sterne. Antike Astrologie und die Etablierung der Monarchie in Rom. Cologne. Schmid, A. (2002). ‘Augustus, Aequinokt und Ara Pacis’, in A. Péres Jiménes (ed.), Homo Mathematicus. Malaga, 29–50. Schmid, A. (2011). ‘Ein Strukturvergleich der altorientalischen und der historisch-mediterranen Königsherrschaften’, in Th. R. Kämmerer (ed.), Identities and Societies in the Ancient East-Mediterranean Regions. Comparative Approaches. Münster, 277–86. Schmid, A. (2016). Die Geburt des Historischen aus dem Geiste der Politik. Zur Bedeutung frühgriechischer Geschichtsschreibung, mit einem Seitenblick auf China. Basle. Schmid, A. (2020). ‘Der Einspruch gegen die eigene Modernität: Prolegomena zur Dialektik der Moderne’, in H. Cancik et al. (eds.), Archäologie der Moderne. Antike und Antike-Rezeption als Paradigma und Impuls. Basle, 89–109. Schmidt, K. (1980). Kosmologische Aspekte im Geschichtswerk des Poseidonios. Göttingen. Schmidt, H. H. (1991). ‘Die Inszenierung des Privatlebens des Hellenistischen Herrschers’, in J. Seibert (ed.), Hellenistische Studien (Gedenkschrift Bengston). Munich, 75–86.

388

The Peace of Augustus

Schütz, M. (1990). ‘Zur Sonnenuhr des Augustus auf dem Marsfeld’. Gymnasium 97: 432–57. Simon, E. (1986). Augustus. Kunst und Leben in Rom um die Zeitenwende. Munich. Solmsen, F. (1972). ‘Die erste Römerode des Horaz’, in H. Oppermann (ed.), Wege zu Horaz. Darmstadt (first published 1947). Spaeth, B. S. (1994). ‘The goddess Ceres in the Ara Pacis Augustae and the Carthage relief ’. AJA 98: 65–100. Svenbro, J. (2000). ‘L’égalité des saisons’, in C. Darbo-Peschanski (ed.), Constructions du temps dans le monde grec ancien. Paris, 341–54. Szabó, A. (1992). Das geozentrische Weltbild. Astronomie, Geographie und Mathematik. Munich. Thome, G. (2000). Zentrale Wertvorstellungen der Römer. Vol. II. Bamberg. Vlastos, G. (1995). ‘Equality and justice in early Greek cosmologies’, in Studies in Greek Philosophy I (ed. D. W. Graham). Princeton. Walde (2000). DNP 8, 1076f., s.v. Nyx. Wallace-Hadrill (1987). ‘Time for Augustus’, in M. Whitby, P. Hardie, and M. Whitby (eds.), Homo Viator. Bristol, 221–30. Weinstock, St. (1960). Divus Julius. Oxford. Wheatley, P. (1971). The Pivot of the Four Quarters: A Preliminary Enquiry into the Origins and Character of the Ancient Chinese City. Chicago. Zanker, P. (1990). Augustus und die Macht der Bilder. Munich. Zecchini, G. (1985). ‘Il “pius de pace” di M. Terenzio Varrone’, in M. Sordi (ed.), La pace nel mondo antica. Milan, 190–207. Zwierlein-Diehl, E. (1979). Die Antiken Gemmen des Kunsthistorischen Museum in Wien. Vienna.

Details of Original Publication Permission to reprint the following items is gratefully acknowledged. 1. Christopher Smith ‘Dead dogs and rattles: time, space and ritual sacrifice in Iron Age Latium’, in J. B. Wilkins (ed.), Approaches to the Study of Ritual: Italy and the Ancient Mediterranean. London: Accordia Research Centre, University of London (1996), 73–90. 2. John Scheid ‘Le prêtre et le magistrate: Réflexions sur les sacerdoces et le droit public à la fin de la république’, in Cl. Nicolet (ed.), Des ordres à Rome. Publications de la Sorbonne, Série Histoire Ancienne et Médiévale 13. Paris (1984), 243–80. 3. Mary Beard ‘Acca Larentia gains a son: myths and priesthood at Rome’, in Mary Margaret Mackenzie and Charlotte Roueché (eds.), Images of Authority: Papers Presented to Joyce Reynolds on the Occasion of Her Seventieth Birthday, Cambridge Philological Society Supp. 16. Cambridge: Cambridge Philological Society (1989), 41–61. 4. Jörg Rüpke ‘Religion und Krieg. Zur Verhältnisbestimmung religiöser und politischer Systeme einer Gesellschaft’, in R. Faber (ed.), Politischer religion—religiose Politik. Würzburg (1997), 309–28. 5. Andreas Bendlin ‘Nicht der Eine, nicht die Vielen. Zur Pragmatik religiösen Verhaltens in einer polytheistischen Gesellschaft am Beispiel Rome’, in R. G. Kratz and H. Spieckermann (eds.), Götterbilder—Gottesbilder—Weltbilder: Polytheismus und Monotheismus in der Welt der Antike. Tübingen: Mohr Siebeck (2006), Vol. 2, 279–311. 6. John North ‘Action and ritual in Roman historians: or how Horatius held the door-post’, in A. Holm Rasmussen and Susanne Williams Rasmussen (eds.), Religion and Society. Rituals, Resources and Identity in the Ancient Graeco-Roman World. The BOMOS-Conferences 2002–2005 (Analecta Romana Instituti Danici, Supplementum 40). Rome: Quasar (2008), 23–36. 7. Olivier de Cazanove ‘Pratiques et rites de la guerre en Italie, entre Romains et Samnites: le passage sous le joug, la légion de lin samnite’, in J.-C. Couvenhes, S. Crouzet, and S. Péré–Noguès (eds.), Pratiques et identités culturelles des armées hellénistiques du monde méditerranéen. Hellenistic Warfare 3, Bordeaux (2011), pp. 357–70.

390

Details of Original Publication

8. Rebecca Flemming ‘Festus and the role of women in Roman religion’, in F. Glinister and C. Woods (eds.), Verrius, Festus, and Paul: Lexicography, Scholarship and Society. Bulletin of the Institute of Classical Studies, Supp. 93. London: Institute of Classical Studies (2007), 87–108. 9. Denis Feeney ‘Interpreting sacrificial ritual in Roman poetry: disciplines and their models’, in A. Barchiesi, J. Rüpke, and S. Stephens (eds.), Rituals in Ink: A Conference on Religion and Literary Production in Ancient Rome Held at Stanford University in February 2002. Stuttgart: Steiner (2004), 1–21. 10. Daniel J. Gargola ‘The ritual of centuriation’, in C. F. Konrad (ed.), Augusto Augurio: rerum humanarum et divinarum commentationes in honorem Jerzy Linderski. Stuttgart: Steiner (2004), 123–49. 11. Filippo Coarelli ‘Il Lucis Pisaurensis e la romanizzazione dell’ Ager Gallicus’, in Christer Bruun (ed.), The Roman Middle Republic: Politics Religion and Historiography c.400–133 B.C. Acta Instituti Romani Finlandiae 23. Rome: Institutum Romanum Finlandiae (2000), 195–205 (4 figures). 12. John Scheid ‘Rome et les grand lieux de culte d’Italie’, in Annie Vigour et al., Pouvoir et religion dans le monde romain: en homage a Jean–Pierre Martin. Paris: Presses de l’université Paris–Sorbonne (2006), 75–86. 13. Mary Beard ‘Cicero and divination: the formation of a Latin discourse’. JRS 76 (1986): 33–46. 14. Alfred Schmid ‘Augustusfriede, Äquinokt und die Mitte der Welt’, in his Augustus und die Macht der Sterne: Antike Astrologie und die Etablierung der Monarchie in Rom. Cologne: Böhlau (2005), 305–35 (revised version of ‘Augustus, Aequinokt und Ara Pacis’, in Aurelio Pérez Jiménez and Raúl Caballero (eds.), Homo Mathematicus. Actas del Congreso Internacional sobre Astrólogos Griegos y Romanos (Benalmádena, 8–10 de octubre de 2001). Málaga: Universidad de Málaga (2002), 29–50).

Suggestions for Further Reading Recent work on Religion in Rome: 2011–21 Early Rome

We are all too short of detailed knowledge about the lives of the earliest Romans, but recent research has been very successful in filling in some of the gaps, as archaeology in particular has proved a rich source of once lost information. Beard, M. (2015). S.P.Q.R. A History of Ancient Rome. London. Beard, M. (2018). How Do We Look? The Eye of Faith. London. Bradley, G. (2020). Early Rome to 290 BC. Edinburgh, esp. 90–102, 155–73, 272–80, 354–9. Graeber, D. (2011). Debt: The First 5000 Years. London, esp. 210–50. Individual and Group

It used to seem obvious to researchers that early Romans had little or no individual experience of religious convictions or emotions and that religion as such belonged to groups, families, cities; today there is a new kind of discussion which throws such beliefs into doubt. Rüpke, J. (2016a). On Roman Religion: Lived Religion and the Individual in Ancient Rome. Townsend Lectures. Ithaca. Rüpke, J. (2016b). Pantheon. Princeton. Scheid, J. (2016). The Gods, the State and the Individual: Reflections on Civic Religion in Rome. Philadelphia. Original in French, Éditions du Seuil, 2013. Workings of Rome

How and where decisions were taken and who had the power to influence them are topics of great significance in understanding Roman life. Pina Polo, F. (2011). The Consul at Rome. Cambridge, esp.13–57. Richardson, J. and F. Santangelo (2014). The Roman Historical Tradition: Regal and Republican Rome. Oxford. Wiseman, P. (2015). The Roman Audience: Classical Literature as Social History. Oxford. Animals

Animals should never be forgotten in any account of the life of the Romans, whether as sources of food, support, and transport, or rather as vital intermediaries between men and women and their gods and goddesses. Faraone, C. and F. S. Naiden, eds. (2012). Greek and Roman Animal Sacrifice. Cambridge.

392

Suggestions for Further Reading

Gilhus, I. S. (2006). Animals, Gods and Humans: Changing Attitudes to Animals in Greek, Roman and Early Christian Ideas. Abingdon. Augury

This section is intended to deal both with Roman practice in seeking divine advice on the future and other ways of predicting what the future had in store. Berthelet, Y. (2015). Gouverner avec les Dieux: Autorité, auspices et pouvoir, sous la République romaine et sous Auguste. Paris. Driediger-Murphy, L. (2019). Roman Republican Augury. Oxford. Santangelo, F. (2013). Divination, Prediction and the End of the Roman Republic. Cambridge. Women

While men, especially in their own view of the matter, remained the dominant group in war and in political life, it is becoming clearer and clearer that women were far more influential in the city than they were once thought to be, especially in the lives of families. Bettini, M. (2013). Woman and Weasels: Mythologies of Birth in Ancient Greece and Rome. Chicago—London. Original in Italian, Einaudi (1998). Gaspar, V. (2012). Sacerdotes Piae: Priestesses and Other Female Cult Officials in the Western Part of the Roman Empire. PhD thesis. Hemelrijk, E. and G. Woolf, eds. (2013). Woman and the Roman City in the Latin West. Leiden—Boston. Gods and Goddesses

It is important to know that neither male nor female deities had sole dominance over the activities and ambitions of the humans who sought their help as so many did. Bispham, E. and D. Miano, eds. (2020). Gods and Goddesses in Ancient Italy. London—New York. Miano, D. (2018). Fortuna. Oxford. Miller, J. F. and J. S. Clay, eds. (2019). Tracking Hermes, Pursuing Mercury. Oxford, esp. 141–224. Religion as Institution

The relationship between Roman religious activities and their political and social lives has recently become an important subject of discussion. Macrae, D. (2016). Legible Religion. Cambridge, MA. Padilla Peralta, D. (2020). Divine Institutions: Religions and Community in the Middle Roman Republic. Princeton.

Suggestions for Further Reading

393

Invisible Romans

A recurrent problem in dealing with early Rome is the unavoidable fact that most source-material concerns—and only concerns—the wealthy elite and their various assistants. The poor, the enslaved, the exploited are all too easily forgotten. Knapp, R. (2013). Invisible Romans. London. North, J. (2020). ‘Slaves and religion’, in S. Hodkinson, M. Kleijwegt, and K. Vlassopoulos (eds.), The Oxford Handbook of Greek and Roman Slaveries. Oxford. Parkin, T. and A. Pomeroy (2007). Roman Social History: A Sourcebook. London—New York. Urbainczyk, T. (2008). Slave Revolts in Antiquity. Stocksfield. War and Triumph

Rome experienced some disastrous defeats in the slow progress of their domination over Italy, but in the long run, Roman power proved irresistible, as is demonstrated by the tradition of triumphalism in the centuries between the fourth century BCE and the first century CE. Albrecht, J. (2020). Die Religion der Feldherren: Vermittlung und Inszenierung des Krieges in der späten römischen Republik. Schriften zur Alten Geschichte. Berlin. Lange, C. H. (2016). Triumphs in the Age of Civil War: The Late Republic and the Adaptability of Triumphal Tradition. London. Östenberg, I. (2014). ‘Triumph and spectacle. Victory celebrations in the late Republican Civil Wars’, in C. H. Lange and F. J. Vervaet (eds.), The Roman Republican Triumph beyond the Spectacle. Rome, 181–93. Rüpke, J. (2019). Peace and War at Rome: The Religious Construction of Warfare. With an Afterword by Federico Santangelo, trans. D. M. B. Richardson. Stuttgart.

Index For the benefit of digital users, indexed terms that span two pages (e.g., 52–53) may, on occasion, appear on only one of those pages.

Apollo 137–40, 301–2, 316–17, 371–4, 377 Ara Maxima 208–11 Ara Pacis 248, 351–5, 357–9, 363, 369–75, 382–3 Arnobius of Sicca 164–5, 218 Arval Brethren 77–9, 84–5, 88–90, 97 Asad, T. 10–11 Astrology 374–6, 383–4 Attus Naevius, myth of 78–9, 90–4, 96–8 Augury 21, 47–9, 56–7, 61, 64–8, 90–2, 268–78, 288, 291–2, 325–7, 338–40, 342–3, 347 Augustine 142, 363 Augustus 37–9, 55, 64–8, 135–6, 139–40, 144–52, 170, 311–12, 314, 316, 319, 321, 351–3, 355–67, 375–9, 381–3 Aulus Gellius 88–9, 213–15, 228 Auspices 56–7, 65–7 Aztec parallel 116, 120–1 Bardt, Carl 36 Beard, Mary 207 Belief(s) 120–1, 126–7, 175 Boll, Franz 360–3 Buchner, Edmund 352–60 Burials 11–17, 26–7 Burkert, Walter 236–7 C. Valerius Flaccus 52–5 Caeninenses 311–13 Calendar 355–60, 375 Camillus 47–9, 301–2, 304–5, 307 Capitoline 132–40, 144–9, 151–3 Cassius Dio 145–6 Caudine Forks 186–95, 236–7 Cazanove, Olivier de 205–8, 212–14 Centuriation 263–78, 288 Ceres 23, 217–26, 246–57, 374–6

Cicero 272, 282–4, 289–90, 313–14, 325–6 Cicero and Ulpian 59–61 Clodius 94, 170–3 Coins 79, 216–17, 374–6 Conscription 107–8, 115–16 Cosmology 231, 289, 368–9 Curius Dentatus 301–2, 305–6 Dionysius of Halicarnassus 47, 49, 78–9, 82, 85–8 Divination 325–30, 339, 342–3, 346–7 Dreams 145–6 Dumézil, Georges 129–30 Etruscans 269–70, 274–5 Fantham, E. 246–9 Ferguson, Brian R. 104–13 Festivals see Robigalia Frontinus see Hyginus Fortuna 302–4, 313–14 Fuchs, Harald 363–4 Functionalism 9–10 Fustel de Coulanges, N. D. 9 Gale, Monica 242 Geertz, Clifford 8, 143–4, 255–6 Gods, goddesses 140–3 See also: Iuno, Iuppiter, Minerva, Ceres; Hercules; Mars; Proserpina; Silvanus Gordon, Richard 237–8, 240–1, 253 Habinek, Thomas 239–40, 243–5 Hercules 88, 130, 132, 134–5, 208–10, 230–1 Hispellum (Spello, Umbria) 317–18 Holy War, concept of 102–4 Horatius 102–4

396

Index

Hornung, E. 141–6 Hyginus Gromaticus 263–5, 267–8, 286–7 Immunity 47–9, 316–18 Inter-disciplinarity 234–5 Iuno 165–6, 301, 305 Iuppiter Optimus Maximus 125, 165–7, 169, 177–8, 304 Knives 13–16, 24–5, 247–8 Le Bonniec, Henri 217–23 Lex Domitia 64–6 Lex Labiena 66 Libra (balance) 134–5, 359–63, 366–8, 374–81 Literature 125–8, 164–5, 234–6, 238–9, 242, 255–7, 332, 351–2, 357–9 Livy 18, 21, 47–9, 51–4, 84–6, 90–3, 111–12, 162–9, 173–81, 218, 271–2, 275, 295–6, 316–17 Lucus Feroniae 316–17 Luperci 44–7, 53–4 Magistrates 33–50, 54–63, 65–70, 92, 110–11, 113–14, 119, 136–7, 179–80, 278–88, 291, 313–15, 318–20 Mars 19–20, 132–3, 185–6, 208–9 Minerva 132, 134–5, 137–8, 151–3 Mommsen, Theodore 19, 40, 47–9, 55–6, 58–9, 64–5 Müller, K. O. 131 Numa, King, and Egeria, myth of 85–90, 177–80, 271–2, 275, 282–3 Octavian 7–8, 18, 22–4, 67–8, 245, 311–12, 317–21 Osteria dell’Osa see also f. 1.1–1.3; burials at 13–15 Ovid 18–21, 25, 28–9, 177, 222–3, 235–6, 246–55 Philodemus, of Gadara 140–2, 148–9 Philosophy, Roman 331–46 Piccaluga, G. 215 Pinarius Natta 46–7, 50–1, 170, 172 Plautus 219, 283–7, 332 Plutarch 47–9, 85–90, 94, 179, 199, 205–6, 210–11, 218 Polybius 63, 162–3, 197–9, 280–2, 339 Polytheism 126–9

Primigenia at Praeneste 313–15 Priests 25–9, 34–9, 42–64, 66–9, 77–8, 80–5, 87–90, 93–8, 114, 179–80, 284, 311–12, 315–17, 320, 344 Proserpina 218, 222–3, 249, 251 Quintilian 33–4, 67–8, 96 Rattles at Osteria dell’Osa 25 Renfrew, C. 8–9 Ritual 19, 25–9, 82, 108–9, 112–16, 125–46, 150–1, 153, 162–78, 180–1, 194, 205–8, 211–22, 224–32, 234–40, 243, 245–6, 249, 253, 255–7, 263, 266–7, 274–5, 281–3, 286–7, 289–91, 311, 382–3 Ritual, sites of, Caenina 311–13 Clitumnus 315–16 Diana Trifatina 314, 316 Fortuna Praen 313–15 Hispellum 317–18 Lucus Feroniae 312–13, 316–17 Robigalia 18–29 Romulus 78–9, 85–6, 88–90, 92, 177, 214–15, 245–6, 268–9, 287, 311–13 Sacra peregrina 225–6 Sacrifice 16–19, 21–6, 28–9, 56, 69–70, 82, 85–6, 88–90, 94, 103–6, 116, 119, 153, 163, 165, 177, 195, 205–10, 216, 220–2, 224–32, 235–57 Salii 102–4 Schaldach, K. 353–5 Schütz, M. 353–6 Silvanus 134–5, 208–10, 230 Szemler, G. J. 39, 46–7, 58 Temples 17–18, 27–8, 82, 110–11, 119, 132–3, 150–1, 205, 264, 270, 272, 274–9, 281, 283, 289–90, 320 Triumph 112–15, 117 Ulpian 33–4, 59–61 Vestals 56, 78, 85–6, 173, 207, 225, 229–31, 305 Virgil 169–70, 181, 185–6, 212, 219, 234–6, 240–7, 250–1, 253–5, 351–2 Vitruvius 132–5, 138–40 Votive deposits in Rome 17–18 Wissowa, G. 44–6, 126–7, 147, 311–12 Women, religious role of 25, 130, 152–3, 165, 205–31