Prophets and Profits

This volume examines the ways in which divination, often through oracular utterances and other mechanisms, linked mortals with the gods, and places the practice within the ancient sociopolitical and religious environment. Whether humans sought knowledge by applying to an oracle through which the god was believed to speak or used soothsayers who interpreted specific signs such as the flight of birds, there was a fundamental desire to know the will of the gods. In many cases, pragmatic concerns – personal, economic or political – can be deduced from the context of the application. Divination and communication with the gods in a post-pagan world has also produced fascinating receptions. The presentation of these processes in monotheistic societies such as early Christian Late Antiquity (where the practice continued through the use of curse tablets) or medieval Europe, and beyond, where the role of religion had changed radically, provides a particular challenge and this topic has been little discussed by scholars. This volume aims to rectify this desideratum by providing the opportunity to address questions related to the reception of Greco-Roman divination, oracles and prophecy, in all media, including literature and film. Several contributions in this volume originated in the 2015 Classics Colloquium held at the University of South Africa and the volume has been augmented with additional contributions.

Richard Evans has taught at the Universities of South Africa and Cardiff. His research has focused on the political and military history of Greece and Rome, and the ancient topography of Sicily and Magna Graecia. He is currently an Honorary Research Fellow in the Department of Biblical and Ancient Studies at the University of South Africa. This page intentionally left blank Prophets and Profits Ancient Divination and Its Reception

Edited by Richard Evans First published 2018 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 711 Third Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2018 selection and editorial matter, Richard Evans; individual chapters, the contributors Published in association with Acta Classica as Acta Classica Supplementum #9 (ISSN 0065-1141) The right of Richard Evans to be identified as the author of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data Names: Evans, Richard J., 1954- editor. Title: Prophets and profits: ancient divination and its reception / edited by Richard Evans. Description: First [edition]. | New York: Routledge, 2017. | Includes bibliographical references and index. Identifiers: LCCN 2017015105| ISBN 9781138290150 (hardback: alk. paper) | ISBN 9781315266527 (ebook) Subjects: LCSH: Divination. Classification: LCC BL613 .P76 2017 | DDC 203/.2093–dc23 LC record available at https://lccn.loc.gov/2017015105 ISBN: 978-1-138-29015-0 (hbk) ISBN: 978-1-315-26652-7 (ebk) Typeset in Times New Roman by Deanta Global Publishing Services, Chennai, India Contents

Notes on contributors vii Preface x Acknowledgements xi Abbreviations xii

1 Introduction 1 Daniel Ogden

2 Was Didyma (Branchidae) a false oracle? 16 Richard Evans

3 Who wrote Greek curse tablets? 31 Olivier Dufault

4 A story of blood, guts and guesswork: synthetic reasoning in classical Greek divination 50 Ralph Anderson

5 value-added divination at Dodona 65 Philip Bosman

6 Divination and profit in the Roman world 76 Federico Santangelo

7 Profiting from prophecy: Q. Marcius Rex and the construction of the Aqua Marcia 87 Alex Nice vi Contents 8 valerius Maximus and the language of stars 106 Jeffrey Murray

9 ‘Arrows fletched from our own wings’: discovering a ‘Delphi of the mind’ in the writings of the Early Church Fathers 114 Daniel Crosby

10 Egyptian necromancy in Heliodorus Aethiopica (6.12–15) and the Witch of Endor narrative (1 Sam 28) 130 John Hilton

11 Sosipatra: prophetess, philosopher and theurgist: reflections on divination and epistemology in late antiquity 144 Crystal Addey

12 One oracle too many? Corippus (and Procopius) on female prophecy in North African divination and profit in the Roman world 162 Martine de Marre

13 Deconstructing divination: superstition, anticlericalism and Cicero’s De Divinatione in Enlightenment England, c. 1700–1730 183 Katherine East

14 Prophecy and Paul Kruger: Robert Grendon’s appropriation of Graeco-Roman prophets and prophetic devices in his South African epic, Paul Kruger’s Dream 199 Szerdi Nagy

15 Cassandra prophesies back: Marion Zimmer Bradley’s Firebrand 209 Elke Steinmeyer

Bibliography 221 Index 245 Contributors

Crystal Addey is a Teaching Fellow in the School of Classics at the University of St Andrews, UK. She has published widely on ancient divination and religion, mainly in Late Antiquity, and on philosophy, especially Neoplatonism. Her monograph Divination and Theurgy in Neoplatonism: Oracles of the Gods was published in 2014. She has also made numerous contributions to journals and volumes of collected papers. Ralph Anderson is Senior Teaching Fellow in Ancient History at the University of St Andrews, UK. His research interests lie mainly in Greek religion, divination and ancient magic. He is currently revising his doctoral dissertation for publication with Cambridge University Press with the provisional title of Dwelling with Divinity: Person, Place and Perception in Athenian Religion. Philip Bosman is Associate Professor in Classics at the University of Stellenbosch, South Africa. He is the author of Conscience in Philo and Paul (2003) and the editor of Mania: Madness in the Greco-Roman World (2009), Corruption and Integrity in Ancient Greece and Rome (2012) and Alexander in Africa (2014). He has also contributed numerous articles to journals in South Africa and in the UK. Daniel Crosby is studying for his PhD at Bryn Mawr College, having previously studied at Fresno Pacific University. His main research interests are Greek and Roman religion, Delphi and early Christianity. He has recently (2016) published in the Journal of Ancient History on Josephus. Martine de Marre is Associate Professor in the Department of Biblical and Ancient Studies at the University of South Africa. Her research to date has spotlighted the role of women in antiquity, especially in Roman North Africa. She has also published on the literary sources of Late Antiquity. Olivier Dufault completed his doctoral studies at the University of Santa Barbara and is currently a Post-Doctoral Fellow in the Distant Worlds Graduate School of Ancient Studies at Ludwig-Maximilians-Universität, Germany. His current research focuses on Greek alchemical inquiry and Late Antiquity. Katherine East is a Leverhulme Early Career Fellow in the School of History, Classics and Archaeology at the University of Newcastle, having previously viii Contributors completed her PhD at Royal Holloway, University of London, UK. She is the author of The Radicalization of Cicero: John Toland and Strategic Editing in the Early Enlightenment (forthcoming). Richard Evans has taught at the University of South Africa, Pretoria, and at Cardiff University, UK. Most recently he has been a Visiting Researcher and Research Fellow in the Department of Biblical and Ancient Studies at the University of South Africa. His publications include: Gaius Marius: A Political Biography (1994), Questioning Reputations: Essays on Nine Roman Republican Politicians (2003), Syracuse in Antiquity: History and Topography (2009), Roman Conquests: Asia Minor, Syria and Armenia (2011), A History of Pergamum (2012), Fields of Death: Retracing Ancient Battlefields (2013), Fields of Battle: Retracing Ancient Battlefields (2015) and Ancient Syracuse: From Foundation to Fourth Century Collapse (2016). He also edited Mass and Elite in the Greek and Roman Worlds: From Sparta to Late Antiquity (2017). John Hilton is Professor of Classics at the University of KwaZulu-Natal, South Africa, and is the current editor of Acta Classica, the journal of the South African Classical Association. He has published extensively on the ancient novel, especially with regard to Heliodorus, in reception studies and in Greek and Latin linguistics. He is co-author of Apuleius: Rhetorical Works (2001 and 2007) and Alma Parens Originalis? Classical Receptions in South Africa, Cuba and Europe (2007). Jeffrey Murray taught at the University of KwaZulu-Natal, South Africa, prior to studying in the Department of Classics at the University of Cape Town. In 2016, he completed a PhD thesis entitled Valerius Maximus on Vice: A Commentary on ‘Facta et Dicta Memorabilia’ 9.1–11. Szerdi Nagy is a Lecturer in Classics at the University of KwaZulu-Natal, South Africa. Her research focuses on the reception of epic literature, classics in Africa and the construction of gender and identity in the ancient world. Alex Nice completed a PhD thesis at Exeter University and was a Lecturer at the University of the Witwatersrand in Johannesburg before teaching in the US. He is currently an Associate of the Free University, Brussels, Belgium. His research has focused on divination, especially in the Roman Republic and Early Principate. Daniel Ogden is Professor of Ancient History and currently Head of Classics at Exeter University, UK. He has published extensively on Greek religion, ancient magic and divination. Among his numerous publications are: Greek Bastardy in the Classical and Hellenistic Periods (1996), Polygamy, Prostitutes and Death (1999), Magic, Witchcraft, and Ghosts in the Greek and Roman Worlds (2002), Aristomenes of Messene: Legends of Sparta’s Nemesis (2004), Night’s Black Agents: Witches, Wizards and Ghosts in the Ancient World (2008), Alexander the Great: Myth, Genesis and Sexuality (2011) and Drakōn: Dragon Myth and Serpent Cult in the Greek and Roman World (2013). He has also Contributors ix edited, among others, A Companion to Greek Religion (2007). He is also an Academic Associate of the University of South Africa. Federico Santangelo is Senior Lecturer in Ancient History at Newcastle University, UK. He is a specialist in the history of the Roman Republic and particularly the period of Marius and Sulla, both of whom he has discussed in monographs: Sulla, the Elites and Empire (2007) and Marius (2016). He has also published extensively and widely on the role of divination in republican politics, notably in Divination, Prediction and the End of the Roman Republic (2013). Moreover, he has authored, edited and co-edited numerous volumes and publications, including Priests and State in the Roman World (with J. H. Richardson, 2011), The Roman Historical Tradition: Regal and Republican Periods (with J. H. Richardson, 2014), Teofane di Mitilene: Testimonianze e frammenti (2015) and Ruin or Renewal? Places and the Transformation of Memory in the City of Rome (with M. García-Morcillo and J. H. Richardson, 2016). He has also edited a collection of previously unpublished papers by Sir Ronald Syme, Approaching the Roman Revolution: Papers on Republican History (2016). In July 2015, he was ‘Visiting Scholar’ to the University of South Africa on the occasion of the Prophets and Profits Classics Colloquium. Elke Steinmeyer is Senior Lecturer and Head of the Classics Programme at the University of KwaZulu-Natal, South Africa. Her main areas of research are in reception studies, Greek tragedy and in Greek and Roman poetry. Preface

The majority of the chapters in this volume had their origins in the Sixteenth Classics Colloquium held at the University of South Africa in July 2015. The objective of the colloquium was to examine the ways in which divination, often through oracular utterances and other mechanisms, linked mortals with the gods and so featured in the sociopolitical and religious environment of antiquity. Whether humans in ancient times sought knowledge by applying to an oracle through which the god was believed to speak, used soothsayers – the μάντεις – who interpreted specific signs such as the flight of birds, or resorted to dice divination, sortation or even the whispered messages from leaves moving in the breeze, there was a fundamental desire to know the will of the gods. In many cases, pragmatic concerns – personal, economic or political – can be deduced from the context of the applicant. Taking a critical look at the wider issues around the writings of ancient authors, and incorporating the vast array of non-literary and material evidence across this research field, has led to stimulating debate in this area of socio-religious scholarship. The concept of pagan divination and communication with the gods in a post-pagan world has also produced an interesting range of literary receptions. The presentation of these processes in monotheistic societies such as medieval Christian Europe, and in later periods, where the role of religion in general had changed radically, provides a particular challenge. This subject has not yet been discussed in any depth by scholars. This volume aims to rectify this desideratum by providing the opportunity to address questions related to the reception of Greek and Roman divination, oracles and prophecy, not only across the genres in which this phenomenon appears but also in various regions around the world, including especially Africa.

Richard Evans Pretoria, March 2017 Acknowledgements

I should like to thank the numerous reviewers of the papers which in due course became the chapters of this volume. Their silent contribution often goes unac- knowledged but reviewing is a taxing process that is nonetheless vital to ensuring that the standards of scholarship are maintained. I take this opportunity to thank you all, especially those hard-pressed colleagues in South Africa. My thanks also to Professor Martine de Marre at the University of South Africa, for her kind support in this project and for organizing the colloquium devoted to the Prophets and Profits theme in 2015. I should also like to thank Michael Greenwood, Matthew Twigg and all their colleagues at Routledge (Taylor & Francis), and Rachel Cook at Deanta Global, for their constant and tireless advice during the preparation of this volume and for making this project a pleasant and rewarding experience. Abbreviations

CIL Corpus Inscriptionum Latinarum, consilio et auctoritate Academiae litterarum regiae Borussicae editum. 17 volumes. Berlin (1863–) DT Wünsch, R., Defixionum Tabellae [Atticae]. Berlin (1897) DTA Audollent, A., Defixionum Tabellae. Paris (1904) FGrH Jacoby, F. et al., Die Fragmente der griechischen Historiker. Leiden et aliubi. (1927–) IG Inscriptiones Graecae. Multiple series, volumes, parts. Berlin. (1903–) IKourion Mitford, T.B., The Inscriptions of Kourion. Philadelphia (1971) LGPN Fraser, P.M. et al., Lexicon of Greek Personal Names. Oxford (1987–) LTUR Steinby, E.S., Lexicon Topographicum Urbis Romae. 5 volumes. Rome (1993–2000) MRR Broughton, T.R.S. The Magistrates of the Roman Republic. New York (1951–2) NGCT Jordan, D.R., ‘New Greek Curse Tablets (1985–2000),’ GRBS (2000) 41 5–46 PG Migne, J.-P. (ed.), Patrologia cursus completes. Series Graeca. Paris (1857–1904) PGM Preisendanz, K. and A. Heinrichs, Papyri Graecae Magicae. Die griechischen Zauberpapyri. 2 volumes. Stuttgart (1973–4) PSI Papiri greci e latini. Pubblicazioni della Società Italiana per la ricerca dei papiri greci e latini in Egitto. 15 volumes. Florence (1912–79) RRC Crawford, M.H., Roman Republican Coinage. 2 volumes. Cambridge (1974) SEG Supplementum Epigraphicum Graecum. Leiden (1923–) SGD Jordan, D.R., `A Survey of Greek Defixiones Not Included in the Special Corpora.’ GRBS (1985) 26 151–197 Tab. Sulis Tomlin, R.S.O., ‘The Curse Tablets.’ In B. Cunliffe (ed.) The temple of Sulis Minerva at Bath. Volume 2. The finds from the sacred spring. 59–277. Oxford (1988)

This page intentionally left blank 1 Introduction

Daniel Ogden

Let us begin, as all accounts of ancient divination should, with the bromide that the activity was not wholly given over to the prediction or navigation of the future.1 It sought the revelation of hidden information, but this information could equally well relate to the past or the present, or indeed to eternity, in the case of escha- tological enquiry. There are of course many other ways to categorize divination. One may, for example, distinguish between procured or ‘impetrative’ methods (such as a purposeful visit to an oracular shrine, or a hieroscopic investigation) and spontaneous or ‘oblative’ methods (such as encounters with unsought dreams or with omens based in nature, as in the case of prodigy births). Or one may distinguish between inductive divination and inspired divination. In inductive divination, one reads signs which the gods have sown, as it were, into the fabric of the universe (in prodigies, in the livers of victims, in bird flights, in the drawing of lots, etc.); in inspired divination, a human soul comes into direct contact with the divine (through a dream, for example, or through more violent forms of possession, as in the case of the Pythia).2 Admittedly, however, future-related divination was and remains the most eye- catching variety, so let us make some observations about it. First, even when divination was sought in relation to the future, queries were seldom framed in terms of a crude demand to know events or outcomes; more typically, one might ask which among alternative courses of action might be more pleasing to the gods, or to which of the gods one might sacrifice for better fortune.3 Secondly, for obvious reasons, divination was a poor and inefficient tool for making sense of the future. But two pleas could be entered in mitigation. On the one hand, it could certainly be a useful tool for managing anxiety: at least one might think that one had done one’s best to discover the will of the gods and to act in accordance with it, and that in itself might be thought an approach deserving of divine mercy. On the other hand, future-divination was a very powerful tool indeed for making sense of the past and for structuring it. It is evident that in many cases a great event, antici- pated or otherwise, was a catalyst for the retrospective collection of future-related omens and prophecies, for the reinterpretation and perhaps also the outright con- fection of the latter (though I suspect that this happened less often that modern sceptics might suppose).4 The work of the chrēsmologoi, the ‘oracle collectors’, keepers of pre-existing oracular pronouncements, was doubtless focused precisely 2 Daniel Ogden upon this activity.5 Pertinent here is Thucydides’ well-known report of the old oracle suddenly recalled (its origin obscure) by the Athenians when they were stricken by plague during the Peloponnesian War, which read, according to an interpretation then favoured, ‘A Dorian war will come, and a pestilence together with it’.6 Pertinent too is Herodotus’ tale of Sparta’s rediscovery of the bones of Orestes: as he tells it, Lichas discovered the bones first, and only then did he work out, intellectually, the relationship between the words of the Delphic oracle supposedly guiding the Spartans to the spot and the circumstances of his actual discovery.7 The enhancement of a given event with retrospectively found omens and prophecies not only conferred significance and grandeur upon it, it also supported the integration of the event into a satisfactory or even a compelling narrative, and above all a memorable one (we have only to think of Herodotus’ marvellous story (7.141) of the Delphic prophecy of the ‘wooden walls’). And, indeed, the promulgation of such narratives legitimated and validated the processes of future- related divination and prophecy more generally.8 We ourselves, of course, are the ultimate beneficiaries of these engagingly woven narratives. Indeed, their engaging nature is doubtless the principal reason why ancient divination remains a topic of fascination for us today. Lucid abstracts are provided with some of the chapters that follow, so the brief treatments here aspire, for the most part, to offer modest additional perspectives upon the chapters’ contentions or upon the material handled in them. Evans (Chapter 2) focuses on Didyma (Branchidae), Miletus’ external oracle. Apollo’s major oracles in Asia Minor, those at Clarus and Didyma, together with the one subsequently founded for him at Daphne near Antioch by Seleucus I (this last, seemingly, in part at least, on the model of Didyma), all seem to have relied upon a sacred spring for their prophetic force, albeit in different ways.9 In history, what is not there, or not said, can often be more telling than what is, and the skill of the historian lies in spotting instructive silences in our sources. And so it is that Evans fashions a suggestive argumentum e silentio about the role of Didyma in the great Ionian Revolt, which was led by the oracle’s own Miletus. One should indeed expect the Didyma shrine to have pronounced a favourable oracle in support of its patron city’s revolt (or at least an oracle which initially seemed to be such), either before or during, and one should equally expect Herodotus to have reported such an oracle, given that he provides such a detailed account of the revolt (his narrative of it spans almost the entirety of his fifth book, albeit with many a digression), given that he is such a great devotee of oracles, and in particular given that he was very much alive to the usefulness of oracles in structuring and shaping historical narratives. Evans’ solution is that the rebel city of Miletus was at loggerheads with its own oracle, a conservative institution and firmly pro-Persian. This antagonism may well be visible in Hecataeus’ advice that Aristagoras should plunder the shrine to fund the revolt.10 Understandably, the wealthy shrines of the Greek world ever preferred order to chaos, and therefore tended to be supportive of the status quo or the great power of the day. Evans’ case may be supported by a fragment of the Hellenistic paroemographer Demon, according to which the Introduction 3 oracle discouraged the Carians from making common cause with the Milesians against the Persians during the revolt.11 We may introduce a further consideration here too. It could be that the role one might typically have expected an oracular message to play in initiating a chain of events in an expansive Herodotean narrative is in this case taken rather by the message of support and advice which the revolt’s leader Aristagoras is sent from the Persian court by his father-in- law Histiaeus (Hdt. 5.35–6). It is noteworthy that the message is conveyed to Aristagoras by means of a tattoo, which in other contexts can serve as a medium for oracles: we think in particular of the oracles tattooed onto the skins of Anthes and Epimenides.12 Dufault (Chapter 3) leaves prophets behind, but remains with issues of profit. He mounts a challenging argument to the effect that there was no significant specialization or professionalization in the writing of curse tablets before the second century ad. Rather, up until this point, users of such tablets were responsible for writing their own. Such a claim, entailing as it does that all users were at least basically literate, has significant implications for their class profile, and suggests that the use of curse tablets was a relatively elite activity.13 Dufault further conjec- tures that the development of specialization in the writing of the tablets, when it came, was a product of or concomitant with the extraordinary developments in the complexity of the tablets’ textual content (in line with the curse recipes and curse texts of the Greek Magical Papyri). Dufault’s hypothesis may draw support from the seeming origin point of the curse tablets. The earliest ones discovered derive from the milieu of the law courts – those of Selinus, c. 500 bc.14 It may at first seem curious that these seemingly visceral artefacts should have originated in such a rarefied and almost intellectual environment. But (relatively advanced) literacy is surely key here: c. 500 bc, the curse tablet represented a new verbal-literary variation upon the wordless voodoo dolls that had long been established in Greek culture. And so it did after all make sense that it should have made its debut in the Greek world’s most intensely literate environment, that of the speechifying rhetoricians. Here, then, we had a series of individuals making physical curses for themselves, but now in the idiom with which they themselves were most familiar, that of the written word. And let us not forget here the intriguing prospect (a fair chance) that a distinctive group of legal curse tablets and voodoo dolls from the Athenian Ceramicus (c. 400 bc), the ‘Mnesimachus doll’ and its associated finds, incorporate an autograph text of the great orator Lysias!15 Faraone’s influential hypothesis that binding curses tended to be produced in agonistic contexts retains some merit, as does the corollary that they were often made between social equals (for all that the curser might be keen to pull ahead of his rival in one respect or other), with all its implications for the identification of the social class of the curse-maker. But this corollary can be taken too far: when shopkeepers or brothel-keepers are identified as the victims of curses, those curses need not have been formulated by directly rival shopkeepers or brothel-keepers.16 The anthropologically widespread notion of the ‘limited good’ may be at play: the 4 Daniel Ogden notion that there is a limited store of good luck out there means that the success of a neighbour comes to be seen as an obstacle to one’s own success: ‘So long as he has it, I can’t have it’.17 And some curse tablets clearly did cross class boundaries in quite emphatic ways, such as the curses made in the first century ad by a ‘wicked’ public slave against the great and good senators of the town of Tuder (the modern Todi in Perugia), the details of which outrage are preserved for us by an inscription (the senators’ names were ‘attached to tombs’).18 A striking implication that emerges from Dufault’s study is its incidental expo- sure of a lack of agricultural or rural content in the earlier tablets. One gets the impression that they were essentially products of an urban environment. This perhaps makes sense if one is indeed to conclude that their use was directly determined by literacy, for this was likely to have been concentrated in cities. However, our evidence may be obscured to some extent here by the practicalities of archaeological investigation, which is more likely to target and indeed simply to find curse tablet deposition sites in urban zones.19 One may still wonder, even so, whether a specialization of a certain kind was required in connection with the earlier curse tablets – a specialization, that it is to say, in the deposition that enabled their activation. Anyone could toss a curse tablet into a sacred spring or a well, whether secretly or ostentatiously, as required. But what about all the curse tablets entrusted to graves? Are we to imagine that they were all inserted by amateur curse-making relatives at the point of burial? In this case, we might think of the famous third-century ad curse against the gates of Rome inserted into the tomb of one Demetrius by his own brother in the confidence that this new and guaranteed well-disposed ghost would enact it for him.20 Beyond this, we might also think of Lucan’s horrid witch Erictho who, in attending the funerals of her relatives, makes a show of smothering the deceased’s head in kisses while biting bits off it for her store of magical supplies: with this ‘stuff’ she will be able to control the corpse’s ghost at a later point (Luc. Phars. 6.564–9). If items could be secretly removed from a corpse in the course of a funeral, they could equally well be added to it. But let us contemplate once again the example of the Mnesimachus doll and its accompanying curse tablet box from Ceramicus: this was found in the grave of a corpse that had been mutilated after burial. We can only imagine that gravediggers were operating a lucrative sideline, taking the opportunity to exploit older burials as they dug down for new ones and, in this case, were at once inserting curse tablets and retrieving cadaverous material for use in other spells – admirable economy. Anderson (Chapter 4) takes a cognitive approach to Greek divination. He makes the point that divinatory interpretation was essentially a relational activity: the key thing was ever to select the right circumstantial, thematic or definitional context within which to site and read the omen or oracle in question. Success in the field of divination essentially depended upon the ability to construct a compelling (if provisional) story around the oracle, the meaning of which could only become apparent when every item was in its place and the story was complete (Anderson speaks of ‘framing, synthesis and narrativization’). ‘Performative efficacy’ in the Introduction 5 construction of the right kind of story out of such elements as were available was ultimately more important than the detection of any definitive meaning in signs. It is easy to agree with Anderson’s identification of the importance of story construction in connection with divination (cf. my own words at the beginning of this introduction). The study of Greek religion from cognitive perspectives is indeed very much in vogue since Jennifer Larson’s remarkable Understanding Greek Religion: A Cognitive Approach (2016), and she has much to say on the cognitive background to Greek divination in particular, while taking a radically different approach to Anderson’s. As she notes, divination assumes the existence of a force of intention- ality out there in the cosmos. Human beings are equipped with the tool of ‘agent detection’, that is, ‘the capacity to recognise intentional behavior from cues in the environment’, and ‘over-detection of agency [is] a better evolutionary strategy than under-detection’. One is more likely to survive, Larson explains, if one over-detects tigers in the rustling of a bush. Hence man’s general propensity to over-read prospective signs of intentional behaviour in his environment, and to commit to the possibility of readable signs in it, as the Greeks certainly did. It is noteworthy, too, that people in all cultures tend to become more responsive to signs of a spontaneous nature in their environment in times of crisis, when a decision is desperately needed in an uncertain context.21 Bosman (Chapter 5) takes us to Dodona. A visit to an oracle was not just about grabbing the info and leaving. Rather, such a visit would normally be a transporting sensory, emotional and intellectual experience, one in which the consulter engaged not only with the divine itself but also with the ambience of the locale and its ancient and venerable myths and traditions (the latter real or confected). A typical visit to the oracle of Trophonius at Lebadeia, as described in some detail by Pausanias, is a case in point: after their multifaceted but ultimately grim experience, consulters supposedly lost the ability to laugh.22 What sort of experience did Dodona offer in this regard? How was it able to recruit clients and pilgrims in the competitive market for divination? Bosman shows that, in the archaic and classical periods, the oracle’s unique selling point was its distinctively and reas- suringly primitive, perhaps even primeval nature. There are several considerations here: (1) its perceived remoteness on the fringe of the Greek-speaking world (this paradoxically combined with a vague notion that it was at the same time a fount of Greekness); (2) its site’s want of architectural development; (3) the partial obscurity of its divine patrons, of whom, however, the name ‘Dodona’ itself seemed to sing; and (4), above all, its focus on a tree, even if the Greeks can have remembered little of the actual Minoan roots of tree cults. It may have been a source of attraction and credibility, too, that the genially primitive but unthreaten- ing Dodona shared its region, Thesprotia, with another oracle probably considered to be yet more ancient, but also rather more terrible, the Oracle of the Dead at the River Acheron, famous in myth and legend (it was some 20 miles distant as the crow flies).23 A notorious enquiry lamella from Dodona asks, ‘To Zeus of the place and to Dione. They shouldn’t use Dorios the evocator [psuchagōgos], 6 Daniel Ogden should they?’24 Here we perhaps find the safe oracle being asked whether it is advisable to progress to the unsafe one, or even being invited to mitigate the terrors and risks of doing so. Santangelo (Chapter 6) discusses Roman anxieties about divination for profit. The Roman state’s hostility to this does not seem to have derived from any general disapproval of free-market enterprise in itself. There was a certain worry that the profit motive might undermine the integrity of divination, and this is expressed repeatedly, for example, in Cicero’s De Divinatione. But the principal concern seems to have been rather that the state should exercise a monopoly over this potentially very dangerous craft,25 with the result that there was a tendency to attempt to suppress the freelance varieties of divination, which would necessarily be those varieties wherein their practitioners required fees, if only to keep body and soul together.26 A key expression of this attitude is to be found in the resolution of the republican senate, recounted by Cicero (Div. 1.92), to attempt to seize control of Etruscan haruspicium for the benefit of the Roman state by the measure of dispatching ten noble lads to Etruria to learn the craft so that they could return to exercise it, without remuneration, on behalf of the city (the precise date of the resolution is unclear).27 As the state subsequently came to be identified with the figure of the (ever-anxious) emperor himself, its control over haruspicium became ever tighter: so it was that Tiberius banned the private consultation of haruspices altogether (Suet. Tib. 63), while Claudius instituted a formal order of state haruspices in ad 47.28 As Santangelo notes, the well-known myth of Tarquin, the Sibyl and her books, for all that it focuses precisely on a sale of divination, contrives paradoxically to affirm the pricelessness of true divination and, notably, divination at state level in particular, through its motif of the unvarying price (Dion. Hal. RA. 4.62). Despite the state’s anxieties, those who found themselves paying freelance diviners will seldom have been looking to its fall, or to the death of the emperor. In the second century ad at any rate, private divination seems to have been dominated by matters of economic concern: as Santangelo demonstrates, such matters occupy a third of the questions of the Sortes Astrampsychi and almost half of the responses of the Asia Minor dice oracles. Most typically, it seems, paying a modest sum to a diviner for his advice would have been viewed as a cautious business investment.

Nice (Chapter 7) reconstructs political and religious battles between 144 and 140 bc over the city praetor Q. Marcius Rex’s attempt to bring fresh water into Rome via a much-needed but massively expensive new aqueduct, the eventual Aqua Marcia, the longest of the city’s aqueducts. The political battles of the Roman republic were often articulated in terms of religion. Accordingly, Marcius’ rivals among both decemviri and the consuls attempted to thwart the project on mantic grounds, finding justification in the Sibylline books and in ominous prodigies of the sort in which Livy revels. But Marcius Rex and his brother were able to push the aqueduct through to completion by virtue of the political and indeed religious capital (gratia) they and their clan, not least their mythical forbears, had accrued. Some of this capital was evidently established through typological methods.29 Introduction 7 Strikingly, it seems that the clan was able to construct a mythical precedent and justification for the aqueduct and in particular for their own construction of it in the claim that one of their supposed ancestors, Ancus Marcius, the fourth king of Rome, had been responsible for creating some sort of forerunner in his own age. The clan was further fortified in the dispute by its own countervailing reputation for expertise in prophecy, a reputation of long standing. Some seventy years previously, one or more of the Marcii had had some prophecies of their own deemed worthy of incorporation into the same Sibylline books. Furthermore, the clan boasted a forebear of remoter mythical status even than Ancus Marcius, in the silenus and sometime river god Marsyas, who had supposedly given his name to them and was responsible for the introduction of the craft of augury to Italy. Marsyas too could be pressed into typological service for the aqueduct. It was said that he had also given his name to the race of the famously snake-bursting Marsi of the Fucine Lake, while the Marcii found that the sources of their new water were ultimately connected, somehow, to this lake. Murray (Chapter 8) demonstrates that Valerius Maximus has a tendency to intervene in the historical material he is reworking for his collection of exemplary deeds and sayings in order to allude to and celebrate the sidereal affinities and catasterizations of various members of the imperial family. Accordingly, he is to be considered a valuable source for what was in fact the acceptable take on these themes in the age of Tiberius. For such dangerous interests ostensibly sat somewhat uncomfortably with the official line on astrology long adopted by the Roman state, as indeed, for that matter, did Tiberius’ own well-attested close relationship with the astrologer Thrasyllus. We repeatedly hear of expulsions of Chaldaeans under the empire, pagan and Christian alike, including by Augustus, Tiberius, Nero, Vitellius and Constantius II.30 The tradition actually stretched right back into the Republic. Valerius himself tells us that as early as 139 bc the praetor Cornelius Hispalus issued a decree against the Chaldaeans, requiring them to depart from Rome and Italy within ten days.31 One has to wonder what actually happened when the Romans issued decrees of expulsion of this sort. How was the decree policed, given the vast numbers that must have been involved? How was it ensured that those expelled did not merely tarry briefly beyond the borders before returning? The questions have only to be posed for the answers to be obvious: such decrees of expulsion could not really be policed, except in the rare cases in which remaining Chaldaeans, or people representable as such, drew attention to themselves by ostentatious activity. The more extreme measure of mass execu- tions would, one accepts, have been rather easier to enforce, however arbitrarily it was done: one thinks, for example of the 2,000 and 3,000 people that Livy tells us were executed for veneficium in the years 184 and 180 bc, respectively (Liv. 39.41, 40.43). In all probability, the decrees of expulsion never effected very much at all in terms of the actual movement of people. Rather, their primary significance must surely have been largely symbolic or performative, as Orlin has argued.32 Crosby (Chapter 9) investigates, in a learned and interesting piece, the multifaceted attacks made upon the great pagan shrine of Delphi in the apologetic writings 8 Daniel Ogden of the early Church Fathers. The material of these attacks was often filched, cyni- cally, one might think, from the critical and sceptical writings on religious matters of the pagans themselves, a phenomenon neatly crystallized by the great Julian who, as Theodoret tells us, attempted to debar Christians from engaging with (possibly even being educated in) the words of classical poets, orators and philosophers, on the basis that they were fletching their arrows against the pagans with feathers taken from their own wings. Crosby rightly observes that the Delphi the Church Fathers were attacking was a straw man, an impressive fantastical construct, ‘a Delphi of the Mind’, far removed from the actual conditions of the sad, silent shrine in their own day, it having been in long-term decline since even before the age of Plutarch. Their game in this regard was evidently a more sophisticated and intellectually abstract one than is immediately apparent – all the better, perhaps, to appeal to and seduce the educated pagan audience targeted for conversion. One particularly appealing feature of the Fathers’ explications of Delphi (and, indeed, of other pagan religious phenomena) is their lucidly mechanistic nature and their wonderful physicality. To modern – perhaps one should say ‘Christian’– eyes, these readings, for all their prejudice and hostility, tend to read rather more accessibly and engagingly than any explanations, such as they are, offered to us by the pagans themselves. We may note the explanation of the Pythia’s enthusiasm formulated by Origen and John Chrysostom (and referred to by Crosby for other purposes): Apollo is an evil demon that emerges from the earth (from its abode in the abyss, no doubt) and enters the Pythia through her genitals as she straddles the tripod, to take violent physical possession of her and set her raving.33 We may add a further example. In an influential discussion in his Apology of ad 197, Tertullian establishes the winged nature of the demons with which the righteous must contend, the corrupted stock of angels presided over by Satan.34 This, he explains, was the category to which the supposed pagan gods belonged, and he illustrates the point with reference to a story famously recounted by Herodotus.35 According to this, Croesus, the sixth-century bc king of Lydia, tested a series of oracles and found Apollo’s at Delphi to be the only reliable one. He tested them by remaining at home in Sardis and cooking up a soup of lamb and tortoise. In the meantime, he dispatched messengers to the oracles to ask them what he was doing. Only Delphi got it right. Tertullian explains that the demon Apollo had been able to fly on its wings to Sardis, cutting through the air in an instant. There it observed directly what Croesus was up to before flying back to Delphi in a moment to give the king’s messenger an immediate response.36 Hilton (Chapter 10) compellingly contextualizes the famous episode of the Old Woman of Bessa in Heliodorus’ Aethiopica, in which an old woman reanimates her son’s corpse for the purpose of divination, against mid-fourth-century ad Christian debates about necromancy, these being enlivened by the conundrum posed by the Book of Samuel’s Witch of Endor episode.37 According to Christian theology, the witch ought not to have been able to call up the ghost of Samuel, which should have been safely confined, like all other ghosts, in the underworld Introduction 9 until the Day of Resurrection. Of course, the date of the Aethiopica’s composition is itself a notorious aporia, but Hilton follows Morgan’s dating of the text to the mid-fourth century ad (subsequent, that is to say, to the siege of Nisibis in ad 350, the historical details of which appear to be reflected in Heliodorus’ account of the siege of Syene).38 But other contexts, we may be sure, were equally live for the Aethiopica’s first readers. For the baser form of Egyptian magic with which the witch is associated corresponds to the series of spells for accomplish- ing skull divination found in one of the grimoires that survive from the sands of Graeco-Roman Egypt. The recipe book in question, one the Greek Magical Papyri, is similarly thought to derive from the fourth century ad. This spell series prescribes rites to be performed on the skull of a dead man to compel his ghost to manifest itself to the practitioner in his sleep, whereupon it may impart its secret knowledge to him.39 Also live for Heliodorus’ first readers would have been the deep tradition of necromantic episodes in classical literature, in particular those featuring the more graphic reanimation variety of necromancy on the one hand and, on the other, those appearing in the novels. As to the former, one thinks of the greatest reanimation sequence to survive in classical literature, that featuring Lucan’s quite Gothic Thessalian witch Erictho (first century ad).40 As to the latter, one thinks of a series of episodes from novels of the second century ad, first among them the episode in Apuleius’ Metamorphoses in which the Egyptian (again) priest Zatchlas reanimates the corpse of the dead Thelyphron to confirm that the man had been murdered by his wife (Apuleius Met. 2.21–30). But the novel tradition evidently pullulated with this sort of thing: we find reanimations of various kinds elsewhere in the Metamorphoses (1.5–19, Socrates); in Iamblichus’ Babyloniaca (a Chaldaean revives a dead girl);41 and in Philostratus’ novelistic third-century ad Life of Apollonius (the eponymous sage restores a dead bride to life).42 Perhaps the most necromantic of all the ancient novels was Antonius Diogenes’ tortuous Wonders beyond Thule, again of the second century ad, in which the murky Cimmerians help Dercyllis to call up the ghost of Myrto, while the wicked Egyptian priest Paapis specialises in inflicting various kinds of death- like states on his victims.43

Addey (Chapter 11) investigates Eunapius’ ‘life’ of the earlier fourth-century ad Sosipatra of Ephesus in his Lives of the Philosophers and Sophists (her ‘life’ is concatenated with the ‘lives’ of her husband Eustathius and her son Antoninus).44 The limited but rich and fascinating details which Eunapius supplies about her are sufficient to indicate that she was in effect, or at any rate was portrayed as, a latter- day Apollonius of Tyana, a female counterpart to the ‘divine man’ (theios anēr) type: an initiate into Chaldaean mysteries, a remote-seeing prophet, a theurgist, a Neoplatonist and a Neopythagorean. Addey’s careful and learned study braces us in salutary fashion against compla- cent analyses of ancient representations of women and their works in accordance with lazy, off-the-peg stereotypes. Johnston’s all-too generic misreadings of her representation are corrected. Sosipatra is not projected as somehow ritually ‘passive’. Rather, Eunapius makes it explicitly clear that Sosipatra was fully trained in 10 Daniel Ogden ritual. When her vine-dressing Chaldaean instructors seal their sacred books into a chest for her upon their departure, they do not do so to leave her somehow passive in relation to them; it is merely a gesture of safekeeping. If Eunapius speaks little of Sosipatra’s physical engagement with ritual, this is because the rituals she performs, as a most advanced practitioner, largely take the form of intellectual and incorporeal worship. Her acts of divination, while appearing spontaneous from an external perspective, similarly proceed from ritual because she enjoys a perpetual internal ritual engagement with the divine. When Sosipatra asks Maximus to perform a counter-ritual against the love spell she suspects has been put upon her by Philometor (is the ‘wood burning’ attributed to him literal or metaphorical?), the request does not reflect her own ritual passivity, but rather her role as an instructor in a ritual of which she is already master; she brings her pupil on by demonstrably showing faith in him, a gesture he receives with pride. We might add to the points Addey makes: that the tendency to silence that Eunapius tells us was exhibited by the grown Sosipatra should not be attributed to any kind of passivity but rather to philosophical discretion (and indeed we are told that her most successful son, Antoninus, would similarly adopt the silence of a statue on occasion; one thinks here too of the popular narrative of Secundus the Silent Philosopher).45 This ‘life’ of Sosipatra exhibits some intriguing thematic correspondences with the first of the engaging tales in the collection of Lucian’s Philopseudes,46 in which a farm’s vinedresser, Midas, is bitten on the toe by viper. A Babylonian- Chaldaean is called in from nearby, and he cures Midas’ rotting leg; indeed, he restores him from the point of death with his spells, which include the tying of a fragment chipped from a virgin’s tombstone to the toe in question. Instantaneously healed, Midas jumps up and carries back the stretcher upon which he had been brought. The Chaldaean then magically summons all the farm’s pestilential rep- tiles to gather before him so that he can burn them all up with his fiery breath. As the animals gather, he somehow knows – presumably by virtue of remote sight – that one big old snake (a drakōn) in particular has failed to heed the summons, and so he sends one of the others to fetch it. Here too, then, as in the Sosipatra material, we find the motifs of (1) farm-visiting Chaldaeans, (2) vinedressers, (3) the stretchering about of men with afflicted legs and (4) the exercise of remote sight.47 Indeed comparison of these two texts might lead one to an enhanced reading of the Eunapian episode in which Sosipatra remote-sees Philometor’s carriage accident. Just as Lucian’s Chaldaean cures Midas’ leg, we may imagine that Sosipatra is not merely remote-seeing Philometor’s accident as it unfolds, but also remotely healing him even as she perceives and as she speaks: ‘his legs are in danger … no, they’re not; his elbows and hands are wounded … but it’s not serious’. Possession of the faculty to effect miraculous cures would, furthermore, bring Sosipatra yet further into line with the ‘divine man’ type, such healing constituting the type’s principal activity. It would be wonderful to know who these marvellous ‘Chaldaeans’ that long stalked the Roman Empire with their mysterious rituals and their astrological expertise actually were. One suspects that few of them had ever set foot in Babylon, and Eunapius’ pair of old men, dressed in animal skins and hawking Introduction 11 their services as itinerant vinedressers, exhibit few signs of having had any link to the city.48 De Marre (Chapter 12), in another learned piece, takes us into the fascinating world of Late Antique North Africa. She evaluates the competing literary and socio-historical contexts of Corippus’ representation of the indigenous Maurians’ (Moors’) use of female prophecy in his Johannis (c. ad 550), his epic devoted to the martial achievements of John Troglita. On one occasion, Corippus (Johannis 6.152–87) gives us a clear example of an inspired prophetess of Ammon. On another, in a slightly confusing passage, he gives us what is formally presented as a description of an inspired Apolline prophetess, though context may lead us to think the description applicable to an Ammonian one too (Johannis 3.85–151). There are no traces of female prophets of Ammon in history or tradition prior to Corippus, so what is going on here? De Marre demonstrates that, on the one hand, the representations depend on classical representations of the inspired Pythia, and indeed Lucan’s famous account of the Pythia in his Pharsalia (5.146–224) constitutes a particular model for Corippus here.49 On the other, they make appeal to the contemporary culture of divination in the region, as evidenced by Procopius. Procopius (Vand. 4.8.12–14; cf. 4.7.28) does indicate quite strongly that the Moors only used female prophets (or at any rate he indicates the Graeco-Roman assumption that such was the case).50 Then early Arabic sources proceed to speak of one Kahina, a seventh-century queen and prophetess of the Berbers.51 This culture of female prophecy was perhaps only loosely tied to cult sites, or not tied to them at all, and so it may have continued to flourish or even come to flourish for the first time in North Africa after the ad 391 Edict of Theodosius closed down the pagan shrines. No doubt the representations of these prophets are also shaped by the highly schematized axis with which Corippus articulates his narrative, in which Christian Byzantines stand against Maurian polytheists. East (Chapter 13) gives us a demonstration of the authority lent by Cicero’s De Divinatione to the anticlerical arguments formulated by English Enlightenment writers (and to the arguments of their respondents too) – a demonstration that will leave the modern Classicist rueful at the neglect of the precious texts of classical literature in the public discourse of his own day. The Enlightenment writers were advocates of ‘natural religion’, a religion bound by the laws of nature: man, they held, was capable of comprehending God and his works through the direct application of reason; indeed, this was the only fashion in which they could truly be com- prehended. The notion, by contrast, of a mystically ‘revealed religion’ was scorned and consigned to the realm of ‘superstition’ with its accompaniments (miracles, portents, etc.). And this is where the blow fell upon the clergy. Insofar as the clergy interposed themselves between the common man and God, and arrogated the role of uniquely authorized interpreters of a revealed religion and of its mystic texts, they could only ever be agents of ‘superstition’. To contentions of this sort the De Divinatione seemed particularly apposite, with the sustained attacks of its second book on bogus varieties of divination and their proponents, men who, precisely, sought to interpose themselves between the common man and the gods, to 12 Daniel Ogden claim privileged access to divine will, all the while exploiting people’s fears and anxieties for their own profit: divinatio was superstitio. A passage in which Cicero (Div. 2.110) speaks disparagingly of the interpreters of the books of the Sibylline Oracles seemed to be of peculiar pertinence. The extent of the Enlightenment writers’ appropriation of this text and the Ciceronian corpus more generally becomes strikingly apparent in John Toland’s aspiration to incorporate into a new edition of the orator’s complete works an index of all passages bearing upon Christianity. Nagy (Chapter 14) gives us some intriguing insights into the reception of the Classics in late-nineteenth-century South Africa with a look at the epic poem Paul Kruger’s Dream by the remarkable Robert Grendon, son of an Irish father and a Herero mother. Only two copies of this poem, published by Munro Brothers in Pietermaritzburg in 1904, survive, though fortunately a new edition is in preparation.52 The poem represents an attempt to give South Africa not only its first national epic, but also an appropriately inclusive one, and so in some respects has much to say to modern generations. If the poem’s control of narrative may leave something to be desired, the 4,750 lines in which its tale is conveyed exhibit highly creditable versification across a range of metres, not least the iambic pen- tameter. The poem tells how a South Africa with a bright future (as Grendon trusted) is forged in a paradoxically negative fashion – through the failure of the ambitions of the supposedly hubristic and tyrannically inclined Paul Kruger. The poem’s themes are conveyed well enough by its somewhat ungainly full title: Paul Kruger’s Dream, The Struggle for Supremacy in South Africa between Boer and Briton; or the Overthrow of ‘Corruption’, ‘Falsehood’, ‘Tyranny’, ‘Wrong’, and the Triumph of ‘Justice’, ‘Truth’, ‘Liberty’, ‘Right’: A Poem. In composing his poem, Grendon made the most of the education he had received in Latin at Zonnebloem College in the Cape (although he seems to have excelled primarily in maths), and in particular his readings in the first half of the Aeneid, that celebration of the founding of the Roman state. So it is that we find Grendon’s Kruger scorning the (truthful) prophecies bestowed upon him by the goddess Fortuna (who owes something to Virgil’s Sibyl), by the god Mars (in ancient tradition, the sponsor of the Trojans with whom Kruger’s Boers are aligned), by Truth in personified form (whose conception owes something to Virgil’s personified abstractions, such as Fama), and by the assembled ghosts of the great Boers of the past (who recall the parades of the dead and the not-yet-born encountered by Aeneas in his visit to the underworld). Given his comfort in the manipulation of the themes and apparatuses of ancient epic, one would give much to know the degree of access Grendon enjoyed to other ancient poems beyond the Aeneid. There is a suggestion of some familiarity with the Iliad, if only in translation. As Grendon’s poem progresses, a Christian voice becomes ever more insistent in these prophecies: once literary legitimacy has been established by means of the use of classical (and necessarily pagan) forms, it seems, attention may then be given to the greater matter of religious legitimacy. And thus the poem closes with Kruger’s deathbed redemption, his return to Christ and his praise of God for the peace he has brought to his land. Introduction 13 Steinmeyer (Chapter 15) reviews Marion Zimmer Bradley’s 2002 novel Firebrand, which attempts to ‘recuperate’ a lost female voice from classical literature, that of Cassandra. Since she is a woman who ever speaks insightful truth and yet ever goes unheeded, one can readily understand the appeal of the Cassandra figure to those seeking to impose upon ancient myths readings of a sort congenial to contemporary feminists. One wonders how long the justifying claim that only (born) females can write myth from a genuinely (gender) female perspective can survive in the current age of transgender politics: it perhaps already feels regressive. Nonetheless, there is evidently much for the Classicist to admire in Bradley’s reworking and kaleidoscoping of the ancient source material. She is to be com- mended for having committed herself to making the most of the further reaches of the Cassandra tradition and indeed the broader field of Greek myth. It is reassur- ing to find her alluding (albeit indirectly) to the obscure but delightful episode in which Cassandra and her twin brother Helenus were left overnight in the temple of Apollo Thymbraeus as babies, whereupon a (or the) pair of the god’s sacred snakes licked out their ears, thereby ridding them of a metaphorical deafness and bestowing upon them the gift of prophecy.53 It is interesting too to see her extend- ing Cassandra’s snake affinities by taking her off to Colchis, there to follow in the serpent-tending Medea’s footsteps.54 It is a pity that Bradley is not able to give greater credit than she does to the one surviving ancient work that is wholly given over to Cassandra’s voice, namely the Lycophronian Alexandra (c. 200 bc), which is admittedly a deeply challenging text. It has recently been rendered slightly more accessible by Hornblower’s new commentary. Pertinently, Hornblower adumbrates the case for supposing that its (pseudonymous) author was actually a woman: the poem adopts female focaliza- tion for most of its extent; Cassandra’s rape is used as an organizing principle; and the author shows remarkable knowledge of female cults.55

Notes 1 Cf. Bonnechere (2007) 145. Note Ogden (2001) 231–50, especially 238: in what aspired to be a comprehensive review of the kinds of answers sought by the specific technique of necromancy (divination from the dead), I could find relatively few examples of straight prediction of the future. There are two caveats here, however: first, our evidence for necromantic consultations is all effectively fictional; second, it might be thought that of all sources of divinatory power, the dead had the weakest natural affinity with the future. 2 For the latter distinction see Bonnechere (2007) 150–5. 3 As with Hdt. 1.158, discussed by Evans in this volume. 4 Fontenrose’s (1978: 240–416 and passim) noble attempt to determine which of the recorded Delphic oracles were genuine and which fictional, etc., may accordingly have been simplistic. 5 Cf. Fontenrose (1978) 145–65; Larson (2016) 79–80. 6 Thuc. 2.54. The text is referred to by Anderson in this volume. 7 Hdt. 1.67, also discussed by Anderson in this volume. 8 I draw inspiration for these remarks from some powerful observations made by Prof. Kai Trampedach at the Tenth Melammu Symposium held at Kassel University in 2016. 14 Daniel Ogden 9 Iamb. De mysteriis 3.11; Amm. Marc. 22.12.8. 10 Hdt. 5.36; cf. 1.92 for the wealth that Croesus had bestowed upon it. 11 Demon FGrH 327 F16. Cf. Parke (1985a) 18–20, 203–4. 12 Stephanus of Byzantium s.v. Ἀνθάνα; cf. Suda s.v. Ἐπιμενίδης. 13 I cannot forbear to mention here the Bath tablets to whose makers illiteracy appears to have presented no obstacle, the mysteries of writing being mimicked with series of repetitive but meaningless marks (Tab. Sulis nos. 112–16; these derive from an age decently subsequent to Dufault’s watershed: the third or fourth century ad). As the exemplary editor of the Bath tablets, Roger Tomlin, sympathetically observes of such illiterate writing, ‘After all, the goddess would be able to read it’: Tomlin (1988) 247. 14 Listed at Jameson et al. (1993)125–6. 15 Jordan (1985a) no. 9 = Faraone (1991a) no. 5 = Gager (1992) no. 41 = Ogden (2009) no. 237; cf. Jordan (1985–1988). 16 DT nos. 41, 47, 52, 72–3, 87, 92, 109; DTA nos. 12, 30, 55, 68–75, 84, 86, 97; Jordan (1985a) nos. 3–4, 11, 20, 43–4, 48, 52, 72–5, 124. 17 Bernand (2000). 18 CIL 11.2.4639; Ogden (2009) no. 180. 19 Let us not leave this point without mentioning another Bath tablet, Tab. Sulis no. 31 (similarly third or fourth century ad, and so again after Dufault’s watershed), in which the aggrieved peasant Civilis asks Sulis for the return of his stolen ploughshare. 20 SEG xiv no. 615, with revisions suggested by Jordan (1985a) no. 129 = Ogden (2009) no. 179. 21 Larson (2016) 74–5, 79. She builds in part on the work of Lisdorf (2007), especially on Roman divination. 22 Pausanias. 9.39; cf. Bonnechere (2003). 23 Hom. Od. 11, Hdt. 5.92; cf. Ogden (2001) 43–60. 24 Christidis et al. (1999) no. 5 (fourth century bc?) = Ogden (2009) no. 31. 25 See Graf (1999) 294–5. 26 In point of fact, however, we may wonder whether Scipio Aemilianus’ clear-out of his Numantian camp in 134 bc really was motivated by a disapproval of divination for profit as such. Appian’s account of the matter might be taken to imply rather that his motivations were discipline and morale (Iberian Wars 85). The traders and the prostitutes were cleared out for the sake of the former, while the diviners and the sacrificers were cleared out quite explicitly for the sake of the latter (ἐλθὼν δὲ ἐμπόρους τε πάντας ἐξήλαυνε καὶ ἑταίρας καὶ μάντεις καὶ θύτας, οἷς διὰ τὰς δυσπραξίας οἱ στρατιῶται περιδεεῖς γεγονότες ἐχρῶντο συνεχῶς). Their presence was a function of low morale in the first place, but no doubt they also sustained a debilitating climate of anxiety even when generally delivering positive messages in individual cases (as it was doubtless in their commercial interest to do). 27 Cic. Div. 1.92. 28 Tac. Ann. 11.15.1. 29 For the term ‘typology’, its roots in Biblical exegesis and its application in Roman thought, see Gransden (1973–1974) and especially (1976) 14–20. 30 Cassius Dio 49.43.5 (Augustus); Tac. Ann. 2.32 (Tiberius); Tac. Ann. 12.52 (Nero); Suet. Vit. 14 (Vitellius); Amm. Marc. 19.12.14 (Constantius II). 31 val. Max. 1.3.3. The decree also bore upon the Jewish proponents of Jupiter Sabazius. 32 Orlin (2010). 33 Origen Against Celsus 3.25.31–4; John Chrysostom Homily on First Corinthians 29.1 = PG 242, 11–19. In this connection, it is a relief to see that the ‘pendulum’ (in Crosby’s phraseology) has swung back again from the revivalist and implausibly historicizing claims of de Boer et al. (2001) to have found evidence of a mephitic cleft beneath Apollo’s temple. I have always been dubious: See Ogden (2001) 245. 34 Tertullian Apology 22; cf. Revelation 12.7–9. Introduction 15 35 Hdt. 1.46–8. The tale also is referred to by Anderson in this volume. 36 Cf. Martin (2001). 37 For which see the most convenient Smelik (1979). 38 Morgan (1996) 417–21. 39 PGM IV. 2006–139. 40 Luc. Pharsalia 6.413–830. Hilton quite appropriately reassures us of the likelihood that Heliodorus could read Latin. I print the name ‘Erictho’ with its classical Latin – and, indeed, Lucan’s – orthography, -cth- being the classical Latin reflex of -χθ-. Many scholars prefer to take the name back to its original Greek form (it is vestigially attested: LGPN s.v.) and then retransliterate it into English as ‘Erichtho’. 41 As summarized at Photius Bibliotheca cod. 94, § 74b. 42 Philostratus Life of Apollonius 4.45. Cf. 4.11–16, where Apollonius summons up the ghost of Achilles. 43 Photius Bibliotheca cod. 166, §§109–11; PSI 1177. See Stephens and Winkler (1995) 101–78; Ogden (2009) 306–9 no. 311. 44 Eunapius Lives of the Philosophers and Sophists 6.5–11. 45 The text is edited and translated by Perry (1964) and the translation is reprinted at Hansen (1998) 68–75. 46 Lucian Philopseudes 11–13. On this see Ogden (2007) 65–104. 47 Improbable as it may seem, one gets the idea that ‘Chaldaeans’ may well have had a tendency to hang around farms. Already in c. 160 bc Cato had been saying of his ideal farm steward (vilicus) that ‘He should not wish to have consulted any haruspex, augur, diviner or Chaldaean’ (haruspicem, augurem, hariolum, Chaldaeum nequem consuluisse velit; De Agric 5). 48 Santagelo (in this volume) perhaps takes the claims of such individuals to Babylonian roots seriously, in his passing reference to ‘Near Eastern astrologers’. 49 Luc. Pharsalia 5.146–224. 50 Procopius Vand 4.8.12–14; cf. 4.7.28. 51 Hendrickx (2013). 52 So reports Christison (2012). 53 Tzetzes on Lycophron Alexandra, introduction; scholl. Hom. Il. 6.76a, 7. 44 (incorporating Anticlides FGrH 140 F 17). 54 Cf. Ogden (2013) 198–209. 55 Hornblower (2015) 40–1. 2 Was Didyma (Branchidae) a false oracle?

Richard Evans

Introduction Miletus you crafty orchestrator of bad deeds, you will become a meal and magnificent prize for the many, your wives will wash the feet of many long- haired men, and our temple at Didyma will be cared for by others. (Hdt. 6.19)

The following discussion attempts to determine the reason for the complete absence from Herodotus’ Book 5 of any reference to oracles originating from the Sanctuary of Apollo at Didyma during the Ionian War (c. 500–493). That problem might not excite much interest initially except when considering the importance Herodotus devotes to oracular messages elsewhere in his history.1 It is therefore curious that there are no oracles – the one exception was from Delphi – dealing with a war which disrupted the whole of western Asia Minor and the islands including Cyprus for nearly a decade, making this a conflict of far greater duration and magnitude than the punitive expedition that Darius sent against Athens and Eretria in 490 or even Xerxes’ full-scale invasion of Hellas in 480.2 Herodotus’ report of the oracle given above is surely also important since it is mentioned not in conjunction with Ionian affairs but in his account leading up to the battle of Sepia in which Sparta defeated Argos in 494 (Hdt. 6.77), and so reflects Argive concerns to know the outcome of that ongoing conflict. Therefore, the oracle dates either to a point in time immediately before or during the Persian siege of Miletus after the battle of Lade.3 It does not precede the beginning of the war but illustrates oracular comment on what was almost inevitable by 494: the destruction of Miletus’ power. It is not relevant to the time when it ought to have been announced: before the war began or when Aristagoras, probably in 500 or the spring of 499, was active in the Peloponnese trying to elicit support for the rebellion.4 It certainly does not explain why the oracle at Didyma appears to have been tight-lipped on the subject of a war against the Persians and why Herodotus chose to ignore any issues about Miletus and its relations with the nearby shrine. The cult of Apollo (and Artemis) at Didyma was administered by Miletus since it was situated in its chorē (Hdt. 1.157–158; Strabo, 14.1.5), and the daily functions of the temenos were undertaken by a local clan named the Branchidae. Was Didyma (Branchidae) a false oracle? 17 Didyma was one of the most respected oracular sites in Asia Minor if not in the entire Greek world and its treasury possessed great wealth deposited there, especially by Croesus the Lydian king (Hdt. 1.92) who had sought advice about making war on Persia. Herodotus specifically states that ‘all the Ionians and Aeolians made a point of consulting the oracle’, but also notes its international reputation (Hdt. 2.159). Yet no encouraging sign seems to have emanated from Didyma, nor any warning of the catastrophe that was to engulf not just Miletus but eventually the whole region.5 And why should the Ionian Greeks appear to have ignored such a prestigious oracle which they would normally have consulted when they were preparing to wage war against Persia? Moreover, can Herodotus have considered the outcome of the Ionian cities’ rebellion against the Persian king so predictable that he did not consider recording any oracle about this war, or was there a darker motive in his failure to transmit messages from deities which elsewhere are deemed to be of interest to the audience? Considering the importance Herodotus devotes to oracular messages elsewhere in his history, the absence in Book 5 of oracles of any origin dealing with any matter, not to mention other supernatural occurrences, shows a startling departure from the topical elements that comprise his methodology elsewhere.

The rebellion of Pactyes and an oracular response from Didyma Herodotus is the sole source for this episode, which evidently occurred shortly after the fall of Croesus in 546–545 bc (Hdt. 1.153–61). The Persian king Cyrus returned to Ecbatana soon after the capture of Sardis but he first appointed Tabalus as military governor (in effect, satrap) and the civilian Pactyes as the head of the local administration.6 This division of control was probably intended to prevent any unrest among the new subject population. However, Cyrus seems hardly to have departed when Pactyes chose to lead a rebellion, raising a force of mercenaries mostly, it appears, from among the Greeks on the coast, with which he took control of Sardis and besieged Tabalus in the acropolis. The ease with which the rebels took much of Sardis suggests that Tabalus possessed little in the way of military resources. On the other hand, Pactyes seems not have been that popular with the local community if his support came essentially from paid troops. On the face of it, this was not a Lydian uprising but a quarrel between two figures in the newly formed government in Sardis.7 Moreover, Pactyes’ revolution was brief and in a matter of days, at most weeks, a relief column sent by Cyrus approached Sardis and what support there was for the rebel melted away. It is the events which took place in the aftermath of this uprising that have some impact of the assessment of the worth of Didyma as a trustworthy oracular site. Pactyes fled from Sardis to Cyme and when the Persians demanded that he be surrendered, its citizens prevaricated and sent a delegation to Didyma to ascertain whether they should accede to this request. The delegates were to ask which action regarding Pactyes would be more pleasing to the gods (Hdt. 1.158). The oracle stipulated that Pactyes should be handed over to the Persians. The citizens accepted this response but a certain Aristodicus interjected as they were about to 18 Richard Evans deport Pactyes. He claimed to have some doubt about the oracle and suggested that the legation sent to Didyma had misunderstood or misinterpreted the message. He created sufficient disquiet that a further delay was called while a second delegation including Aristodicus was dispatched to pay another visit to Didyma. Aristodicus acted as spokesman at the shrine and explained that Pactyes had come to the Cymaeans as a suppliant and would be executed if delivered to the Persians; he stated that they did not wish to be a party to violence unless the god gave a specific command. The oracle again stated that Pactyes should be handed over to the Persians. Yet Aristodicus remained unconvinced, which is certainly an indication of scepticism about the validity of the oracle. Herodotus relates that Aristodicus remained at the shrine and walked around the temple, removing birds wherever he found them nesting in the walls. Then a threatening voice was heard to come from the ναός of the shrine which reprimanded the Cymaean for his behaviour, for the birds had sought refuge in the god’s sanctuary. Far from being awed by the god’s anger, Aristodicus responded with equal vigour, calling out the double standards of the god who chose to take care of the suppliants in his temple – the birds – but ordered the Cymaeans to give up Pactyes, a suppliant in their city. Still it seems that Apollo had the last word. The voice dismissed Aristodicus’ protest with the comment that the god did indeed protect suppliants in his own temple but that the Cymaeans must surrender theirs. The sooner this was done, the quicker Cyme would pay the penalty for its sacrilege and the god ordered that the Cymaeans must never again approach Didyma for advice about future suppliants. The Cymaeans returned to their city but, rather than handing Pactyes to the Persians they sent him to Mytilene on Lesbos. When they heard that the Mytileneans had been offered money to deliver him to the Persians, they rescued Pactyes and transported him to Chios. But the rebel was soon given to the Persians by the Chians, who were rewarded for their treachery with land on the coast at Atarneus (Hdt. 1.160). The Cymaeans’ actions may initially have deferred their punishment by the gods for delivering their suppliant to the enemy but the Pactyes affair ushered in a period of great upheaval and destruction to Ionia, as Herodotus clearly reveals (Hdt. 1.161–71). The Persians were intent on punishing any city which had contributed to Pactyes’ rebellion and from Herodotus’ account it appears that there had either been widespread support for his cause in Ionia and elsewhere or that the rebellion had caused instability across the region.8 The Persians quickly and thoroughly regained control. Along the west coast of Asia Minor, Phocaea and Teos were sacked (Hdt. 1.163–8), and other Ionian cities were similarly subju- gated. Herodotus fails to mention how all this must have affected the Aeolians, who lived directly to the north of Ionia, and he offers no account of the fate of Cyme, which had perhaps been mostly closely linked with Pactyes and his revolt.9 Moreover, Miletus was completely unaffected by any of these events. Sole rule in the form of tyranny appears to have been the regular system of government in Miletus for much if not all of the sixth century. Herodotus (Hdt. 5.92) refers to Thrasybulus the tyrant of the Milesians as a contemporary of Periander of Corinth, who ruled from about 625 to 587 bc.10 There was also close cooperation Was Didyma (Branchidae) a false oracle? 19 between Miletus and Lydia whose king, Croesus, contributed as liberally to the treasury at Didyma as he did to Delphi.11 Following the destruction of Lydian power, Miletus, like the other cities along the coast of Asia Minor, came under the control of Persia and so the existing system of tyrannies was maintained, although there is no record of the ruler of the city at the time of Croesus’ fall or in the subsequent uncertainties of Pactyes’ revolt. However, from about 520 the tyrant was Histiaeus, a loyal partisan of Darius who was rewarded with the position of adviser to the king, whom he accompanied to Susa in about 510.12 At that juncture the administration of the city passed to Aristagoras, Histiaeus’ son-in-law. In 501 or 500, Aristagoras dispensed with a government controlled by a single ruler and the Milesian demos elected generals to conduct the revolt against Persian rule.13 And so a long-standing system of government at Miletus was replaced and this political upheaval will surely have affected all institutions in the polis, including the religious sites such as nearby Didyma. Nonetheless, the Branchidae, who supervised the cult, seem to have been unaffected by the turmoil which had developed all around them in the 540s and then again nearly fifty years later. It is therefore very curious indeed that instead of appealing to an oracle unconnected with the Persians, the Cymaeans chose to send a deputation to Didyma, although it is also worth noting that the suppliant was not a Greek, according to Herodotus (Hdt. 1.159), but of Lydian origin. The Cymaeans could easily have sent a delegation to another oracular centre in the region.14 Perhaps their reasoning was based on Didyma’s long relationship with the Lydian king, who had by then been taken captive or killed by Cyrus.15 And Miletus had evidently not engaged in any hostile act against Persia and had in fact already negotiated a peace with Cyrus. Therefore, the response of the oracle can hardly have come as a surprise to Aristodicus and his fellow delegates. Understandably, not wishing to jeopardize that peace, the Branchidae were not helpful to the Cymaeans. It is of course possible that Herodotus’ account, with its supernatural tendencies, disguises a more pragmatic reason for the Cymaeans’ presence in Didyma. Since all the other Ionian cities had declared themselves free when Lydian power came to an end, was the intention to force the hand of the Milesians into joining a common rebellion? In any event, the strategy failed and backfired since the Branchidae simply followed the policy of their patrons the Milesians. Brown argues that the negative portrayal of the Branchidae, and hence of the shrine at Didyma, found in Herodotus’ account was originally the creation of Xanthus, the Lydian historian whose Lydiaca influenced all later accounts. He suggested that to a Lydian, ‘it might have seemed appropriate that a corrupt family of priests … betrayed the Lydian cause after taking Croesus’ gold,’16 yet there can be no certainty that Pactyes was actually Lydian, and besides, Lydian participation in this rebellion had already ceased. It was far more realistic of the Branchidae to recommend an accommodation with Persia and avoid certain destruction, a denouement which the oracle clearly presaged. Contrary to Brown’s thesis, therefore, the reason is more likely to be a much less complex one of political obliga- tion and, as such, unconnected with any literary tradition or bias, especially since 20 Richard Evans the special political attachment of the Branchidae at Didyma can also be identified later in the greater war after 500 bc.

The Ionian War, Aristagoras, Hecataeus and the Branchidae Miletus maintained its particularly close ties with Persia, especially when Darius made Histiaeus its ruler some years before the king campaigned in Thrace and across the Danube (Hdt. 4.83–143). But following Histiaeus’ forced exile in Susa from about 510, the internal politics in Miletus turned against Persian rule and there appears to have been a widespread desire to rebel. This change of allegiance was in part the work of Aristagoras, who ruled Miletus in Histiaeus’ absence. Histiaeus encouraged Aristagoras in his scheme, although the motives of both were driven more by personal ambition – the former to return to his city, the latter to avoid losing his position – rather than to achieve any long-term independence for the Ionian cities. Aristagoras’ singular lack of success in an attempt to capture Naxos for the Persians, and his lack of funds afterwards to pay the troops he had employed in the adventure, accelerated an existing political crisis and helped the personal intrigues of the protagonists, namely Histiaeus and Aristagoras. Following his return from Naxos and before he could be reprimanded or punished by Artaphernes, the Satrap of Lydia, Aristagoras sounded out the level of support for a rebellion among his friends. He may well have regarded this option as his only future course, but Histiaeus sent a message to his son-in-law, also encouraging revolt, which arrived at this precise moment (Hdt. 5.35).17 In this meeting a vote for military action would have been unanimous if not for the opposition of the Milesian philosopher Hecataeus, who was present and argued for maintaining the status quo.18 Hecataeus offered some interesting advice at this stage. He urged the Milesians to invest in an enlarged fleet of warships and to gain command of the sea.19 In order to finance this strategy, Hecataeus recommended the seizure of the treasures deposited at Didyma by Croesus (Hdt. 5.36). Although this advice was not taken up, in such an open discussion, perhaps recorded by Hecataeus, the news will have reached Didyma very quickly. It would be strange if the Branchidae’s reaction to this denuding of their sanctuary’s wealth had been positive, and if it had become a subject of debate, the possibility remained that the Milesians would remove the treasures when or if they were required in the future. It is therefore highly probable that after a half-century of close cooperation with Persia the relationship between the Milesians and the Branchidae at Didyma now became rather strained to say the least. If, as seems very likely, given the tradition, the Branchidae were not in favour of a war against Persia, this stance made the shrine vulnerable to attack by the Greeks but meant that it could also be surrendered willingly to the Persians by the keepers of the shrine. The Milesians missed an opportunity to finance their war, but the Branchidae did not forget the threat. It should further be noted that the mainland Greeks did not pillage sanctuaries such as Delphi in this period and the first recorded occasion was almost certainly during the Third Sacred War in the 350s (Diod. 16.14.3, 23.1). At that time, the Phocian leaders decided to fund hostilities against the Amphictionic League Was Didyma (Branchidae) a false oracle? 21 and Philip II of Macedon by drawing on the contents of the various treasuries at Delphi for the payment of mercenary soldiers.20 One hundred and fifty years before this, Hecataeus had presented a cogent argument for ransacking Didyma to finance the war against Persia. Could this rupture between shrine and polis be the reason that when oracles feature so prominently in the history of Herodotus there are so few references to those of Didyma and none related to the Ionian War, when it should have been a usual feature of the narrative? Moreover, why would any Ionian Greek at war with Persia seek advice from an oracle so assuredly and consistently pro-Persian? The roles of Aristagoras and Histiaeus consume Herodotus’ account of the Ionian War (Hdt. 5.35–6.21).21 Thus, while Herodotus has room to tell the tale of the slave sent from Histiaeus to Aristagoras bearing a message urging rebellion tattooed on his head (Hdt. 5.35), and even of oracles received by other Greek cities, none seem to have been deemed worthy of mention for the Ionians or their allies among the Aeolians or Carians, whose fate is the focus of the narrative.22 In other words, Herodotus gives the impression that all military decisions made by the Ionians in this war were governed not by religion or divination, but by rational thought. Yet the attack on Sardis (Hdt. 5.99–103) by the Ionians and their allies, the Athenians and the Eretrians, would normally have been preceded by advice from the deity. The campaign ended in a serious defeat and the Athenians refused to provide further aid for the rest of the war. Was it because an oracle had been ignored or never sought? We are given no reason for the Athenians’ about-face. Moreover, why, when only advice from Hecataeus is alluded to – that the former tyrant should evacuate just to the island of Leros, close to Mount Mycale, so that he might return more easily – did Aristagoras decide to desert the cause he had been instrumental in launching (Hdt. 5.124–6)? Aristagoras chose to flee to Myrcinus in Thrace, where he was soon killed. Had he sought an oracle for this action and why was Hecataeus ignored? Undeterred by their leader’s departure, the Milesians elected new generals and carried on their war, although they did so now in the face of overwhelming odds. The Milesians had few options by then, unless they defeated the Persian fleet, and so warships were gathered at Miletus by the Ionian allies (Miletus, Priene, Myus, Teos, Phocaea, Erythrae, Chios, Lesbos and Samos).23 But there is no word that an oracle was sought or delivered on the Greeks’ capacity to defend themselves and gain mastery of the sea. Hecataeus had advised this strategy long before, in 500, but with just 353 warships in the Greek fleet against a force of 600 Persian ships, the outcome looked bleak. Most of the Samian contingent broke ranks at the start of the battle, followed by the ships of Lesbos, leaving the Greeks outnumbered three to one. It was a wonder that the fighting (Hdt. 6.15–16) was so long drawn out until the Greeks retreated either into Miletus or beached along the coast of Mount Mycale.

The destruction of Miletus As the Delphic oracle predicted, according to Herodotus (Hdt. 6.19), in early 493 bc, Miletus was destroyed by the Persians and with it the oracular site of Apollo 22 Richard Evans at Didyma. After the Persians had concluded the fighting with the Carians, their strategy was to reoccupy the Ionian cities which had rebelled one by one and so they advanced overland to Miletus. Today the site of Miletus lies inland but for much of antiquity the city lay on a peninsula in the Latmian Gulf that has since silted up. In late 494 bc, an army could approach the city only from the south. This must also mean that Didyma must have been occupied before the Persians could begin their siege. Once the battle of Lade was lost, Herodotus gives some detail about the aftermath as it affected certain of the losing participants, such as the Chians, massacred in error perhaps, outside Ephesus and the journey into exile of many of the Samians. Yet there is no coverage of a siege at Miletus, implying possibly that the fall of the city occurred very soon after the naval defeat. Still, there is some evidence to indicate a more complex situation than Herodotus’ narrative presents, which runs contrary to his claim that the city was actually left in a ruinous state. And if Miletus escaped relatively unscathed by the defeat of the Ionian fleet, then Didyma most probably did too. Thus, some Milesians may have escaped and resettled, temporarily if not permanently, in Zancle (Thuc. 6.4.5–6: ‘Ionians who sailed to Sicily to escape the Persians’). The main obstacle to Herodotus’ devastating picture, however, is the obvious presence of Milesians in the Persian forces at Mycale in late 479 (Hdt. 9.99),24 and the city’s subsequent membership of the Delian League after 478. It is therefore highly likely that only known and active supporters of Aristagoras’ rebellion were actually removed from the city in 493 and that the city itself was quickly re-established. Delphi’s supposed warning about Milesian women washing the feet of the Persian victors was at best an exaggeration and Herodotus must have known this. If Herodotus is guilty of embellishing the fate of Miletus at the end of the Ionian War, then the destruction of Didyma, which he also claims was foretold by the oracle, is, on balance, not likely to have occurred at this time, but can be safely redated to 479 simply because of its pro-Persian sympathies in the past. It is non- sense to suppose that Didyma would have been destroyed when it was not associated with the rebellious Ionian Greeks. If the fabric of the city was spared, then its nearby shrine was also left untouched. Why Herodotus gave so dramatic an ending to the Ionian Revolt will never be known, but there may have been some personal hostility towards members of the Milesian elite such as Aristagoras, although the latter was already dead by the time the Persians recaptured Miletus. Herodotus certainly considered Aristagoras a ‘worthless’ person for the manner in which he chose to desert Miletus before the end of the war. Yet Aristagoras must have had some positive character traits to win over so many of the Ionian cities to his self-proclaimed cause. Moreover, Herodotus’ account of the Delphic oracle’s prediction of the end of Didyma may be questioned further. The oracle had insisted that the shrine would be the concern of others but does this mean reassignment of the land to new ownership or simply that the Persians established direct control from Sardis? The fact that Herodotus states that the land above the plain was granted to Carians while the Persians occupied the plain itself led Hammond (1998) to assume that the Carians were given control of the site at Didyma. The juxtaposition of ‘Carians’ with the Was Didyma (Branchidae) a false oracle? 23 destruction of Didyma in the narrative forces an assumption that they occupied the oracular site. But Didyma does not belong to the land above the plain and it was on the sole overland route through the plain from Miletus into Caria and Lydia. It is far more likely that the Persians turned over much of the territory of Miletus to the care of the Satrap of Lydia, and that peripheral land on Mount Grion was leased to farmers who lived in adjacent Caria. If Miletus had been destroyed, it would not have had control over Didyma in the period between the defeat at Lade and its resurgence after Mycale. If so, who took control of Didyma? Was it in fact the Carians, as Hammond seems to have accepted as implicit in Herodotus’ statement, although ‘above the plain’ does not suit the geographic location of this sanctuary? Moreover, would Darius have rewarded the Carians of Pedasa, who had proved to be some of the most obdurate of Persian enemies in the Ionian War with the ‘most famous’ oracular site in Western Asia Minor?25 Even considering the Persian king’s conciliatory temperament, Hammond has overlooked both the geographical evidence and the role of the Carians in the recently concluded war. Herodotus claims that the oracle was fulfilled, and describes the end of Didyma very dramatically: ‘shrine and temple and oracle’ were ransacked and burned. There is surely no doubt that the historian wished to convey this message, but the question of whether he thought he was retelling a fact or inventing or perpetuating a myth cannot be answered. Whatever, the condition of Miletus after the siege in 494, it returned to its status as a subject city of Persia within the Lydian satrapy just like the other cities of Ionia, and so too, presumably, Didyma.

The massacre of the Branchidae While this subject is not central to the discussion here, the historicity of Alexander’s massacre of the descendants of the Branchidae, as related by Curtius Rufus (7.5.28–35), is the subject of much debate and is an element that needs to be addressed since it does have an impact on the worth of the oracle and the activities and loyalties of the temple keepers.26 Tarn’s argument for regarding as an ancient invention the massacre of the Branchidae – earned for their ancestors’ ‘treacherous’ behaviour in sacking the temenos at Didyma, which was their charge, and then taking refuge with Xerxes in 479 after the battle of Mycale – hinges on the ‘rule’ that a late unsound source should not take precedence over a reliable early source.27 Besides Curtius, therefore, Tarn has also to include Diodorus, Strabo and Plutarch in this category while he omits Pausanias and Aelian, who also mention the affair. Herodotus’ evidence is therefore to be preferred, but is this evidence as secure as Tarn argues? Herodotus is never a thoroughly reliable historical guide. Two examples ought to suffice: the Samians are said to have occupied ‘the beautiful town of Zancle’ (Hdt. 6.24) but Herodotus surely knew that these settlers were expelled – or most of them at least – within two or three years by Anaxilas of Rhegium, who installed new citizens; while the claim that the Ionian cities were destroyed and their inhabitants deported (Hdt. 6.31–2) is simply a gross gener- alization at best and untrue at worst. Herodotus was probably employing poetic elements rather than a good historical source.28 24 Richard Evans There were certainly other writers who covered the Ionian War, some earlier than Herodotus, one or more of whom could have been his source since Book 6 of his history cannot have been written earlier than the mid-430s, already fifty years after the event. Herodotus did not write his history from memory and the Ionian War was concluded before he was born, hence his reliance on these earlier sources. The nature of these sources is difficult to categorize or even identify. Currently, it is thought that his sources were mostly of an oral rather than a literary nature although it is also recognized that there is a range of written material that he could have accessed during the composition of his work.29 Hecataeus is the most likely written source for events in Ionia before and during the war and Herodotus will not have found much if anything other than hostile information about Didyma and the Branchidae in this material, as is obvious from the discussion above. Other works like those of Xanthus or Hellanicus may have thrown a quite different light on events, while others again, such as a playwright or a poet, would have emphasized the dramatic aspects. Herodotus used Aeschylus’ The Persians for his account of Marathon, and referred to at least one epitaph cred- ited to Simonides in his account of Thermopylae. Herodotus could easily have employed Phrynichus’ Fall of Miletus for events in the Ionian War. Such dramatic versions could account for the elaborate and highly theatrical account for the apparent destruction of Didyma in 494–493. But it was not history, since any destruction at Didyma can only have occurred in 479, after the battle at Mycale and in response to this Greek victory and the reoccupation of the site by the victors.30 Didyma’s century-long loyalty to the Persians at that point would inevitably have led to revenge by the Greeks and not the other way around. The absence of oracular utterances from Didyma may have something to do with Herodotus’ audience, of course, although it seems unlikely that Athenians would have found reference to this site as emotionally distressing as they apparently found Phrynichus’ tragedy at the end of the 490s. But if this tragedy had contained material about treachery and betrayal, and ended with the Milesian women washing the feet of their Persian conquerors, the population forced into exile and Didyma destroyed, it is not inconceivable that Herodotus would have incorporated this tale into his own narrative. He refers to Phrynichus’ play (Hdt. 6.21) and concludes by stating that ‘Thus Miletus was emptied of its citizens’. Yet Miletus simply could not have been deserted from 493 and so it seems that tragedy and history became muddled and, writing fifty years or more after the event, the historian chose a more dramatic but also less accurate account.31

Conclusion The discussion here has to some extent highlighted the vulnerability of Herodotus’ evidence for a study of the Ionian War. Notwithstanding that conclusion, this weakness does not impinge on the question of whether or not Didyma had come to be considered a false oracle by that time, and thus its utterances of no use to the Greeks. Therefore, it seems unlikely that there would be any record of oracles from Didyma when there was a general uprising against Persia, since no requests Was Didyma (Branchidae) a false oracle? 25 would have been made because of the shrine’s long-standing loyalty, first to the Persian backed tyrants and, second, to the Lydian kinship. It is also not irrelevant to suggest that the lack of reference to oracles in Herodotus Book 5 might reflect the notion of Ionian rationalism compared to earlier Lydian complacency about divine intervention, and that, granted the prominence Hecataeus appears to occupy as a source, this could well have influenced the methodological approach at this point in the Histories. Still, the fact that Hecataeus failed to convince the Milesians, even the otherwise unscrupulous Aristagoras, suggests that as yet the plundering of temple treasures remained beyond the bounds of acceptable action. Elsewhere, before and after the Ionian War, the pervasive supernatural features witnessed in the Marathon campaign and the prominent role of oracles in the later invasion of Greece by Xerxes seem to indicate that Didyma’s was silent. With regard to the charge of ‘treachery’, who was guilty? The Branchidae were killed because their ancestors were said to have behaved in a treacherous fashion and this is supposed to have been manifest in the surrender of Didyma to the Persians. But Didyma had never been sympathetic towards to the Greeks and if there was any guilt regarding treachery it was the consistent loyalty to Persia, at first a friend and overlord but later the enemy. Curtius (7.5.31) claims that some of the Milesians were hesitant about the guilt of the Branchidae but he does not mention why. Perhaps it was because they remembered that some of their ancestors had been ready to pillage temple treasures.32 The theme was treachery and betrayal and so, in Curtius’ view, or that of his source, this crime was not one of simply siding with the Persians but had a much greater significance, and it is possible that, in revenge for Milesian contempt of Didyma, the Branchidae had played some role in the sack of Miletus in 494 or in the defeat at Lade. Moreover, it is possible that the oracles from Didyma came to be recognized as poor and unreliable. It was convenient that the spring associated with the delivery of the oracles in the sanctuary dried up and that oracles were no longer available after 479, and it seems at least likely that the Branchidae were responsible for blocking up the oracular spring when they fled from Miletus. It seems likely too that the oracle continued to function under Persian patronage after 494 and that the renaissance of the shrine had more to do with Alexander than the physical functioning of the cult. In conclusion, it seems unfair to have described the actions of the Branchidae as treachery or betrayal when they surely knew about the speech delivered by Hecateaus even before the Ionian War commenced and that the Milesians, or some of them, would have been quite prepared to plunder the treasury at Didyma to fund their rebellion.33 The fact that Herodotus recounts this also indicates that there must have been some animosity in the relationship between the shrine and Miletus. However, this state of affairs cannot be traced back to the Pactyes episode, when the oracle demanded that the Cymaeans return their suppliant to the Persians. Herodotus appears again to present a misleading picture since Pactyes was not an Ionian Greek but a Lydian and hence a subject of the Persian Empire. Jurisdiction over his case belonged not to the Ionians but to the Persians. The oracle therefore behaved impeccably but it is as well to remember the context. 26 Richard Evans Miletus had come to terms with Cyrus but the other Ionian cities chose to fight for independence following the collapse of Lydia and were then incorporated into the Persian Empire by force. In that case, therefore, the oracle and the city acted together and at that stage both were pro-Persian. It was only during the consultations for the start of war in 500 that the allegiance of the city and the shrine diverged. The Branchidae could justifiably claim betrayal by sentiments in the city even if the sanctuary’s treasury remained untouched during this war. At least, it is not recorded that the Milesians raided the treasure. If Hammond is correct, Herodotus then maintains his misleading impression by heavily underscoring his description of the end of Didyma in 494, when Darius may well not have ordered such an act. If the Branchidae had been threatened in 500, this could have been because of the oracle’s perceived pro-Persian stance, which is likely to have been regarded as having a negative impact on the rebellion of Miletus. As a result, leaders such as Aristagoras would have avoided this ‘false’ oracle. Treason seems evenly divided between Miletus and Didyma, and, unfortunately for the Branchidae, the Milesians or their admirers wrote the history to their perpetual disadvantage. Brown (1978) may be partly correct in his argument but the Branchidae were perhaps less guilty than Aristagoras and the Milesian elite. Didyma became a false oracle for the new democracy at Miletus since it had long links with the Milesian tyrants, Lydia and, ultimately, the Persian king. No democratic leader, even a turncoat tyrant like Aristagoras, could possibly hope for favourable messages from Apollo at Didyma and he would have been wiser to seek out other oracular centres, such as Delphi. This commitment to the tyrants and their link with Persia explains the Milesian leaders’ antipathy towards the sanctuary in 500 although they did not adopt the ruthless move suggested by Hecataeus, however sensible it may have been. The failure to support the rebellion against Persia must be the ‘betrayal’ perceived by Curtius and the reason why Alexander took violent revenge on the descendants of the Branchidae. Alexander had destroyed Thebes in part because of its Medizing; the Branchidae were slaughtered for the same crime of collaboration committed by their ancestors. Finally, it should also be noted that Herodotus was himself guilty of some distortion (Hdt. 6.19), deliberate or not, since he certainly redated any destruction of Didyma from the aftermath of Mycale to the capture of Miletus in 494, perhaps in order to enhance his own dramatic presentation of the events in Book 5.

Notes 1 The extent of dependency on oracular messages in the narrative may be gauged by noting references in Books 4 to 6 of the Histories. In Book 4, Herodotus (155–9) mentions no less than five Delphic utterances connected to the foundation of Cyrene, while in Book 5 there are several references to Delphi (Hdt. 5.43, 79, 82, 89, 92) but none relate to the war in Ionia. There are other references to Delphi (Hdt. 6.77, 6.86c, 6.139), but these are also not connected to events in Ionia. In the entire history, there are just two references to oracles from Didyma (Hdt. 1.46 and 1.57–9), which deal with Croesus and the revolt of Pactyes. On Pactyes, see below. Was Didyma (Branchidae) a false oracle? 27 2 However, Herodotus (6.98) illustrates his belief in the supernatural by dwelling on an earthquake on Delos said to be an indication of the troubles ahead, and this notion becomes a vivid and integral element in the Marathon campaign well before the Persian fleet arrived in Greece in 490. See further discussion in Cobet (1986) 11. 3 Herodotus’ text (6.18–20) raises some interesting questions about the events because he implies that a part of the oracle meant for Miletus was to be forwarded by the Argives. There is no evidence that it was ever sent, since by then it was too late to influence events; furthermore, could oracles be unsolicited or was this a unique event? Fontenrose (1978: 70–1, 169) notes that Herodotus is alone in stating that the oracle granted to the Argives also contained a message for the Milesians. Other sources – Pausanias, the Suda and Tzetzes – seem to be under the impression that there were two oracles. Fontenrose (1978: 169 n. 6) and others recognize the episode’s dubious historicity since there is no record of a Milesian delegation to Delphi and the Pythia would only have addressed a suppliant who was present. For the oracle specifically directed to the Argives see Hdt. 6.77. 4 There is no mention that he went to Argos before or after his failed bid to win Spartan aid although it is quite logical that he should have visited other cities in Greece besides Athens and Eretria. 5 Other important oracular sites were at Chryse in the Troad, Claros on the coast at Colophon and Gryneum north of Cyme to name just three. For Gryneum see Brown (1978) 70. Croesus seems to have regarded advice from Didyma as less reliable than that from Delphi or Amphiaraus (Hdt. 1.49), yet his gifts to the Asia Minor cult of Apollo equalled his donations to Delphi (Hdt. 5.36). 6 Brown (1978: 65 and n. 6) notes that Herodotus’ use of the verb κομίζειν leaves Pactyes’ position a little vague since some see this to mean that he was in charge of ‘transporting the treasury to Cyrus’ (de Sélincourt/Marincola, trans. 1996) or of ‘the administration’ (Godley, trans. 1920). Herodotus (1.55) states that Tabalus was a Persian and Pactyes a Lydian. For a discussion of their relative positions in the new province’s hierarchy, see Asheri in Asheri, Lloyd and Corcella (2007) 181 and Hornblower (2013). 7 Herodotus (1.155–6) claims that Cyrus blamed the Lydians and intended to sell them all as slaves, but was dissuaded from this drastic action by their former king Croesus. This tradition surely presents a negative picture of the Lydians, who do not in fact appear to have been involved at all, and perhaps emanates from a writer such as Xenophanes of Colophon. See Asheri in Asheri, Lloyd and Corcella (2007) 181. 8 The later rebellion in c. 500 was probably on a broader scale, but there are indications of unhappiness with Persian rule fifty years earlier. Herodotus (1.141–3, 148, 170.1) refers on a number of occasions to the meetings of the Ionian cities – the Panionium. However, Miletus, with its special relationship with Persia, chose to take little or no part, although the meetings were held on Mount Mycale and administered by Priene, the nearby city across the Gulf of Latmus. As in the greater Ionian conflict, Herodotus chose to highlight personal rather than sociopolitical or economic motives for Pactyes’ uprising. 9 Herodotus’ survey and enumeration of the Ionian fleet at Lade in 393 (6.8) does not include a contingent from Cyme. However, Cyme had contributed to the force that Aristagoras led against Naxos in 501–500 (Hdt. 5.37–8) and so would have possessed a fleet of its own, and indeed the city joined the rebellion. 10 Robinson (1997) surveys early examples of democracies outside Athens in the sixth century but makes no mention of Miletus, where there is plainly no record of this form of government before 501–500. 11 Croesus controlled Ionia, including Miletus, and when Cyrus conquered the Lydian kingdom in 545, by default he also acquired control of the Greek cities of western Asia Minor. Croesus bestowed gifts on both Delphi and Didyma, and others copied this patronage. 28 Richard Evans 12 Histiaeus was a member of Darius’ entourage during the king’s expedition across the Bosphorus and the Danube in 513–512. His loyalty was rewarded with a gift of the city of Myrcinus in Thrace. However, Darius’ general Harpagus warned the king of Histiaeus’ personal ambitions and as a result the tyrant was called to Susa. 13 Herodotus (5.37) refers to a state of ‘isonomia’ or an equitable interpretation of the laws. He also provides the primary cause for rebellion: Aristagoras’ failure to seize Naxos for the Persians, who had financed the expedition (Hdt. 5.30–5). 14 See n. 4 above. 15 Herodotus (1.155–6) relates that Croesus advised Cyrus on the best actions for the Persians to take against the Lydians supporting Pactyes, but this episode is not historical and the Lydian king may well have died at the end of the siege of Sardis. 16 Brown (1978) 77. Brown also notes (1978: 72) that Herodotus’ account shows that the oracle maintained control of the situation and that Aristodicus did not defeat the god. It seems unlikely that Xanthus could have been responsible for the extant account, which is plainly supportive of Didyma. 17 Evans (2015) 8–9. On the historicity of the episode and the likelihood that Histiaeus was not responsible for Aristagoras’ actions, see Georges (2000) 14. 18 His presence at this meeting indicates that Hecataeus was not only a member of the Milesian elite but that he was also close to Aristagoras and presumably also to Histiaeus. As such, Hecataeus was well placed to record the events that unfolded in the region, as is duly noted by Asheri in Asheri, Lloyd and Corcella (2007: 182). 19 This bears a striking similarity to the Delphic oracle’s advice to the Athenians (Hdt. 7.141), as interpreted by Themistocles (Hdt. 7.143), to rely on their war fleet as a defence against the Persian invasion of Greece in 480. 20 Philomelus is described as the author of this sacrilege by Diodorus (16.30.1, 16.31.4, 16.61.2). Note also the discussion of Buckler (1989: 38–9). Some sources suggest that it was not Philomelus but rather Onomarchus, his successor, who first pillaged the treasuries, but it is most likely that the latter emulated his predecessor and took the sacrilege to new heights, as suggested by Buckler (1989: 47). See Buckler (1989: 38 n. 10) for references to the discussion of Philomelus’ guilt and the ancient sources. The ‘Sacred Wars’, three in all, that were fought for control of Delphi, highlight the complex perceptions among the Greeks regarding the control of their sanctuaries and the wealth these contained. No such conflict is attested for the early control of Didyma for there the dominant powers, initially Lydia and then Persia, probably foiled local Ionian Greek rivalry in this area. 21 Neither Aristagoras nor Histiaeus was present in Miletus at the end of the hostilities. Histiaeus was captured and executed in the Troad. 22 The Thebans consulted Delphi (Hdt. 5.79), as did Epidaurus (Hdt. 5.82), Athens (Hdt. 5.89, 90), and Corinth (Hdt. 5.92b, 92e). Herodotus relates the military campaigns of the Cypriot cities and the Carians without any mention of oracular activity. However, a fragment of Diodorus (10.25.2) appears to contradict Herodotus and indicate that the Carians sought and obtained at least one oracle, perhaps from Delphi, and seemingly in the last stages of the Ionian War. The context suggests that Miletus was still a dominant force in the region and that the Carians needed allies. The latter had made peace with the Persians before the fall of Miletus, when considerable concessions were granted to the rebels since the Persians needed a pacified Caria before they could regain control of Ionia. See Evans (2015 38–9). Notably, Diodorus has nothing to say about Didyma in the context of the war. 23 The absence at this stage of Ephesus, Smyrna, Colophon and Clazomenae suggests that the Persians had retaken these cities or that, like Smyrna, they had not in fact been involved in the war. The attention to tactical details proposed by the Phocaean Dionysius (Hdt. 6.11–12) prior to the crucial battle, although ultimately unsuccessful because of the arrogance of certain crews, is possibly also indicative of Herodotus’ illustration of human artfulness over a dependence on religion. Nevertheless, naval training must have Was Didyma (Branchidae) a false oracle? 29 been equally important and more decisive in subsequent naval battles against Persia rather than simple or naïve trust in oracles. 24 This discrepancy has long been noted. See How and Wells (1928) 2.330; Marincola (1996) 601. 25 Hammond (1998) 339–44. 26 Hammond (1998: 339–44) discusses the ancient references for the massacre of the Branchidae. It seems to me that Hammond has shown conclusively enough that the Branchidae migrated to Sogdiana and were killed by Alexander. But he overlooked two significant points that support his assertion. The first is the date at which the Branchidae abandoned Didyma, and just who had overall control of the temenos between 493 and 479. 27 See Tarn (1922) 63–6. Cf. Tarn (1948) 2.272–4, whose argument that Callisthenes (Strabo 17.1.43; Jacoby, FGrH Callisthenes fr. 14) had invented the massacre dominated the debate until twenty years ago. Thus Hammond (1980: 298) notes that the episode ‘is generally regarded as unhistorical.’ However, opinion has now changed. See Dascalakis (1966) 221–3; Brown (1978) 64–78; Parke (1985b) 59–68; Bosworth (1988) 108–9; Flower (2000) 117–18. Didyma was conveniently reborn in 333, when it again provided oracles, primarily to Alexander. See Worthington (2003: 48). Both Jouguet (1928: 86) and Radet (1950) place the destruction in 493–494 and then the treacherous behaviour of the Branchidae in 479. For Callisthenes as the original source for the episode see Brunt (1965) 205–15; Worthington (2003) 48. 28 All the Ionian cities were functioning as civic entities in the 470s. Ephesus and, particularly, Smyrna do not appear to have been affected by the war. 29 For further analysis of Herodotus’ sources see, for example, How and Wells (1912– 1928) 1.21–36; Pearson (1939); Evans (1982); Fehling (1989); Marincola (2001); Cawkwell (2005). The debate clearly remains closely contested and without obvious conclusion, although the argument here is that some literary material must have caught Herodotus’ imagination. 30 Neither the destruction of the site at Didyma nor the desecration of the shrines is attested; they are assumed, although some structure must have been available for Alexander to visit. It should also be noted that the Branchidae may have fled their sanctuary believing that pillage was inevitable, along the lines of Xerxes’ objective after Thermopylae to capture Delphi as related by Herodotus. This episode is usually regarded as an invention created to excuse the pro-Persian leanings of the Pythia. Still, an intention to sack Didyma could have been presented as revenge for damage done to Delphi, although neither Herodotus nor any other source hints at this. The destruction of sanctuaries became a more regular event, perhaps as a result of the financial demands made on the warring states in the Peloponnesian War. The Syracusans fully expected the Athenians to pillage the temple of Zeus at Polichne when the invaders began their siege in 414. Olympia was sacked in 369 by the Arcadians, while Dionysius I sacked the temple of Eileithyia at Agylla (Diod. 15.14.3–4; Strabo 5.2.8). 31 The date of the play is unknown and both 492–491 and 479–478 have been suggested. For a discussion see Rosenbloom (1993) 159–96. The later date poses an interesting context since the Athenian audience would have watched a series of events played out over nearly a decade but been reminded of the brevity of their own city’s defence even if that strategy had contributed to ultimate success over the Persians. For a discussion of the relationship between Miletus and Sybaris, to which Herodotus draws attention (Hdt. 5.21), and the fact that the historian is again manipulating the evidence to suit his narrative plan, see Evans (2013). 32 Curtius expresses no doubt about the episode but asserts that it was unfair to kill the descendants of the Branchidae, who knew nothing of Miletus and Xerxes. Moreover, Alexander’s motives may have been ambivalent since the Milesians’ presence was not necessarily voluntary. Alexander had besieged Miletus and great damage must have been done to the city as a result. The Milesians may well have been serving with the 30 Richard Evans Macedonians to ensure the good behaviour of their fellow citizens at home. If the episode has any basis in historical fact, it was therefore much more complex than is evident in Curtius’ brief notice or in modern scholarship thus far. The charge made by Alexander against the Branchidae, as reported by Callisthenes (fr. 14), was sacrilege and therefore the same crime committed by the Phocians in the Third Sacred War (Diod. 16.35.5). Had the Milesians followed the advice of Hecataeus to remove the temple treasures from Didyma, their city may well have been destroyed by Alexander. 33 Note also Hecataeus’ negotiations with Artaphernes at the conclusion of the war, as related by Diodorus (10.25.4), again hinting that the so-called destruction of Miletus was not as dramatic as Herodotus would have his audience believe. 3 Who wrote Greek curse tablets?1

Olivier Dufault

Many scholars of ancient Greek religion would probably agree that the use of curse tablets in the ancient Mediterranean world ‘cut across all social categories’.2 In practice, however, it is also a common working assumption that the use of curse tablets was typical of the non-elite (i.e. of those who did not actively participate in politics and could not accumulate surplus wealth).3 I believe this last assumption to be dominant in part because these two hypotheses can easily turn out to be the same: if one were to think that curse tablets were produced at a similar rate by each social class, one would also have to assume that non-elite classes wrote or commissioned most curse tablets. This is indeed the assumption which readers are sometimes expected to make.4 In the following, I argue that the bulk of published Greek curse tablets, which come from Attica in the fourth century bce, do not support this assumption. To test the assumption that, generally speaking, the writing of Greek curse tablets from the classical and Hellenistic periods was a non-elite phenomenon, I had to reconstruct the missing argumentation. This reconstruction forced me to consider a subsidiary question: since it is generally agreed that illiteracy was common in ancient Greek societies, it would be impossible to assume that curse writing spread across all social classes unless we also imagined that those who were illiterate or insufficiently literate acquired tablets from literates. In other words, the claim that the use of curse tablets cut across all social categories is inextricably linked to the idea that those without sufficient writing skills acquired curse tablets from those possessing an adequate level of literacy.5 It is not surprising, therefore, that scholars typically assume that specialists wrote many if not most curse tablets.6 Consequently, I will also consider the assumption that curse tablets were generally written by professional curse writers, and I will argue that this second assumption is similarly impossible to substantiate for the classical and Hellenistic periods. The goal of this study is to address the question of the social distribution of curse writers directly and to suggest at least two different avenues of research. On the one hand, the evidence of curse tablets could support the hypothesis that democratic institutions stimulated the expansion of literacy in Athens. It could well be that Athenians in the classical and Hellenistic periods were exceptionally literate and that people from all classes came to see the writing of curse tablets as 32 Olivier Dufault an effective way to solve problems. The fact that most of the earliest curse tablets come from Attica could be a manifestation of exceptionally high literacy levels in Athens. On the other hand, if one is to assume that ancient Greek or Latin literacy was connected with the possession of wealth or political power (an assumption that must be partly accepted by those assuming that democracy influenced literacy levels positively), one should rather consider the hypothesis that the writing of curse tablets was typical of ancient literate milieux. I have chosen to explore the second avenue of research. Richard Wünsch bought most of the tablets studied here in 1894 from a certain Rhousopoulos, who acquired them in unknown circumstances. The majority of the tablets date from the fourth century bce and come from Attica, and were published by Wünsch in the Defixionum Tabellae Atticae (DT). They are currently being re- edited by Jaime B. Curbera.7 In addition to these works, I have consulted Auguste Audollent’s Defixionum Tabellae (DTA), which includes Greek tablets that were missing from Wünsch’s edition. I have also consulted David R. Jordan’s cata- logues (SGD and NGCT) as well as the Supplementum Epigraphicum Graecum, which cover the rest of the evidence published up to 2006. The most common way to classify curse tablets is to divide the material according to the occasion for which they are assumed to have been written. Curse tablets are usually classified into four different categories: (1) ‘relationship’ curses, also called amatory or erotic curses; (2) ‘judicial’ curses (a group of curses usually binding the linguistic abilities of a target and often set in a judicial context); (3) ‘competition’ curses, used in the context of dramatic competitions and games; and (4) ‘commercial’ curses targeting businesses. The fact that curse tablets from the last category targeted shopkeepers (κάπηλοι), women, servitors and craftsmen must have contributed to the impression that curse tablets cut through all social categories. This material, however, represents only a fraction of the evidence. Moreover, the four categories do not inform us about cursers since they did not necessarily targeted individuals from their own social class. Even if they did, and if the use of classical and Hellenistic curse tablets offered a true cross-section of Athenian society, we would expect to find references to peasant or agricultural work. But these are conspicuously absent from the tablets.8 In fact, most classical and Hellenistic curse tablets (around two- thirds) only list names and do not provide enough information to be classified into any of the four categories above. Curse writers rarely referred to themselves and the literary tradition does not explicitly discuss the writing of curse tablets (see below). In sum, curse tablets tell us next to nothing about the social context in which they were written. This observation emphasizes the importance of an obvious fact: those who wrote curse tablets were literate to some degree. I see two ways in which curse writers could have come from every social classes. First, we would need to believe that literacy was widespread in Athens and that high levels of literacy were achieved there (and elsewhere) in the first century bce, the time at which the simplest of Greek curse tablets (i.e. those listing names and lacking in sentence structure) stopped being written.9 The second and more realistic hypothesis Who wrote Greek curse tablets? 33 would be to consider that curse writing was the product of a specialized trade was run by literates. Ancient literacy plays a fundamental role in this argument and I will (1) discuss this issue before looking at (2) literary and (3) epigraphic evidence that could tell us more about the social origins of curse writers. Finally, (4) I will present the three sets of tablets that were most certainly written by professional curse writers. These all date to the second century ce or later. In short, I argue that the assumption that curse tablets were generally written by professional curse writers who worked on behalf of the non-elite cannot be substantiated for the classical and Hellenistic periods. Rather, the evidence suggests that the professionalization of curse writing was a post-Classical phenomenon.10

Ancient Greek literacy From a comparative perspective, it would be surprising if high levels of Greek literacy had been achieved by all social classes in classical and Hellenistic times.11 There are, however, different types of literacy since people learn how to read and write for different reasons and in different situations.12 Since two-thirds of classical and Hellenistic curse tablets only list names, ‘name literacy’ (the ability to read and/or write names) could have been sufficient to write the majority of ancient Greek curse tablets. To know who could have written these curse tablets, we consequently need to evaluate the spread of name literacy in ancient Athens. The level of name literacy can be approximately gauged by the fifth-century bce practice of ostracism in Athens. If each person wrote their name individually, the ostracism quorum of 6,000 suggests that at least 20 per cent (6,000 out of c. 30,000) of Athenian citizens were expected to possess name literacy. Again, if we assume that democratic institutions and wealth increased the chances of obtaining higher levels of literacy, we can further assume that this represents the minimum level of Greek literacy possessed by Athenian citizens in the fifth century bce. It is more complicated to estimate the name literacy rate of the total population of Attica. First, population estimations for the region in the fourth century bce, which are based on the estimation of the total number of citizens, vary between c. 75,000 and c. 250,000.13 If we assume that for each Athenian citizen there was one woman and two adult metics or slaves, we can estimate the total adult population at an average of 120,000. The minimum percentage of name literacy among the inhabitants of Attica would then have been somewhere around 5 per cent (6,000 citizens out of a total population of 120,000). Following the same reasoning, we could also concede name literacy to all those who were part of the hoplite class. The percentage of the population of Attica able to write two-thirds of classical and Hellenistic tablets (i.e. name-only tablets) thus represented probably somewhere between 5 per cent (the minimum percentage of the population of Attica with name literacy) and 7.5 per cent (the proportion of the population of Attica expected to have been of hoplite class or higher).14 Simply considering the practice of ostracism, we can estimate that two-thirds of classical and Hellenistic curse tablets from Attica could have been written by 7.5 per cent of the population or less. 34 Olivier Dufault These estimates depend on the hypothesis that Athenian citizens took democratic institutions seriously enough to learn how to read and write. There are signs that the democratic regime of Athens might not have increased the general level of literacy. Athenians did leave many public inscriptions, and certainly expected some to read them, but they did not take public measures to spread the use of writing. In fact, it was not necessary to possess any level of literacy to participate in basic Athenian democratic institutions.15 Arguments based on ostracism should also be qualified by epigraphic and literary evidence showing that some ostraka were pre-written for illiterate citizens.16 In any case, if we are to assume that around 5 to 7.5 per cent of the Athenian population could have written name-only curse tablets – and which disappeared by the first century bce – it would be difficult to affirm that the ability to write curse tablets cut through all social classes and through all periods. The poorly drawn letterforms, aberrant orthography and deficient grammar found in some private inscriptions and curse tablets do not prove that basic literacy was widespread in Athens since non-official writing standards are unknown. As Mabel Lang has shown, the letterforms and orthography of almost all types of official and non-official writing fluctuated during the fifth century bce and only stabilized towards its end.17 This means that writing in Attica around the fourth century bce had been fixed only three or four generations earlier. If even those who had experience with writing did not think much about orthographic norms in non-official writing, we cannot take variations in handwriting, orthography and grammar seriously when establishing literacy levels. The obscurity and sheer illegibility of some tablets can also give the misleading idea that curse writers were barely literate. Many curse writers throughout antiquity apparently thought that curse texts should be encrypted or at least anomalous.18 In classical and Hellenistic times, this mostly involved the misplacement or reversion of words and letters.19 The practice of sealing tablets with nails was perhaps a result of the assumption that reading a curse would dispel its power, as mentioned on a fourth-century bce tablet from Pella.20 Other curse writers might have attempted to make their curses difficult or impossible to read in order to ensure their efficacy.21 The unpredictable nature of curse texts could have induced Wünsch’s assumption that some curse writers were barely literate. For instance, the text found on DTA 66, which Wünsch said was ‘written by an illiterate’, is strangely placed and sometimes leaves out or misspells vowels. But its handwriting does not appear to be less practised than those of most tablets from the oracle of Dodona (c. 500–250 bce)22 or than those from a fourth-century cavalry archive from Athens.23 There are other reasons to believe that letterforms, orthography, syntax and the general appearance of curse tablet texts are not good indicators of their writers’ literacy levels. The lead tablets recording consultations of the oracle of Dodona all appear to have been written by different persons and show a general fluidity in orthography and grammar.24 Some consultants wrote on behalf of communities, some planned to acquire ships or to do business and could have been educated. At least one appears to have been a peasant, and one a fisherman.25 With the exception Who wrote Greek curse tablets? 35 of one writer who appears to have been exceptionally learned, the spelling and grammar of the tablets do not follow the rules of official inscriptions.26 In other words, the range of writing styles, orthography, grammar or spelling found in classical and Hellenistic Greek curse tablets is similar to that found in comparable inscriptions, such as the oracular tickets from Dodona and the archive of the Athenian cavalry. Neither of these sets are representative cross-sections of Athenian society or of the wider ancient Greek-speaking world. If anything, they show that the writing of texts that were not supposed to be put up publicly in a city did not usually follow strict writing norms. Letterforms also appear to be a particularly poor indicator of the social origins of curse writers. For instance, we would expect that the handwriting found on DT 66, categorized as ‘ignorant of letters’ by Wünsch, would have been inferior to that of an Athenian cavalry archive. The only reason for this judgement appears to have been the misspellings found on the tablet and the strange disposition of the text. The first anomaly is in fact common to all types of non-official inscription and the second might have been an encryption method. We will have to wait for the publication of Curbera’s new edition of Wünsch’s tablets to judge the level of literacy of classical and Hellenistic curse writers with more precision. For the time being, considering that the general appearance of curse tablet texts cannot give reliable information about the social origins of their writers, we should turn to the Greek literary tradition to see if it can help determine who wrote Greek curse tablets.

Literary sources and curse tablets Athenian women – especially courtesans and procuresses – were often associated with cursing in classical and Hellenistic literature.27 Greek literature, however, always represents these women using non-literary cursing techniques. As I will argue in the next section, this is not specific to the representation of ancient witches. While references to curses or bindings are relatively common in ancient Greek texts, no explicit mention of curse tablets can be found in texts from the classical and Hellenistic eras. In fact, the only representation of a professional curse writer, Pamphile, in Apuleius’ Metamorphoses (3.17), dates from the second century ce.

Curse tablets and the idiom of ancient Greek cursing Katadesmos, one of the technical terms now used for curse tablets, comes from the verb καταδέω, which primarily means the action of physically binding something to something else. In Homer (Od. 5.383–5, 7.272, 10.19–24) the verb can also denote how divinities ‘bind’ the ‘path of the winds’.28 The verb was used on a third of legible tablets from Wünsch’s collection and it must have been intended in the sense found in Homer.29 It is certainly because of the presence of the verb in curse tablets that Plato’s use of its past participle (καταδέσμοις) in Republic 364 is sometimes understood to refer to the writing of curse tablets rather than to the action of 36 Olivier Dufault stopping or binding. In support of this interpretation, scholars have also noted the occurrence of the past participle form of καταδέω in the collection of recipe books from late antique Egypt known as the Papyri Graecae Magicae (PGM).30 A first problem with this reading is that we cannot assume that the practice of curse writing in Attica around the fourth century bce was similar to that found in Egypt six or seven hundred years later.31 Moreover, when the PGM writers and a late antique curse writer denoted a curse tablet, they used κάτοχος rather than κατάδεσμος.32 Elsewhere, Plato used the verb καταδέω to refer to the binding of the soul to the body. Once again, the reference to the Homeric use of the word seems more likely although Plato did add an important detail. The soul, Plato wrote (Phaedo, 83d; Timaeus, 73c) is bound (καταδεῖται) to the body through pleasure and pain ‘as though with a nail’ (ὥσπερ ἧλον). With this metaphor, Plato was most probably alluding to cursing, but it would be difficult to tell which type since curse figurines were also sometimes run through with nails. Moreover, even if this practice was relatively common, less than half of classical and Hellenistic curse tablets were found sealed with nails.33 While the use of figurines and spoken words were common features of ancient Greek cursing,34 specific mentions of curse writing are absent from classical and Hellenistic sources. The evidence suggests that καταδέω and cognates could des- ignate the act of cursing in general rather than the act of cursing through writing. We need to look into the Latin tradition for the first explicit references to written curses.35 The only ancient term for curse tablets comes from the aforementioned PGM, a collection of recipe books from late antique Egypt. The word used there, however, was always κάτοχος.

Professional curse writers in literary sources The text from Plato’s Republic mentioned above is one of only two passages that could be used in support of the assumption that professional curse writers were common in classical and Hellenistic Greece. In the Republic, Socrates’ interlocutor Adeimantos mentions the activities of ‘itinerant priests and seers’ (ἀγύρται δὲ καὶ μάντεις) who are said to have peddled ‘enchantments and bindings’ (ἐπαγωγαῖς τισιν … καὶ καταδέσμοις) to the rich and to have claimed that they could convince the gods to serve them.36 Adeimantos adds that these individuals also brought forward books attributed to Orpheus and Musaeus and convinced citizen bodies and individuals of their capacity to purify them from wrongdoings (ἀδικήματα), or from those of their ancestors (364e–365a). It is unproblematic to claim that Athenian religious professionals would have ‘driven the practice’ of curse writing,37 or that this practice could involve the work of ‘different kinds of specialists’.38 It is less so, however, to base the argument that curse tablets were usually written by religious professionals or ‘so-called magicians on the evidence provided by Plato.39 Support for this theory can also be sought in a passage of the Laws (932e–933e) concerning healers and religious professionals injuring others with φάρμακα (‘drugs’). Under that heading, Plato classified a first type of offence ‘in which Who wrote Greek curse tablets? 37 injury is done to bodies by bodies according to nature’s laws’. The second group includes cases in which harm was done ‘by means of trickery, incantations and by what are called bindings’ (ἣ μαγγανείαις τέ τισιν καὶ ἐπῳδαῖς καὶ καταδέσεσι λεγομέναις). Those targeted by Plato here are μάντεις and τερατοσκόποι (‘seers’ and ‘interpreters of portents’), whom he assumed to have harmed others through the use of the second type of φάρμακα. Plato must have been referring to the same individuals he targeted in Republic 364. The exact meaning of κατάδεσις, however, is difficult to ascertain. Apart from the passage in Plato’s Laws, it occurs in only two other texts, where it refers to knots.40 The meaning appears to be very similar to that of κατάδεσμος. Again, nothing indicates that Plato was referring specifically to written curses. The problem is not simply that the words used by Plato do not specifically refer to curse tablets. By implicating diviners and ‘itinerant priests’ (i.e. ἀγύρται, a term whose meaning is far from clear), Plato was making indirect accusations that supported his claims and his position in Athenian society. The cathartic rites that Plato ascribed to religious professionals contradicted the theodicy he presented in his dialogues since their presumed efficacy would have enabled one to commit injustices without facing universally sanctioned consequences. Those whom Plato opposed provided means to cope with past and present injustices in the same way that his theodicy provided readers with a reason to pursue a life in accordance with justice. The distance created between Socrates and those whom Plato criticized was perhaps not as great as it might seem. Like Plato’s Socrates and Plato himself, those criticized by Adeimantos also had recourse to myths to frame their claims about the afterlife.41 Trials for impiety in Athens were often related to the Eleusinian mysteries42 which, for Plato at least (Gorgias 493a–b), were related with the afterlife. We can expect that these trials raised the stakes of engaging critically with popular soteriological practices. Moreover, if we are to follow the dialogue to the letter, the Athenians did not consider the so-called diviners and itinerant priests to be charlatans; on the contrary, Greek cities welcomed them and celebrated their rites.43 Plato appears to have been speaking against religious professionals who garnered popular support and it is probable that the citizens who profited from their rites were not ready to accept Plato’s criticisms or the ethics he advocated. This antagonism suggests that Plato’s statements that diviners practised bindings were not incidental remarks about Athenian society. On the contrary, it was particularly clever of Plato to suggest that those who pretended to remove ἀδικήματα – and which manifested themselves in the form of a curse – would have also sold curses.44 His accusations should be read in the same light as any rhetorical defence. That he made a similar association in the Laws is not further evidence but a repetition of a similar claim in a more explicitly normative tone.

The social distribution of curse writers and epigraphic evidence Literary texts do not provide us with evidence to back up the double assumption questioned here, namely, that curse tablets were generally written by professional 38 Olivier Dufault curse writers who worked on behalf of the non-elite. Curse tablets do not give more conclusive evidence. As argued above, inferences drawing upon the handwriting, orthography or grammar of curse tablets are unconvincing. To infer social class based on literacy levels implies a drastic simplification of reality; jumping from handwriting alone to social class is even less likely to be accurate. Another problem is that Wünsch and Audollent only informally gauged writing proficiency. It is also problematic that they mixed writing proficiency with other criteria, such as orthography and grammar, to describe the education level of writers. Of the complete and legible tablets from the DT, Wünsch noted that only three or perhaps five writers were particularly experienced, while three were not.45 One might assume that Wünsch studied differences in handwriting but did not systematically annotate the tablets since he also claimed that most writers ignored grammatical rules.46 Surveying the same corpus, Curbera mentioned the presence of many clumsily written tablets and cited four examples.47 He also noted that the DTA inscriptions were similar to those found on monuments.48 From the reports of these two scholars, it appears that the handwriting of the majority of classical and Hellenistic curse tablets from Attica was unremarkable. We can conclude that most of those who wrote the tablets had some experience with writing. Scholars mentioning the large distribution of handwriting types found in curse tablets usually cite three such tablets. Statistically, this small and mixed sample does not tell much about the penetration of curse writing in Attica in the classical and Hellenistic age.49 To get a better idea of the social distribution of cursers, we can also look at the content of the curses themselves. While we can assume that curses were aimed at people known by those who wished them ill, the targets were not necessarily competitors. The mention of the target’s craft, for example, could have been a way to identify them. As Esther Eidinow has pointed out, it is unlikely that curse tablets were mainly a result of the agonistic nature of ancient Greek society when curses meant to influence the result of dramatic or athletic competitions represent only a fraction of the evidence.50 In any case, to assess the only argument that could support the idea that the use of curse tablets cut across all social classes, it is necessary to follow the assumption that cursers at least shared the same social environment as their targets. I have separated these curses into three groups: curses related to theft, curses related to craftsmen or to service providers and curses related to lawsuits. I have rejected curses related to relationships and those related to competitions. The number of tablets related to competitions is too small to be representative of any class interest. I have also assumed that relationship curses could be intended for almost any target and not necessarily someone belonging to the curser’s social milieu.

Curses related to theft Curses related to theft give tantalizing information about the socioeconomic origins of cursers.51 Some stolen items, such as bathing clothes or cloaks, might Who wrote Greek curse tablets? 39 suggest that the cursers were not particularly rich.52 Market price, however, was not the only value that these garments could have. Curses mentioning the theft of money provide a better way to assess the economic level of cursers. DTA 42 from Megara, undated but certainly written after the Roman occupation, cursed an individual who accused the curser of borrowing twenty denarii. This seems to have been a relatively small sum for the first century ce.53 The other example, DTA 212 (third century bce, found in Bruttium) involves the theft of three gold coins. This category might show a relatively wide social distribution depending on the various degrees to which the economy was monetized in these two different places. In regions where coins were not usually circulating, the simple possession of coinage would suggest a wealthy individual. However relevant the comparison with theft-related curses might be, the proportion of classical and Hellenistic tablets in this group is insignificant (14 out of the 156 which provide enough information besides the name of the targets; i.e. 9 per cent).54 Consequently, it does not offer a representative sample of the spread of the use of curse tablets among different social groups.

Curses related to craftsmen or service providers These curses, the few tablets that targeted artisans, shopkeepers (κάπηλοι) and prostitutes, appear to illustrate non-elite interests if we assume that they were written by professional rivals.55 Curse tablets binding establishments (καπηλεῖα) or their managers could have been written by people frequenting these places, by corporations as well as by rival owners or shopkeepers. However, if naming a trade was a way to identify targets, we do not know whether the curser targeted them for their business unless he or she targeted the work itself (ἐργασία). Even if tablets were intended to impede competing businesses, they were not necessarily written by owners of small shops or by their employees. One of the few commercial examples – a fourth-century bce tablet from Athens (DT 87) – curses a neighbouring shopkeeper along with five other individuals including a woman, a linen seller (σινδο[νο]πώλην), and a carpenter (καναβιο[υ]ργόν), and, moreover, anonymous servants and several other women whose trades are not mentioned. Sosimenes, one of the targets, must have been a business owner since a shopkeeper and the linen seller are said to have belonged to him. It would be surprising to find a single artisan or a small retailer who could have been in direct competition with so many people in such a variety of trades. Since the curser mentioned that he/she was a neighbour of his victims, Eidinow has suggested that this and similar curses might have been prompted by animosity coming from living in close quarters, or perhaps by rivalries between clubs or societies.56 This is plausible, as is the hypothesis of a competition between landlords being transferred and fought out between their employees’ or their slaves’ shops. Tablets targeting nondescript workplaces (ἐργαστήρια) or work (ἐργασία) are more likely to have been written in response to competition.57 These curses targeted different kinds of business: one cursed a leather worker (DT 12), another healers (SGD 124). Since Athenian workplaces could be relatively large, this is 40 Olivier Dufault still no proof of curses having been written by relatively poor individuals. For example, the orator Lysias (12.19) owned a shield-making ἐργαστήριον operated by 120 slaves. Even if most commercial tablets had been targeting owners of small establishments, two important facts run against using this evidence to argue that curse tablets cut across all social categories. First, it was not the norm to be an artisan, a shopkeeper or a prostitute in classical and Hellenistic Greece. More importantly, only a handful of classical and Hellenistic tablets could be said to have been targeting artisans, shopkeepers or prostitutes. Even if we were to assume that most of these curses were written by direct rivals, they still represent only 4 per cent of the Greek tablets available in the two main corpora and in David Jordan’s first survey (DTA, DT and SGD). This represents 23 per cent of the classical and Hellenistic tablets which provide information about their targets.58

Curses related to lawsuits The classical or Hellenistic tablets potentially written in the context of lawsuits outnumber those of any other group.59 As many have argued, Athenian litigation was more of a way to resolve honour-based feuds between powerful groups than an attempt to impose the rule of law on all citizens.60 If we add curses naming Athenian aristocrats to the 55 judicial tablets, this would mean that one out of every four of these curses was probably aimed at an Athenian politician.61 This would also mean that almost half of the classical and Hellenistic tablets providing information about their targets (43 per cent; 67 out of 156) were aimed at politicians or written in a judicial context.62 Keeping with the assumption that curse tablets were aimed at competitors – or, at least, at persons whom the curser knew – we could then conclude that about half of these tablets were probably written in an elite context. On the other hand, scholars have responded to the claim that Athenian courts mainly served the interests of the rich by arguing that legal literature is not representative of the whole spectrum of Athenian legal experience.63 Acknowledging the existence of other legal procedures could also buttress the hypothesis that Athenians of all stripes had access to legal institutions. Eidinow lends support to this position by showing that Athenian judicial curse tablets imply the presence of many attendants at court proceedings. Her discussion of several tablets is particularly interesting as it shows that women were present in courts even if they were not represented in the speeches of Athenian orators.64 However, the question of the representative nature of the Athenian courts themselves remain unresolved. Since we are not in a better position to judge the extent to which curse tablets are representative, one of these two domains of evidence cannot be used to support a hypothesis concerning the other without external support. In sum, the epigraphic evidence supporting the claim that curse writing in classical and Hellenistic Athens cut across all social categories is not particularly good. Assuming that curse writers targeted their rivals, the little that can be inferred is that a minute portion of these tablets were written by workers or business owners. Workers among this group may have been relatively poor but they are not representative of the population of Attica. Those who worked the land – the majority of the population – were never alluded to in classical and Who wrote Greek curse tablets? 41 Hellenistic were tablets. It is consequently not possible to conclude that the use of curse tablets effectively cut across all social classes or that they were mostly used by the non-elite. Rather, since we know that curse tablets were written documents, we can infer that their authors were literate and consequently more likely to be part of the elite. To leave the argumentation there, however, would be to ignore that professional curse writers or literates could have provided their skills to the illiterate. The following and final part of the argument assesses what curse tablets can tell about this trade.

Professional curse writing and epigraphic evidence As was argued in the second section, the only literary source that could support the theory of an ancient Athenian curse tablet market is not conclusive: Plato did not specifically mention curse tablets and his mention of the sale of curses is suspect. It remains to be seen whether the epigraphic record can nonetheless support the claim that religious professionals commonly provided literates and non-literates with curse tablets. John Gager has offered the most extensive argument for the existence of an ancient curse tablet market. Noting the presence of elegant handwriting among curse writers, ‘highly formulaic texts’ and large caches of tablets found in a single place, he concluded that ‘on balance the scales appear to favor professionals [by which he means either magi or scribes], at least in the Roman period, both for inscribing the tablets and for providing the formulas’.65 Three assumptions are at work in the argument. First, elegance in handwriting does not provide reliable evidence about individuals who wrote curse tablets on a professional basis. As Gager has noted, scribes could have written one or several curse tablets. Scribes, however, or any experienced writer, could have also done so for him- or herself. The second assumption associates the use of complex formulas with professionalism. But any literate (or barely literate) person could have copied a recipe for his or her own profit.66 Moreover, even if late antique curse tablets sometimes show textual or graphic devices similar to those found in the documents from the PGM collection, it appears that no published curse tablet has been (faithfully) copied from one of the extant recipes.67 It is also possible that some curse writers created formulas on the spot. Both assumptions ignore the possibility that experienced writers could write curse tablets for themselves. Moreover, Roger Tomlin’s edition of the tablets found at the baths of Aquae Sulis (modern Bath) cannot be used to support the assumption that curse tablets were written by professional curse writers. As Gager has noted, Tomlin mentioned that most of the curse writers from Aquae Sulis were experienced, that a few had ‘calligraphic’ hands and that some ‘were so clumsy as to suggest semi-literacy’.68 While Tomlin noted the ‘clerical’ appearance of the handwriting found on the majority of the tablets, he did so in part to suggest that scribes or clerks from the procurator’s office nearby might have written the tablets, not that tablets had been written by professional curse writers.69 In fact, Tomlin concluded that all the tablets must have been written individually.70 Considering that a professional curse writer would have certainly left more than one tablet, it is unlikely that even a single one worked in late Roman Aquae Sulis. 42 Olivier Dufault The best evidence for professional curse writing is found when multiple tablets written by the same hand have been explicitly commissioned by different individuals. Only one extant set presents this combination of characteristics (see below). Anybody who wrote a tablet once can be expected to have tried again, at least a few times. I would expect professional curse writers to have left at least a dozen curse tablets. Wünsch, the editor of the DT, did not notice the presence of large finds of tablets written by single individuals despite looking for similar handwriting. He pointed out when a single person had written multiple tablets but he did not notice more than two tablets in each instance.71 Similarly, Curbera’s preliminary study of the DT corpus uncovered only four sets of two to three tablets each.72 The following section lists the three sets of tablets that were most probably written on behalf of others according to one or more of three criteria: they include a relatively high number of tablets written by the same person (twelve or more), proof that they were commissioned,73 and/or show very similar or identical formulas. All three sets were written in Greek and and are dated from the second century to the fourth century ce.

Tablets from two wells in the Athenian agora (c. 250 ce)

According to Jordan, a single person from the mid-third century ce inscribed fifteen curse tablets and deposited them in two wells in the Athenian agora. The tablets were found below a destruction layer associated with the Heruli invasion of 267 ce.74 Most of the tablets follow the same formulas and the writer probably worked from memory since the wording often differs slightly. One tablet calls upon the spirit of a dead person in what appears to be an attraction curse, or ἀγωγή.75 Four tablets attempt to separate women from male acquaintances and one hands over a certain Tyche, daughter of Sophia, to Typhon. The other tablets bind three wres- tlers, a charioteer and one man called Eros son of Isigeneia, who is not otherwise identified. Considering the relatively small number of tablets written, it is not entirely clear at first sight whether their author wrote them for him- or herself or for others. The tablet targeting Tyche, however, was probably pre-written to be sold afterwards. Jordan pointed out that the name of the target was slightly compressed, as if the writer, intending to sell the curse in the future, had left a space that was too small for the names of Tyche and her mother. Fifteen tablets is already more than the average number found in most same-hand sets, but closer to the small sets of same-hand tablets than to the two others listed below (respectively thirty-seven and one or more). Nevertheless, the fact that one of the tablets was pre-written indicates the work of a professional curse writer.76

Tablets from a columbarium near Rome (c. 400 ce) This set was found among the urns of a columbarium on the Via Appia and was published over a century ago by Wünsch.77 Most of the tablets were written in Who wrote Greek curse tablets? 43 Greek and targeted charioteers. Comparing the names of the charioteers with the names of Roman charioteers found on contorniati – medallions on which the names of emperors and charioteers both appeared – Wünsch dated the tablets to c. 390 ce.78 Of about fifty Greek tablets, thirty-seven were written by the same person.79 Considering the number of tablets and the fact that the formulas are very similar, it is likely that the curse writer operated professionally. What kind of professional is difficult to tell. Literary evidence relating to curses and chariot rac- ing does not suggest that charioteers were expected to know how to write curse tablets.80 Any literate person with a recipe book or a little imagination may have inscribed the tablets.

Tablets from Amathous, Cyprus (c. 200 ce) A large deposit of curse tablets from a shaft near Amathous on Cyprus provides the best evidence of professional curse writing (DT 18–37).81 Of the twenty tablets published so far, sixteen show the same handwriting. There are, however, 260 tablets in the lot and more tablets with the same handwriting could probably be found.82 According to preliminary observations and a partial publication, the tablets appear to have been written by several persons with handwriting dated to the late second century or the early third century ce. All of the published tablets show the same prayers, invocations and divine names, features which appear almost in the same order each time. It is also certain that these tablets were written on somebody else’s behalf since they show the same handwriting and list the names of different petitioners. According to Jordan, the unedited tablets bear the same formulas.83

Conclusion These three sets of curse tablets are the only epigraphic evidence attesting to the practice of professional curse writing. This fact is well worth repeating since the implicit double assumption that I have discussed here – that curse tablets were usually written by religious professionals on behalf of the non-elite – is often generalized to all curse tablets. The only literary evidence that has been proffered in support of the first part of the double assumption is problematic. Similarly, epigraphic evidence suggests that professional curse writers were exceptional or inexistent during the Classical and Hellenistic periods. Clear evidence only sur- faces in sources from the second century ce and later. Of course, my analysis is also based on assumptions. The first is that socio-historical studies should value epigraphic corpora over literary traditions. From this, I have also assumed that the absence of epigraphic evidence should be valued over unsubstantiated expectations (e.g. that curse writing or that ‘magic’ in general is typical of the non-elite); that an act which we recognize as religious must have been performed by a religious professional. I expect that my second assumption will be less widely shared than the first since it runs against the testimony of Plato. 44 Olivier Dufault However, Plato’s comments appear less reliable once we assume that his socio- logical imagination was limited by his social position. Since Plato’s own philosophy and theodicy were directly contradicted by the do ut des sacramental theology implied by his opponents’ practices, he must have been particularly interested in believing that the religious professionals in question also sold curses to the rich. More importantly, the verb κατάδεω and its substantivized form (which Plato used in the same sense as κατάδεσις in the Laws passage) referred throughout antiquity to knots or to the act of binding, not to curse tablets. ‘Curses’ (ἀραί etc.) and the act of ‘binding’ (κατάδεω) were certainly common in Greek literature but no extant author described a curse tablet before Tacitus and Apuleius in the second century ce.84 The assumption, based on Plato’s Republic and Laws, that it was common to seek the services of professional curse writers throughout the Mediterranean world from the fifth century bce to the fifth century ce cannot be supported by the evidence provided by the published curse tablets. Similarly, to assume that curse tablets were used by people from all social categories cannot be substantiated by literary or epigraphic evidence. In other words, the evidence suggests that, before the second century ce, curse tablets were usually written by individuals who were literate to various degrees and who cursed for their own profit. Both literary and epigraphic sources suggest that the professionalization of Greek curse writing was a late phenomenon. It is difficult to say whether these professionals made their first appearance during the second century ce or if they merely increased in number from that point on. In any case, it is more likely that the use of curse tablets started to reach illiterate or partially literate individuals during the first centuries ce. The results of this study are not simply negative. They also raise new questions: Why did professional curse writing appear or become popular around the second century ce? Could larger socioeconomic changes be partly responsible for this change? Could the diffusion of new textual devices found on late antique curse tablets and other media be related to the appearance or acceleration of this trade?85 In other words, could the professionalization of curse tablet writing be the symptom of a new trade that involved a new type of professional (religious or otherwise) dealing in techniques such as those found in the Greco-Egyptian recipe books? Finally, I should also offer some explanation to the reader who thinks that the negative arguments of sections one to three were unnecessary. Unlike many other ancient corpora, the corpus of ancient Greek and Latin curse tablets is in constant evolution. Since new evidence will likely come to light in the future, the study of curse tablets can afford bold hypotheses as well as falsification attempts. First, new tablets are regularly found and the rate of discovery may increase as late antique archaeology develops. Second, around 750 post-Classical tablets have never been edited or indexed. Of the three sets studied here in section four, over 200 tablets from Amathous and around forty tablets from the wells of the Athenian agora are still unpublished. The case of Latin tablets from the UK is similar. According to Daniel Ogden, Tomlin’s publication of the tablets from Bath Who wrote Greek curse tablets? 45 (c. 100 legible tablets) represents only a sixth of the entire deposit.86 This means that the number of Greek and Latin tablets still unedited represents a third of those now listed or edited by modern scholars (about 750 of a total of about 2,100).87 New editions could radically change the makeup of the corpus. In the event that more late antique tablets are published, the Thesaurus Defixionum Magdeburgensis will certainly become useful in tracking these changes.88 I would certainly not claim to provide a definitive answer (or refutation) as to who wrote Greek curse tablets. The nature of the field of study, however, gives hope that some hypotheses will one day gain in certainty.

Notes 1 Research for this paper was made possible thanks to the Fonds de Recherche Québécois, Société et Culture and the Distant Worlds Graduate School at LMU, Munich. I would like to thank friends, colleagues and reviewers for commenting on previous drafts. 2 Gager (1992) 24. By ‘curse tablets’ (also called defixiones and katadesmoi), I mean short texts written in various languages (usually ancient Greek or Latin) found on various supports (usually lead), and in which writing played a role in wishing others to be paralysed or somehow harmed. They were in use from early classical times to late antiquity and deposited in tombs, sanctuaries or underground bodies of water (e.g. wells) throughout the Roman and Greek spheres of influence. For a concise introduction to the topic, see Ogden (1999) 3–90. For scholars sharing Gager’s point of view, see Graf (1997) 84–6; Ogden (1999) 67; Graf and Johnston (2017); Parker (2007) 129; Riess (2012) 176. For a critical, sophisticated and exhaustive study of curse tablets that repeatedly but indirectly approaches this question, see Eidinow (2007). 3 See Dickie (2001) 1–2; Lambert (2004) 76; Toner (2009) 39–40, 81–3; Watson (1991) 198. According to Bernand (1991: 20, 30–4), curse users mainly came from the margins of ancient societies. By ‘the elite’ in Attica at the end of the fourth century bce – the time and place where most of the evidence considered here comes from – I mean about 9,000 individuals possessing 2,000 drachmas or more, ‘those who owned at least enough land to make an independent living as a working farmer’ (Van Wees 2011: 111). According to Van Wees, these c. 9,000 males formed around 30 per cent of the citizen body (c. 27,000–30,000) and probably controlled around 80 per cent of the land. 4 See Graf and Johnston (2017); Jordan (1988) 276–7. 5 The only explicit theory of this kind was first proposed by Burkert, who argued that professional curse writing evolved from the practice of eastern religious professionals (1997: 69–70). This theory was adopted by Johnston (1999: 119). For a different interpretation, see Carastro, who made a case that the language of Greek curse writing is better explained by Greek representations of binding (2003: 1–14; 2006: 163–88). On curse tablets as the work of religious specialists, see also Dickie (2001) 47; Graf (2005b) 257, 269–70 and (1997) 134; Rosenberger (2012) 72. Versnel (2006) notes that the existence of professional curse writers can be guessed from ‘series of identical defixiones … found together and written by the same hand’. But Versnel did not point out that most of these series of tablets are rather small (from two to five tablets) and that all large series (c. twelve tablets or more) date to the second century ce or later and were written in Greek (see below). Gager (1992: 5) provides the only discussion of professional curse writing, and looks for professional curse writers beyond the realm of the religious professional (see discussion below). For similar approaches see Ogden (1999) 54–60; Jordan (1985) 205–55; Collins (2008a) 71. 6 See Bernand (1991) 20; Dickie (2001) 47; Riess (2012) 174; Stroud (2013) 86. 7 See Curbera (2015) 109–13. On the rediscovery of Wünsch’s collection and Curbera’s new edition, see Curbera (2012) 193–4. 46 Olivier Dufault 8 This is also the case for curse tablets for all places and all periods. See, however, Tab. Sulis 31 from Bath and written in Latin c. 175–275 ce, which concerns a lost ploughshare. 9 See Faraone (1991b) 5. 10 To my knowledge, the only authors to have expressed the view that curse writing was further professionalized in later times are Collins and Gager, who argue for an increase in professionalization starting in the first century ce, and Faraone, who sets it in the first century bce. See Collins (2008a) 71; Gager (1992) 4; Faraone (1999) 16. 11 For summaries of the argument see Harris (1989), esp. 11–12, 327. 12 For a critique of Harris’ book, see chiefly Thomas (2009b) 357. See also Thompson (1994) 65; Hopkins (1991) 133–58. On the study of literacy types rather than of literacy considered as a single spectrum, see Johnson (2009) 3–10, 13–43. 13 See Hansen (2006) 93. 14 Thus 9,000 citizens out of a total population of 120,000. See note 3 above. 15 Thomas (2009a) 23–4. As noted above, the fact that the earliest Greek curse tablets mostly come from Attica could further support the theory that Athenian citizens were exceptionally literate. For a recent argument on the effect of democratic institutions on literacy levels in classical Athens, see Missiou (2011) together with the reviews by Jim (2011) and Stauner (2012: 152–5). 16 See Thomas (2009a) 18–19. 17 Lang (1982) 75, 87 and (1976) 24. On the adoption of the Ionian script and the disap- pearance of Attic letters in all texts shortly thereafter, see also Threatte (1980) 26–51. 18 On what Ogden calls the ‘twistedness’ of curse tablets, see (1999) 29–30. 19 See DTA no. 60; DT nos. 24, 86, 95. 20 See Curbera (2015) 105–7. This notion is expressed on at least three Greek tablets from the Hellenistic period: NGCT no. 31 (Pela, fourth century bce); DTA nos. 42–3 (Megara, fourth century bce) and DTA no. 52 (Attica third or second century bce). 21 See Curbera (2015) 105–6. 22 See Dakare, Vokotopolou and Christide (2013). 23 See Kroll (1977) 83–140, esp. nos. 1–26. 24 Parke (1967) 101. 25 Eidinow (2007) esp. ch. 5. 26 Eidinow (2007) 130. 27 Dickie (2001) 77–92. 28 Cf. Carastro (2003) 11–13. 29 καταδ(έ)ω alone: DT nos. 40, 41, 43–7, 52, 54, 56–8, 60–2 (‘of Roman times’), 64, 66, 68–71, 77, 79, 81, 84–6, 88, 90, 91, 98, 100, 101, 104, 105–7, 119, 124, 125, 141, 159; in combination: DT nos. 55, 89, 92–3, 95, 109. 30 See e.g. PGM IV.2175–8; Ogden (1999) 5. 31 Gager (1992) 5–7. A comparison of curse tablet recipes from the PGM with classical and Hellenistic curse tablets also makes this point clear. Cf. PGM V.304–69; VII.396– 404, 417–22, 429–58; IX, X.24–35, 36–50; XXXVI.231–55; LVIII.1–14. 32 PGM III.1–164; VII.396–404, 417–22, 429–58; XXXVI.1–34. A professional curse writer operating in Rome in the late fourth century also used κάτοχος to refer to a curse tablet (DTA 187, line 55). 33 Plato refers to this practice at Laws 933a. Curbera counts seventy-five nailed tablets out of the 200 complete tablets in the DT corpus (2015: 105). Examples of figurines pierced with nails are listed by Faraone (1991a) 200–5. See item no. 7 (found in a grave, Attica, third century bce), no. 12 (Sanctuary of Zeus Hypsistos, Delos, first century bce), no. 21 (unknown provenance, with Latin inscription, first century ce) and no. 27 (Middle Egypt, fourth century ce). 34 See Watson (1991); Faraone (1991a). 35. Neglecting passages which mention devotiones or defigere only – two words referring to forms of cursing that do not necessarily imply the use of curse tablets (see Livy, 8.9; Ovid, Heroides 6.91–4) – we can list the following passages: Ovid, Amores 3.7.27–30; Who wrote Greek curse tablets? 47 Tacitus, Annals 2.30, 69; Apuleius, Metamorphoses 3.17. Apuleius provides the sole evidence implying professional cursing activity. 36 Plato, Republic 364b–c: ἀγύρται δὲ καὶ μάντεις ἐπὶ πλουσίων θύρας ἰόντες πείθουσιν ὡς ἔστι παρὰ σφίσι δύναμις ἐκ θεῶν ποριζομένη θυσίαις τε καὶ ἐπῳδαῖς, εἴτε τι ἀδίκημά του γέγονεν αὐτοῦ ἢ προγόνων, ἀκεῖσθαι μεθ᾽ ἡδονῶν τε καὶ ἑορτῶν, ἐάν τέ τινα ἐχθρὸν πημῆναι ἐθέλῃ, μετὰ σμικρῶν δαπανῶν ὁμοίως δίκαιον ἀδίκῳ βλάψει ἐπαγωγαῖς τισιν καὶ καταδέσμοις, τοὺς θεούς, ὥς φασιν, πείθοντές σφισιν ὑπηρετεῖν. 37 Parker (2005) 122. 38 Ogden (1999) 54. 39 See Dickie (2001) 48–9; Graf (2005b) 269; Johnston (1999) 119. Gager offers the qualification – ‘on balance the scales would appear to favour professionals, at least in the Roman period, both for inscribing the tablets and for providing the formulas’ (1992: 5). For a similar assumption, see Stroud (2013) 86. Curbera implies professional writers by use of the term ‘magician’ throughout his paper but contradicts it in the conclusion (2015: 108–9, 116). 40 See Plutarch, Dialogue on Love (Amatorius) 771a; Josephus, JA 3.165.1. 41 Plato, Republic, 358b–368a. Cf. Phaedo, 107c2–d5, 114c2–8. Plato also limited the discussion of soteriology to myths, negative formulations and off-hand remarks. See Menn (2013) 191–216. 42 Gagné (2009) 211–47. 43 For a similar reading, see Stratton (2007) 42; Collins (2001) 482, 484. 44 For a similar observation cf. Graf (1997) 23. 45 Well-written tablets: DT no. 55, ‘very elegant writing imitating the beauty of those found on public monuments’; no. 87, ‘most beautiful writing’; no. 109 is an ‘example of those [tablets] that are written with care’. One could also count DT no. 68: ‘written with very small letters’ and DT no. 107, which shows some level of literacy and perhaps even an attempt to imitate classical examples. Badly written tablets: DT no. 66: is a ‘judiciary inscription written by an uneducated man and consequently full of mistakes’; no. 75 is ‘very badly written, perhaps by a foreigner’; of no. 94, Wünsch writes that ‘If the tablet is genuine – which I do not doubt at all – it is a very curious example of an attempt [at cursing] done by an unlearned man’. 46 DT 2. 47 Curbera (2015) 113. 48 Curbera (2015) 109. 49 Eidinow and Ogden point to DTA no. 85, Boeotia, third or second century bce, no later than the Hellenistic period, or second/third century ce; SGD no. 48, Athens, c. 323 bce and no. 173, Olbia, between third and first century bce. See Eidinow (2007) 143; Ogden (1999) 59. 50 These are DT nos. 33 and 34, both from Attica and the third or fourth century bce; DT no. 45, Athens, third or second century bce; SGD no. 91, Gela, c. 450 bce. See Eidenow (2007) 156. 51 These do not follow the formulas generally found on curse tablets and were perhaps handled more openly. See Versnel (1991) 60–106 and (2010) 275–354. They are consequently less easily comparable with other curse tablets. 52 Bathing clothes: Tab. Sulis no. 63; cloaks: Tab. Sulis no. 62; DTA nos. 6, 212. We might compare these with the lost blankets requested on tablets from Dodona. See Parke (1967) 272. 53 The expenses for food for a Roman soldier’s salary in 81 ce (in Egypt) was 240 drachmas (roughly 240 denarii) a year, which means that the price of food for one day was evaluated at 0.7 drachma. See the P. Gen. Lat. 1 in Erdkamp (2007) 309. 54 Following Audollent: DTA nos. 2–4, 6, 8, 11, 12, Cnidus, 300–100 bce, various items; DTA nos. 74–75, Achaia, no dating and no indication of stolen object; DTA nos. 212, Bruttium, third century bce, three gold coins. 55 These are: DT no. 12, leather worker; DT nos. 30, 70, 72–3, 75, SGD no. 43, shopkeepers and shops; DT no. 55, pipe maker and carpenter; DT no. 68, multiple targets, includ- 48 Olivier Dufault ing shopkeepers, a miller, a boxer, pimps and prostitutes; DT no. 69, a helmet-maker and a gilder; DT no. 71, 74, 84, workshops; DTA no. 87, multiple shopkeepers, women, a linen-seller and a frame-maker; DTA no. 52, female slaves; DT nos. 86, 97 DTA nos. 41, 47, 72–3, 92, 109, SGD nos. 43, 73, 75, work; DTA no. 74, stonemason; SGD nos. 3–4, blower from silverworks; SGD no. 11, stallholder, household slave, innkeeper and pimp; SGD no. 20, blacksmith; SGD no. 44, potters; SGD no. 48, scribe; SGD no. 52, net makers; SGD no. 72, seamstress; SGD no. 124, workplaces and men belonging to healers; SGD no. 170, helmsman. For a survey of this category, see Eidinow (2007) 191–205. 56 Eidinow (2007) 198–9. 57 DT nos. 71, 74, 75 (84 and 86 also curse κάπηλοι); DTA nos. 41, 52, 72, 73; SGD nos. 75 and 124. Following Eidinow (2007) 425. I have excluded SGD no. 88. 58 See Faraone (1991b) 10. 59 I follow the count in Eidinow, who excludes tablets that do not contain forensic language: DT nos. 25, 38, 39, 63, 65–7, 81, 88, 94, 95, 103, 105, 106, 129, 158; DTA nos. 39, 43, 44, 49, 60, 62, 63, 67, 77, 87–9, 90; SGD nos. 6, 9, 19, 42, 49, 51, 61, 68, 71, 89, 95, 99, 100, 107, 108, 133, 176, 179; NGCT nos. 1, 9, 10, 12, 14, 15, 24, 38. See Eidinow (2007) 168–9. Gager noted three judicial curses involving the elite: DT nos. 38, 103, SGD no. 14. See Gager (1992: 116–50). The names of Athenian aristocrats have also been found together on DT nos. 24, 26, 47–50 and SGD nos. 14, 48, 107. 60 See Cohen (1995) ch. 4, esp. 113–15; Osborne (1985) 40–58, esp. 52. 61 Considering that, of the sixty-seven judicial curses listed by Faraone (1991b), fourteen involve aristocrats (DT nos. 24, 26, 38, 47–50, 95, 103, DTA no. 60 and SGD nos. 14, 48, 107, 162). On the identification of Greek aristocrats, see the prosopographical studies cited by Faraone (1991b 16, n. 76). On the links between curse tablets and Athenian aristocrats, see Parker (2005) 129–31. 62 See the discussion in Eidinow (2007) 172–3. 63 See Bers (2009) 7–24; Carey (1994) 172–86; Lanni (2006); Rubinstein (2000). 64 Eidinow (2007) 172–3. 65 Gager (1992) 5. 66 See the comments of Tomlin on Tab. Sulis 8. 67 On SGD nos. 152, 153, 155, 156, 159 and NGCT no. 93 and their similarity to PGM IV nos. 336–406, see Kambitsis (1976) 213–23; Martinez (1991). Other curse tablets most probably written from a recipe: Tab. Sulis 8, which includes the mention carta picta perṣc̣[ripta] – ‘the written page has been copied out’; SGD no. 167, which includes a title such as those found in the PGM; DTA nos. 15–16, which show the same drawing; DTA no. 188, similar to PGM LVIII.1–14; NGCT no. 115, similar to PGM IV.1443–1457. 68 Tab. Sulis 84, 100. 69 Tab. Sulis 88, 98. 70 Tab. Sulis 99. 71 See DT 45 and DT nos. 35, 36, which both date from the Roman period. 72 Curbera (2015) 111–13. 73 I have excluded SGD no. 91, which dates to c. 450 bce, from Gela, even though it was written for somebody else since it specifies that it was written on account of friendship. 74 These are SGD nos. 22–35 and 38. For publication and dating, see Jordan (1975) 245–8 and (1985b). 75 See PGM XXXVI.69–160. 76 See discussion in Jordan (1985b) 77 See Wünsch (1898) = DT nos. 140–87. 78 See Wünsch (1898) 56–63. 79 Wünsch (1898) tablet nos. 6–8, 10–11, 16–33, 35–48 (= DT nos. 145–47, 149–50, 155–71, 173–86). Tablet nos. 9, 12–15, 34, 49 show different letterforms. For handwriting identifications see Wünsch (1898) 53–6. Who wrote Greek curse tablets? 49 80 See the three occurrences of curse accusations involving charioteers in the Histories of Ammianus Marcellinus (26.3.3, 28.1.27, 28.4.25). See also Dickie (2001) 282–7. 81 For a study of these tablets, their ancient setting and their discovery see Wilburn (2012) 169–218. 82 Aupert and Jordan (1981) 184. Jordan subsequently mentioned that he saw the work of several hands in those that he studied. Cf. Gager (1992) 133 n. 46. Mitford, who published the third edition in the 1970s, remarked that the writing of two tablets – IKourion nos. 130 and 132 – differed markedly from those of the fifteen remaining defixiones, while two others – IKourion nos. 137 and 127 – had very similar handwriting. 83 See Gager (1992) 133 n. 46. 84 Tacitus, Annals 2.69. The mention of mysterious writing in 2.30 also strongly suggests curse tablets. See also Apuleius, Metamorphoses 3.17. 85 E.g. voces magicae, palindromes, word-pictures (i.e. carmina figurata), references to exotic or unknown divinities. See Gager (1992) 6–9. Some of these features are also found on engraved gems; others became part of philosophical discussions in late antiquity. 86 Ogden (1999) 4. 87 Ogden (1999) 4–5. 88 See http://www-e.uni-magdeburg.de/defigo/wordpress/ 4 A story of blood, guts and guesswork Synthetic reasoning in classical Greek divination1

Ralph Anderson

‘Bird signs? You want to plan a strategy based on bird signs?’ scoffs Eric Bana’s Hector in Wolfgang Petersen’s Troy, when he is confronted with an attempt at divination. When the Homeric Hector encounters exactly the same omen, he too rejects it and condemns the hapless Polydamas, who reported it, as a coward. Yet, although these two Hectors reject the advice of divinatory signs, they do so on very different grounds. Where the Hollywood Hector denies the relevance of irrational factors such as divination, and indeed religion more generally, to the pragmatic realm of warfare, the Homeric Hector dismisses only the specific omen that Polydamas reports and remains quite convinced that Zeus has promised him victory. The gap between these two Hectors is emblematic of the modern difficulty not only of accepting quite how much divination there was in antiquity and how seriously it seems to have been taken, but also of comprehending how such seemingly irrational practices could have been employed in even the most serious of undertakings.2 To be ‘modern’, in this context, is not simply to be contemporary, but to espouse particular assumptions about rationality, causation, knowledge and the secular nature of world, all of which militate against the ready comprehension of divination. The purpose of this chapter is to explore the cognitive process by which divinatory signs are brought to bear on practical situations. The fundamental question it seeks to answer is where the meaning generated by divination comes from. This question will be explored through consideration of a number of well-known oracle stories from Herodotus, and some examples of various kinds of divination from other authors. The chapter’s central claim is that the reasoning which underpins divinatory thinking, and hence the meaning of oracles and omens, is fundamentally relational. In other words, the meaning that is found in divination is akin to the meaning that is found in a story, in which each component of the narrative influences the meaning of the other components, so that the overall meaning does not emerge until the story is complete. To plan on the basis of divination, therefore, is not to bring to bear a piece of knowledge independently derived from (for example) a bird sign upon a separate and, in principle, unconnected situation, but to bring the divinatory sign into relation with a practical setting so that each illuminates the other. It is this act of relating which generates the knowledge and insights supposedly gained by divination. A story of blood, guts and guesswork 51 In western scholarship, divination has frequently been evaluated against scientific models of truth and prediction. Such models place heavy emphasis on replicability and objective verification as the essential hallmarks of truth. However, as anthropologist Rosalind Shaw argues, this ‘positivist preoccupation with knowledge as verifiable observation’ tends to ‘obscure … the existence of alternative conceptions of truth’ and readily casts divination as ‘failed scientific explanation’.3 The alternative conceptions of truth to which Shaw alludes include notions such as the truth of the binding word – for example, the oath of a witness or the oath sworn to seal a vow – and the model of truth which Shaw herself espouses in her own analysis of divination, the truth of ‘performative efficacy’.4 In this view, the truth of divination ‘is in the action generated, the social reality reconstituted, and the resultant well-being of the people; it is not to be found in an abstract system or specific verifications of separate oracular pronouncements’.5 That is, divination is valuable not because it makes objectively verifiable predictions of future events or diagnoses of the past causes of present problems, but because it enables a society to resolve its current crises by generating an authoritative understanding of the current situation, its aetiology and opportunities for future action.6 But how does divination do this? Analyses such as Shaw’s, which evaluate divination by its power to effect change, tend to emphasize the sensitivity of divinatory practices to the nuances of the societies in which they take place. Central to the success of divination is the active involvement of both diviner and client. David Parkin, for example, commenting on Kenyan divination, regards the diviner and client as making up a ‘co-operative enclave’ which jointly arrives at a divinatory outcome. In similar fashion, commenting on Nyole divination in eastern Uganda, Susan Whyte offers a detailed discussion of the back-and-forth interpretative play between the two sides, in which the diviner’s skill is complemented by the client’s detailed knowledge of the specific situation. Significantly, this creative dialogical play is often downplayed or effaced completely in indigenous accounts of divination, in which the divinatory outcome is what matters.7 Metaphorical, ambiguous and polysemic language is also widely found. While a scientific or positivist approach might conclude that such language is there to protect divination from disproof and to allow the wily soothsayer to evade criticism for poor results or to claim credit for what was really just good luck, Shaw and Parkin argue that ambiguity and polysemic language are in fact central to the ability of divination to generate meaningful answers to questions. Parkin in particular highlights a transition that takes place in the course of the divinatory session, from ambiguous language at the beginning to unambiguous and specific language as a decision is reached. He characterizes this as a transition from an uncertain and chaotic ‘simultaneity’, in which multiple possibilities are entertained in parallel, to a ‘sequential’ order- ing of possibilities which eventually yields a stable answer. The initial problem of divination is a ‘superabundance’ of potential meaning which must be skilfully reduced to a single, stable solution.8 Cryptic, allusive and often fractured language is vital to the early stage of this process, as it allows the diviner to convey a rich multiplicity of allusively formulated, partially grasped and cognitively simultaneous alternatives out of which sense must be made to emerge.9 52 Ralph Anderson The prominence of cryptic and polysemic language in these anthropological examples naturally brings to mind the famously cryptic and polysemic Delphic oracles in Herodotus’ Histories. The series of high-profile prophecies in Book 1 is characterized by this kind of language, from the famous prediction that Croesus would destroy a great empire if he fought the Persians, to the subsidiary response that he should not be ashamed to flee when a mule sat on the Persian throne, and the equally cryptic oracles given to the Spartans concerning Tegea.10 We might even see in the Pythia’s response to Croesus’ ‘test question’ an allusion to the problem of the superabundance of potential meaning. Before Croesus entrusts his main question to an oracle, he first tests a number of oracles in order to establish which, if any, of them can be relied upon, by sending them a question to which he already knows the answer (Hdt. 1.46). Even when the question is simple – ‘What is Croesus doing now?’ – and the answer already known to the inquirer, the Delphic oracle insists on reminding Croesus how much more complexity there is to the world than he or any mortal can fully grasp. The oracle prefaces its answer to Croesus’ question with the statement that it can number the grains of sand, measure the sea, understand the dumb and hear the voiceless (Hdt. 1.47). Given this complexity, the wise approach would be to scrutinize the oracle’s response very carefully, and not to press on impetuously, as Croesus does, missing the ambiguity of the oracle and simply assuming that it supports his plans. Croesus’ failure to explore the meaning of the oracle properly and his subsequent downfall have been explained in many ways. For example, Matthew Christ argues that Herodotus presents Croesus’ testing of the oracles as a shrewd example of rational research, similar to Herodotus’ own, and attributes Croesus’ fall to his arrogance, which leads him to abandon his rational evaluation at the critical moment.11 Elton Barker emphasizes the political context of Croesus’ fall, linking his failure to perceive the ambiguity of the oracle to his status as the arrogant head of a tyrannical regime in which only one voice, his own, is ever heard, and in which the kind of open and meaningful debate which the polysemic oracle requires is impossible.12 Julia Kindt sets the story in a religious and epistemological context and attributes Croesus’ downfall to his failure to perceive the gulf between mortal and divine perspectives and his arrogant belief that he can approach Apollo as an equal, despite the oracle’s warning that divine perception far exceeds mortal.13 It seems that in Herodotus, then, the ambiguity of the oracle can be an opportunity for rational exploration, a test of the moral or political character of the inquirer, and an emblem of the limits of human knowledge. However, Herodotus’ tales of oracular ambiguity can also shed light on actual divinatory thinking in antiquity. Herodotus’ Delphic oracles cannot be taken at face value as transparent historical sources however. The editors of both major collections of Delphic responses regard them as spurious in one way or another. Parke and Wormell accept contact between Lydia and Delphi before and during Croesus’ reign, but reject both Croesus’ testing of the oracle and the responses given to him afterwards. They do, however, accept as genuine the two oracles given to the Spartans concerning Tegea.14 Fontenrose is more sceptical and judges inauthentic the oracles given to both Croesus and the Spartans.15 It is not only the folkloric elements of these A story of blood, guts and guesswork 53 oracles but also their elaborate hexameter verse forms and careful ambiguity that arouse suspicion. These elements contrast markedly with those responses that are more readily accepted as historical, in which a straightforward answer was given to a question posed in a binary format, along the lines of ‘would it be more profitable and better to do x?’16 By these standards, then, Herodotus’ Delphic oracles represent, at best, very distorted versions of actual consultations and, at worst, products of Herodotus’ historical imagination. Nonetheless, it can be argued that there is more than one approach to establishing their authenticity. Lisa Maurizio notes that Parke, Wormell and Fontenrose judge Delphic oracles authentic if they can be proven to contain or closely reflect the actual words spoken by the Pythia.17 She criticizes this kind of approach, arguing that, in strictly logical terms, no oracle in the Delphic tradition can be demonstrated to record the actual words of the Pythia, because we have no way of accessing that original utterance and therefore no sure means of establishing this kind of authenticity.18 Moreover, she argues, this approach to authenticity is based on assumptions that properly belong to a literary tradition in which original authorship is both important and susceptible to proof or disproof. It is more accurate to regard our written reports of Delphic oracles as oral-derived accounts, that is, written snapshots of and contributions to a long tradition of oral transmission.19 In the case of oral tradition, transmission is really a misnomer. In a textual tradition, at least as now understood, the important thing is the verbatim transmission of the original text, with nothing added or removed; in an oral tradition, every performance is in effect a recomposition of the previous version. Often this is masked by an explicit assertion that the current performance is ‘just the same’ as the previous version and thus, by implication, the original. The authenticity of such recompositions in performance is determined by the audience, which has the power to reject the new performance or accept it as a valid part of the tradition.20 In such a setting, it is a mistake to judge authenticity by estimating the fidelity of the historical record to the hypothesized original words of the Pythia. Indeed, so central are the processes of reperformance, recomposition and authorization to the divinatory process, and so important is the interplay between performer and audience, that one may even speak of the authorship of oracles being shared between the Pythia and the recipients of her words, both the original enquirer and those who follow afterwards.21 When this approach is applied to oracles, it suggests that the Greek world was awash with orally transmitted oracles that were, in effect, recomposed in performance to suit the current circumstances. A clear example of this is the oracle ‘remembered’ by certain Athenians during the plague that struck Athens early in the Peloponnesian War. Certain elderly Athenians claimed to remember an old prophecy that ‘war with the Dorians comes, and a plague will come at the same time’.22 Others said that the prophecy foretold a famine, the operative Greek words being λοιμός (plague) and λιμός (famine). Thucydides’ comment on this is that people adapt their memories to suit their current sufferings, and that if, in future, a war with the Dorians were accompanied by a famine, the other version of the oracle would be recalled.23 Thucydides’ comment may be dismissive, but 54 Ralph Anderson if we view this way of using oracles through the lens of performative efficacy, then it is likely that this flexibility is actually central to the practical value that oracles hold. In this case, the oracle has an explanatory function, giving some kind of meaning to the terrible and unprecedented plague the Athenians were endur- ing. It is noteworthy that the attempt to challenge the oracle on grounds of literal authenticity – that is, fidelity to a supposed original source – fails: the fact that the city is gripped by plague gives relevance and utility to an oracle that relates to a plague and outweighs any quibbles about whether the oracle originally referred to a famine instead. Several important points emerge from Maurizio’s discussion. First, while her model of an oral tradition of Delphic (and other) oracles undermines any possibility of using Herodotus to recover the actual words spoken by the Pythia on any particular occasion, it offers a strong defence of the idea that Herodotus’ account must have been plausible to his audience in his own lifetime.24 If we accept that his Delphic oracles are written recompositions of stories already in circulation in the oral tradition, and that his recompositions were accepted by his audience as valid, and possibly even authoritative, then we may take a further step and infer that Herodotus offers us an insight into contemporary thought processes of using oracles. That is, Herodotus’ Delphic oracles were plausible to his audience because they conformed to contemporary ideas of oracular divination. Second, Maurizio’s idea of oracles being recomposed in performance, as in the ‘plague’ oracle in Thucydides, presupposes a theory of knowledge that differs markedly from that implied by a literary tradition. A literary tradition, in contrast with an oral tradition, is sustained not by recomposition but by reproduction – that is, by the verbatim transmission of an original text which preserves unchanged the exact words of the founding author. When the concept of transmission by recomposition is applied more broadly, it has far-reaching implications not only for our understanding of how knowledge is gained, used and passed on, but also, more radically, for the very nature of that knowledge. As the anthropologist Tim Ingold suggests, a model based on transmission ‘implies that knowledge already acquired is imported into contexts of practical engagement … and therefore that the knowledge is, in itself, context independent’.25 By contrast, Ingold maintains that knowledge is inherently relational and is generated in the context of practical action. Knowledge is not a fixed object to be passed along and applied mechanically to new contexts, but is generated and regenerated in the process of practical action, which always takes place in its own particular context. Where previous knowledge is brought to bear, it is via an active process that seeks relationships between remembered experience and the current situation, and in doing so, changes the meaning of both. The upshot of this, he argues, is that ‘we can understand the nature of things only by attending to their relations’.26 In other words, we know what things are only through how they relate to other things and to ourselves. As in a story, the meaning of the elements does not become clear until the story is complete, and the relationships between all elements are known.27 It follows from this that a change in relationships between things amounts to a change in their very nature and meaning. As Jerome Bruner puts it, ‘parts and wholes in a A story of blood, guts and guesswork 55 narrative rely on each other for their viability’. That is, ‘The act of constructing a narrative … is considerably more than “selecting” events either from real life, from memory, or from fantasy and then placing them in an appropriate order. The events themselves need to be constituted in the light of the overall narrative’.28 The same principles may be applied to oracles, and to the practical situations in which oracles are used, but in that case, crucially, part of the story is still in the future, still unknown and hypothetical. The rest of this chapter will explore this relational approach to divinatory knowledge through a series of case studies. The first case study concerns an oracle given to the Spartans in their lengthy campaign to subdue Tegea. Having already failed to conquer Tegea once, the Spartans have consulted Delphi and been told to retrieve the bones of Orestes. Unable to find them, they ask Delphi for further guidance, and are given a deeply cryptic response:

There is a place, Tegea, in the level plain of Arcadia, Where two winds are blowing under strong constraint; Striking and counter-striking are there, and woe is laid upon woe; There the life-giving earth holds Agamemnon’s son. Bring him home, and you will be Tegea’s protector. (Hdt. 1.67)

The oracle is eventually deciphered by Lichas, a former Spartan cavalryman, by a combination of ‘luck and wisdom’ (συντυχίη … και σοφίη, Hdt. 1.68). Visiting Tegea, Lichas enters a smithy, where he expresses wonder at seeing the iron being worked. If the Spartan marvelled at iron-working, the smith remarks, he would have been truly astounded at the gigantic coffin he had found while digging a well in his courtyard (Hdt. 1.68). At this point, realization dawns on Lichas. Herodotus talks us through his reaction:

He considered what had been said and put it together for himself [συνεβά λλετο] that this was Orestes, according to the oracle, putting it together συμβαλλόμενος] like this: he found [εὕρισκε] that the blacksmith’s two pairs] of bellows were the ‘winds’, the hammer and anvil were the ‘striking and counter-striking’, and the beaten-out iron was the ‘woe laid upon woe’, drawing this inference [εικάζων] because the discovery of iron had been an evil for mankind. He put these things together [συμβαλόμενος δε ταῦτα] and, returning to Sparta, he declared the whole matter to the Spartans. (Hdt. 1.68)

Herodotus’ account heavily emphasizes a process of ‘putting together’ that leads to ‘discovery’ (συμβάλλω and ευρίσκω) – in other words, the bringing into relation of the ambiguous words of the oracle on the one hand and, on the other, elements of Lichas’ immediate environment, the smithy. Lichas puts together the words of the oracle, the words of the smith and the equipment of the forge and, as a result, discovers the identity of the enormous body buried outside. Crucially, this 56 Ralph Anderson is not a one-way process in which the oracle is mechanically applied in a context- independent manner. Before Lichas enters the smithy and talks to the smith, he has no idea that the oracle means Orestes is in a smithy. The oracle is so ambiguous that it is effectively meaningless to him, or at least without stable, usable meaning. (Or rather, we might say that it has only a jussive meaning for him – he must search for its true meaning.29) Likewise, the smithy itself is an opaque environment: it is not obvious, even to the smith who dug him up, that it is Orestes who is buried in the back yard. The meaning of the elements in play – the oracle, the smithy and the body – becomes clear only when they are brought into the right relationship. This relational, synthesizing aspect of oracle use is further reflected in the participle εικάζων (‘drawing this inference’, Hdt. 1.68). Εικάζω may also be translated, rather less charitably, as ‘guess’ and indeed this sense is prominent in Euripides’ well-known remark that ‘the best seer is the one who guesses [εἰκάζει] well’.30 We might suspect Euripides, or his unnamed character, of tendentiously exploiting the range of meaning of this verb. The semantic range of εικάζω includes to liken, compare, infer from comparison, estimate and conjecture as well as to guess.31 Though the inclusion of this last sense, albeit something of an outlier, allows hostile speakers, such as Euripides’ character, to disparage divination as mere guesswork, the central cluster of meanings connected with representation, likening and comparison make this verb curiously appropriate to the activity of divination, a task which crucially involves seeking resemblances, putting things together, connecting features and bringing them into new relationships with each other. Indeed, Plutarch, who preserves this remark at de defectu oraculorum 432c, is quick to emphasize how much thought goes into divination, stressing that the best mantis (seer) is an intelligent (ἔμφρων) man who follows the promptings of that part of his soul that possesses sense (νους) and which operates in accordance with reasonable probability (μετʼ εἰκότος). This relational and synthetic principle is further illustrated by another famous Herodotean oracle, the ‘wooden wall’ oracle. This prophecy, which promises the Athenians that the ‘wooden wall’ alone will resist the Persians, is presented by the Pythia as a ‘word made firm as adamant’ (ἔπος … ἀδάμαντι πελάσσας, Hdt. 7.141). However, while the word itself is inflexible, its meaning is anything but.32 Once again, the problem is how to derive meaning from an ambiguous oracle, and at stake is the proper way to relate the oracle to the physical substance of Athens – its houses and temples, its Acropolis, its navy. In the ambiguity of the oracle lies both a deficit and a superabundance of meaning – it could mean anything and, therefore, in practical terms, it means nothing. As Herodotus presents the story, the Athenians have successfully whittled down the excess of meaning to just two options, a wooden fence round the Acropolis or their wooden warships (7.142). However, the final step in specifying the meaning of the oracle cannot be achieved until the relationship between its source, Apollo, and its recipients, the Athenians, is taken into account. Themistocles points out that the oracle would not have called Salamis ‘divine’ if the Athenians were destined to be destroyed there, and that the destruction the oracle foretold must therefore be intended for the Persians (Hdt. 7.143). The oracle’s meaning is so intensely relational that A story of blood, guts and guesswork 57 it depends not only upon the relationship between the Pythia’s words and the environment of Athens but also upon the relationship between the ultimate author of the oracle, Apollo, and its recipients.33 A caveat must be added to the idea that Herodotus offers us an insight into contemporary thought processes of using oracles. If the Delphic oracle did not in reality speak in ambiguous riddles, but instead gave straight answers to binary, yes/no questions, then how can it be said that Herodotus offers us any insight at all into Greek divinatory thinking? At this point, I shall merely offer some hypotheses. It may be profitable to draw a distinction between responses delivered directly by the Pythia to questions posed in a binary format and Delphic responses that have been cut free from Delphi itself to some extent and have entered an oral tradition of oracles. Yet this distinction should not be seen as impermeable. Maurizio has argued that the ambiguity of Herodotus’ Delphic oracles results from active interpretation of responses by the immediate enquirers, others whom they consulted at home and those who recalled the oracle later, who between them transformed simple answers into complex riddles through their strenuous efforts to adapt oracular responses to their circumstances and objectives.34 We might say, therefore, that the interpretative enclave identified by anthropologists in contemporary divination should be extended in ancient Greece to include the afterlife of Delphic responses as they are passed on, reperformed, recontextualized and recomposed, yielding new meanings for new circumstances.35 It would follow from this that Herodotus’ Delphic oracles can indeed shed light on divinatory thinking in general, even if they do not shed light on the historical specifics of particular consultations of the Delphic oracle, because they are products of the same thinking practices extended over time. In any case, since the focus of this chapter is on the process of interpretation in divination, rather than on the historicity of Herodotus’ oracles, one might still argue that the process of distilling a complex and problematic situation down to a single question that could be put to the oracle in binary format itself requires a considerable effort of synthetic reasoning, albeit before rather than after the consultation. Moreover, whatever one’s view of Herodotus, a wide range of other modes of Greek divination offer good evidence of an important role for ambiguous signs that demand interpretation through synthetic and relational reasoning. Examples may be found across the full range of Greek divinatory practice, from sacrificial divination, divination by chance events (cledonomancy) and the use of pre-existing prophecies.36 A first example may be found in book two of Xenophon’s Anabasis. After the Battle of Cunaxa, Xenophon and the ten thousand Greek mercenaries are at a loss. Though they triumphed in their section of the battlefield, their patron, Cyrus the Younger, was killed, leaving the Greeks stranded far from home. Trying to work out what to do next, the Greek commander Clearchus conducts a sacrifice to investigate possible courses of action. A distinctive feature of Greek sacrificial divination, in contrast with Etruscan or Mesopotamian techniques, is that there seem to have been very few signs that had a clearly defined meaning. If the lobe of the victim’s liver was missing, this was invariably taken as a very bad omen, but otherwise it seems that subtle variations of colour and texture were the main 58 Ralph Anderson elements a Greek diviner had to work with.37 In any case, the connection between the entrails of a sacrificial victim and the disposition of troops is far from transparent, so interpretation would require great creativity. Although Clearchus himself was confident in the prowess of his men, and even proposed carrying on the fight on behalf of Cyrus’ friends (Xen. Anab. 2.1.4), the sacrificial signs urged retreat. We might suspect a hunch on the part of Clearchus, his officers or his mantis that such a proposal would be unwise, nevertheless Xenophon’s account of Clearchus’ reporting of the omens suggests that he was initially somewhat puzzled by the results:

My friends, when I sacrificed with a view to advancing against the king, the omens were not favourable. And with good reason [εἰκότως], indeed, they were not favourable; for, as I now ascertain [ὡς γὰρ ἐγὼ νῦν πυνθάνομαι], between us and the king is the Tigris, a navigable river, which we could not cross without boats – and we have no boats. On the other hand, we cannot stay where we are, for we cannot get provisions; but as for our going to join the friends of Cyrus, the omens were extremely favourable. (Xen. Anab. 2.2)

The phrase ‘as I now ascertain’ (ὡς γὰρ ἐγὼ νῦν πυνθάνομαι) suggests a process of enquiry subsequent to the sacrifice that aimed to relate the sacrificial signs to features of the practical setting in which Clearchus was operating and which he was testing for possible courses of action. The sacrificial signs made good sense to Clearchus (that is, they appeared εἰκότως) only once he had set them in the context of the geography of the region: any advance would soon be halted by the River Tigris, which the Greeks could not cross. In the light of Ingold’s ideas about the relationality of knowledge and Bruner’s remarks about narrative, we might regard this process of enquiry not as a process of uncovering the pre-existing and independently given meanings of the sacrificial signs or of the river, but as a process of constituting the significances of both by actively bringing them into a relationship with each other in the context of the Greeks’ practical capabilities. We might say that the signs were validated in their negativity by being brought into relation with an impassable obstacle, while the impassability of that obstacle may have been confirmed, or at least emphasized, by being linked with the omens. After all, a more gung-ho commander might have declared that there was no good reason for the signs to be negative, accused his mantis of incompetence or cowardice, and pressed ahead either in ignorance of the barrier the Tigris would present, or in confidence that he would find boats when he got there.38 In such an admittedly counterfactual case, the signs would be drawn into relation with the character of the mantis and dismissed as meaningless, while the Tigris would not be brought into consideration at all. In either case, as Ingold’s remarks on the nature and generation of knowledge might suggest, the nature of the signs (valid or invalid) and the nature of the Tigris (passable, impassable or irrelevant) become known to Clearchus only through the relationship between them, a relationship which he himself actively constituted through the synthesizing process of divination. A story of blood, guts and guesswork 59 The operation of this relational mode of thinking can be seen in two further pairs of episodes, in each of which ostensibly identical signs are interpreted in very different ways. They illustrate the way in which the act of setting a sign in different contexts gives different meanings to the sign itself and to the setting in which it is put, and creates different narratives that link them together and man- date different proposals for action. In the first pair, sneezes are the sign. In the first episode, the sneeze is taken as confirmation of a proposed plan; in the second, it is skilfully diverted so as not to disrupt action that is already underway. The first example occurs shortly after Xenophon enters the action of the Anabasis as a character in his own right. With the Greeks once again at a loss, this time after the treacherous execution of their commanders during negotiations with the Persians, Xenophon resolves to take the lead. He makes a rousing speech, in which he attempts to persuade the Ten Thousand that – with the help of the gods – their best hope of safety lies in fighting their way out of Persia. As he utters the word ‘safety’ (σωτηρία), one of the men sneezes. As Xenophon-the-author presents the incident, this is spontaneously interpreted by the rest of the army as a positive omen. In one movement, they fall to their knees and worship ‘the god’ (τὸν θεόν). Xenophon-the-character swiftly identifies this god as Zeus Sōtēr, on the basis that he had been speaking of σωτηρία when the sneeze occurred, and proposes that they sacrifice to him when they reach a friendly land, a proposal which is universally accepted (Xen. Anab. 3.2). The second incident is recounted in Polyaenus’ Strategemata 3.10.2. As the fourth-century Athenian general Timotheus was preparing to set sail with his fleet, one of the sailors sneezed. The helmsman took this as a bad omen and ordered a halt, but Timotheus disarmed the sign by saying with a smile that it was hardly surprising if, with so many men present, one of them should sneeze. The sailors laughed and set sail. Neither account can be regarded as an unmediated account of real events. Xenophon represents himself as a model commander throughout the Anabasis, while the second example is preserved without further context in a didactic collection of deeds and sayings of great commanders compiled some five hundred years after the incident supposedly took place. However, for this very reason, both can be taken as examples of the principles governing the use of chance omens. In both cases, what is at stake is the frame within which the sneeze should be understood and within which its significance should be constituted. In the Anabasis, the coincidence of the sneeze with the word ‘safety’, occurring as it does during a moment of crisis and indecision, is intuitively seized upon as meaningful by the army. It allows Xenophon’s plan of action to be seamlessly connected to a vision of future safety, a connection which is quickly consolidated by Xenophon’s proposal of thank offerings to Zeus Sōtēr. In Polyaenus’ example, by contrast, all such future- orientated narratives are quashed by Timotheus, who places the sneeze in a framework of pure contingency rather than entailment, a frame in which meaningful links between events are denied, no future can be foretold, and the sneeze means nothing at all. The contrast suggests that in neither case is the significance of the sneeze inherent in the sneeze itself. Instead it emerges through the relations that 60 Ralph Anderson are created between the sneeze and the setting in which it occurs. This synthesis is an active process and, in both cases, it is a collective undertaking, requiring validation by the audiences of mercenaries and sailors. How active and open to contestation the processes of framing, synthesis and narrativization can be is suggested by the second pair of episodes, both of which concern eclipses. In 413 bc, the Athenians’ withdrawal from Syracuse was famously delayed by an eclipse of the moon. According to Thucydides (7.50), most of the Athenians were greatly disturbed by the eclipse and urged their commanders to wait. Fatally, Nicias, the overall commander, concurred. In 357 bc, an eclipse also threatened to derail Dion’s campaign to overthrow the tyranny of Dionysius of Syracuse. Plutarch grants Dion sufficient scientific knowledge not to fear eclipses, but states that his men were unsettled and needed reassurance from Dion’s mantis, Miltas. Miltas persuaded them that the omen should be read not against them, but against Dionysius. The omen, he said, portended the eclipse of something that was currently resplendent, and there was nothing more resplendent than the tyranny of Dionysius (Plut. Dion 24.1–2). Where Nicias was unwilling or unable to reframe the earlier eclipse in favourable terms, Miltas successfully redirected the ominous force of his eclipse by placing it in an alternative relational frame, one which linked the light of the moon to the splendour of Dionysius. By changing its referent, Miltas did not deny the significance of the eclipse, as Dion’s scientific outlook might have done, but redirected it. He thus effectively recomposed an omen that the men had already interpreted for themselves, and did so convincingly enough for them to accept his version as authoritative. One might note, however, that all these depictions of divinatory foresight were created with the benefit of hindsight. The perspective of the authors therefore differs markedly from that of the historical actors they describe. The author knows how events turned out and can thus seamlessly match divinatory predictions to the actual outcome, but the protagonists had no such certain knowledge and could only hypothesize about the future, however actively they were trying to shape it. Accounts of past divination are therefore constructed on a radically different epistemological footing from the events they purport to describe. In the episodes discussed so far, we have seen that divination generates useful insights through a process of synthesis and relational reasoning. Divination is thus a particularly acute illustration of the idea that we may understand a thing only by attending to its relationships. This lends divinatory foresight the structure of a narrative in which, as Ingold and Bruner suggest, the meaning and indeed the nature and identity of the individual elements and of the overall story itself are not given in advance but are formed and known through their interrelationships. It follows from this that the meaning of neither the overall story nor the individual elements within it can be known until the story is complete, the episode has run its course and all its elements are known. Since divination plays on the past, present and future, it projects a narrative which contains elements that are still to come and which, at the time of divination, are merely hoped for, feared or imagined – the destruction of the Persians, the discovery of Orestes, the fall of Dionysius and so A story of blood, guts and guesswork 61 on. Lying in the future, they must remain provisional until they are definitively realized through action. However, this provisionality is problematic only if we expect divination to function as a form of scientific prediction, rather than as a tool for guiding action. If instead we see the value of divination as lying in its performative efficacy, then provisionality is an inherent feature of the divinatory process. The only way to establish definitively that the omens spoke with good reason is to pursue the path that they recommend. Only with such hindsight can the foresight yielded by divination be fully vindicated. In other words, the final element of the divinatory episode is not the mantis’ announcement or the Pythia’s utterance, nor even its interpretation by the enquirer, but the success or failure of the action taken as a result. One might therefore expect the Greeks not to have obeyed omens and prophecies mechanically or slavishly but to have been alive to the possibility of fresh interpretations arising as a course of action unfolds. Indeed, the active interpretation displayed by the successful protagonists in the examples discussed earlier suggests just this point. However, the provisionality of divination is most clearly visible in accounts of failed interpretations, such as Croesus’, which unfolds from a failure of such awareness. Apollo even rebukes him for failing to recognize that other interpretations were possible: he should have asked which ‘great empire’ was meant (Hdt. 1.91). In such cases, the final element in the divinatory episode – the outcome of the actions taken – alters the nature and meaning of the preceding elements. This suggests that omens were, in principle, open to reinterpretation, right up to the point at which they were confirmed or denied in action. We can see this more clearly still in Mardonius’ disastrous use of prophecy before the Battle of Plataea, as recounted by Herodotus. Frustrated by inac- tion, Mardonius decides to attack the Greeks, against the advice of his allies. Summoning his officers, he asks them if they know of any prophecy about the Persians being destroyed in Greece. When they are silent, cowed by his authority, he announces that he knows of a prophecy that states that the Persians will come to Greece, plunder the temple at Delphi and then be destroyed.39 He declares that the Persians should therefore neither approach the temple, nor try to plunder it, and that they will thereby escape destruction (Hdt. 9.42). He then orders the army to prepare for battle. Herodotus then intervenes to correct Mardonius: he believed that the prophecy related to the Persians, but Herodotus relates it instead to the Illyrians and the army of the Encheles (Hdt. 9.43). Mardonius’ defeat and death at the Battle of Plataea brutally expose the provisionality of his interpretation, and demand that the prophecy either be abandoned or remapped.40 Rather than dismissing the prophecy altogether, Herodotus finds it a new relational frame and thus preserves its potency as a prophecy. To conclude, the examples presented here show ancient Greek divination to be a skilled and sensitive process of making practical sense out of situations that escape normal comprehension by providing a means of synthesizing numerous factors in the practical environment of the people concerned and reducing a surfeit of possibilities to just one. Because of its synthetic and improvisational character, the use of divination, whether by oracles, sacrifice or other means, 62 Ralph Anderson was not a matter of mechanically applying previously acquired knowledge to a novel situation, but of generating new knowledge and insights via a process of bringing prophecies or divinatory signs into a productive relationship with salient features of the lived-in world. This involved participants in a wide range of decisions: which oracles to bring to bear; what to take as an omen and what not; what counted as a salient feature of the environment and what was irrelevant; and how to relate them all to each other. Since the practical meaning and referents of both oracles and omens were underdetermined and negotiable, a stable meaning could be settled only by this process of synthesizing omen and world, a process which generated new meanings and significances for both the divinatory sign and the features of the world with which it was brought into relation. These meanings were negotiated communally, and the acceptance of an interpretation depended on many things, such as the persuasiveness of the interpreter, the trust placed in him or her by the audience and the plausibility of the interpretation. It would be easy to view this flexibility as a sign of charlatanry, but that would be to apply an inappropriate model of truthfulness to divination, a model based on ideas of scientific prediction. Divination is better viewed in terms of its performative efficacy, its ability to help people through difficult situations. Dealing with the relative merits of proposed future actions, divination yields insights that are akin to a story whose end is imagined but not yet known. Divination’s insights can be fully validated only by being put into practice and converted step by step from suggestion to reality. As befits a practice that offers performative efficacy not objective prediction, its insights remain provisional until the practical outcome is known. Divination offered the Greeks a means of gaining foresight by synthesizing what they knew and what they suspected about their practical setting, but it was a foresight that could be confirmed only in hindsight.

Notes 1 I would like to thank the participants in the UNISA colloquium and the two anonymous readers for their help with a number of points of detail and argument. 2 Petersen (2004) 00:55:25; Homer, Iliad 12.195–250. In both cases, the omen is an eagle attempting to carry a live snake, which bites it. In the Iliad, this omen is spotted in the heat of battle, and Polydamas is presented as one Trojan warrior among others. In Troy, the scene is transposed to a council chamber and Polydamas is presented as an aged priest, superstitious and ineffectual, in contrast with Hector’s practical man of action. This transposition accentuates the gulf in understanding between modern and ancient responses to divination. Cf. Price (2008) 125–6. 3 Shaw (1991) 137. When divination is seen in these terms, those who practise it must either be ‘charismatic charlatan[s]’ who manipulate their credulous clients, or just as credulous as those who consult them. See Peek (1991) 3. 4 Shaw (1991) 139–40. 5 For this summary of Shaw’s argument see Peek (1991) 135. 6 Parker (2000: 77–78) similarly highlights the pragmatic orientation of much divination as a ‘guide to action’ and its value in offering ‘apparently objective and uniquely authoritative’ advice, though he does not explicitly focus on the particular models of truth that underpin this orientation. A story of blood, guts and guesswork 63

7 Parkin (1991) 187; Whyte (1991) esp. 170–1. Classical scholars have also begun to recognize the part played by the consulter in divination, and increasingly regard the interpretation of an oracular response as integral to the divinatory process. See Maurizio (1997) 316; Parker (2000) 80; Maurizio (2013). 8 Parkin (1991). For ‘superabundance’ see Whyte (1991) 165, 170; Werbner (1973). 9 Shaw terms this ability of divination to encompass many potential meanings in its ambiguous and allusive language ‘cryptic potency’ (1991: 139–41). 10 Destroy a great empire: Hdt. 1.53; mule: Hdt. 1.55; Spartan oracles: Hdt. 1.66, 1.67. 11 Christ (2013) 237–42. 12 Barker (2006) esp. 9–14. Barker contrasts Croesus’ mishandling of his oracle with the Athenians’ open approach to interpreting the ‘Wooden Wall’ oracle by public debate (2006: 19–22). See also Maurizio (1997) 328 n. 67. 13 Kindt (2006) 37–41. 14 Parke and Wormell (1956) 1.131–9, 1.94–7. 15 Fontenrose (1978) 111–15 (Croesus), 123–4 (Sparta’s Tegean oracles) = Q99–101 and Q88–90 respectively. 16 For a recent summary of the debates on verse oracles and the typical format of questions, see Bowden (2005) 21–4, 33–8. 17 Maurizio (1997) 309, 311. 18 Maurizio (1997) 312. 19 Maurizio (1997) 313–14. 20 Maurizio (1997) 314–15. 21 Maurizio (1997) 316. Cf. Maurizio’s statement that ‘Delphic divination belonged to creative and engaged clients and story-tellers as much as to its shrine’ (2013: 75). This is a development of the now-common observation that debate over the interpretation of the oracle’s words did not distort the divinatory process but was integral to it. See Parker (2000) 80; Harrison (2000) 149–50; Kindt (2006) 39. 22 Thuc. 2.54: ἥξει Δωριακὸς πόλεμος καὶ λοιμὸς ἅμ᾽ αὐτῷ. For discussion see Maurizio (1997) 317–18. 23 Thuc. 2.54: οἱ γὰρ ἄνθρωποι πρὸς ἃ ἔπασχον τὴν μνήμην ἐποιοῦντο. ἢν δέ γε οἶμαί ποτε ἄλλος πόλεμος καταλάβῃ Δωρικὸς τοῦδε ὕστερος καὶ ξυμβῇ γενέσθαι λιμόν, κατὰ τὸ .εἰκὸς οὕτως ᾄσονται 24 Cf. Maurizio (1997) 317. 25 Ingold (2011) 142. 26 Ingold presents this position through a contrast with what he terms the ‘classificatory’ system of knowledge, in which ‘everything is what it is due to its own given nature’ which is fixed and inherent, so that ‘we do not have to attend to their relations to know what things are’ (2011: 160–1). 27 Ingold (2011) 162. For fuller discussion of these ideas, see Ingold (2000) particularly on transmission (132–51), on wayfinding, navigation and storytelling (219–42), and the fundamental discussion at 13–26. Antecedents to the relationalism which Ingold espouses can be found in, among others, Bateson (1973) and Merleau-Ponty (1962), both of whom inform Ingold’s position. A parallel emphasis on ‘perspectivism’ has emerged in Americanist, particularly Amazonian, anthropology (see Viveiros de Castro (1998); Vilaca (2005)), while studies that emphasize the embodiment of consciousness challenge the division between subject and object and thus also move towards relational approaches (see Csordas (1990); Kapferer (1997)). See also Bruner (1991). 28 Bruner (1991) 8. 29 Maurizio also argues that Greek oracles instigate a search for meaning but do not dic- tate how to achieve it (2013: 77). Cf. Kapferer on dreams in Sinhalese sorcery practices, which ‘impel those who experience them to search for their meaning’ and which are ‘packed with the portent which extends towards a meaning which is not yet there’ (1997: 229). 64 Ralph Anderson

30 Eur. fr. 973 Nauck: μάντις δ᾽ ἄριστος ὅστις εἰκάζει καλῶς. 31 LSJ s.v. εἰκάζω. For the history and core sense of εἰκάζω, see Chantraine, who views its meaning as progressing from image and resemblance to comparison and conjecture (1968–80: 2.354–5). 32 Cf. Maurizio (2013) 70. 33 The historicity of these oracles has been fiercely contested. Bowden usefully surveys the debate and suggests that some form of consultation would have been needed to sanction the evacuation of Athens, since evacuation entailed abandoning sanctuaries to be looted and destroyed (2005: 100–7). The historicity of these oracles is less important for present purposes than the insight which Herodotus’ reconstruction of the consultation and debate gives into contemporary conceptions of divination. 34 Maurizio (2013) 75. 35 In the case of the ‘wooden walls’ oracle, these meanings could range from ‘How shall we escape the Persians?’ to ‘How did our ancestors defeat them?’, the former having a prospective, practical force and the latter having a retrospective and perhaps ethical force. 36 Both Maurizio (2013: 70–2) and Harrison (2000: 122–3) see parallels between Delphic oracles and cledonomancy and discuss them in tandem. 37 Collins (2008b) esp. 320–23. 38 Cf. the disastrous decision of Neon to ignore negative sacrificial omens at Calpe Harbour in Anab. 6.4. For accusations against manteis see Soph. OT 380–403 and Ant. 1033–63. See also discussion in Flower (2008: 132–52). 39 Maurizio notes how Mardonius suppresses debate here (1997: 328 n. 67). 40 We might note that the Persians had in fact tried to sack Delphi (Hdt. 8.36–9), though Herodotus does not comment on this here. 5 Value-added divination at Dodona

Philip Bosman

Introduction The theme of this collection encourages us to look at proceedings at ancient oracles from a transactional point of view. If the reader will allow some indulgence in economic jargon, an oracular consultation may, from such a perspective, be viewed as follows. The sector may be described as services, and the commodity as information. A client visits an acknowledged point of sale, directs an enquiry and receives an answer. The client, or enquirer, reciprocates by offering a reward to the source of the information (i.e. the divinity). To this transactional skeleton, we may now add some flesh. Peculiar to this type of transaction is the separation of the two parties (the person requesting and the person supplying the information) by an otherwise impassable divide. A large part of the service thus involves brokerage mechanisms that facilitate a smooth transfer of information. Rewards or payments are culturally determined: intangible rewards are given in the form of prescribed ritual (worship) and esteem; tangible rewards take the form of votive offerings (e.g. tripods). The latter cannot cross the divide between client and service provider, and remains with the medium/broker at the point of sale. From an economic point of view, the transaction appears rather dubious. Suspicion increases when one adds that the peddled information is either some form of permission/sanction or pertains to decisions to be made involving future uncertainty, and that the service provider is per definition invisible (and has since disappeared)! Few classicists will be content with such a rendition of what happened at an oracular session in ancient Greece. We intuitively feel that much more was involved, a feeling supported by the success of the practice across the ancient world. Divination was not only spread far beyond the Graeco-Roman world, but also proved to be durable: all indications are that, with the odd exception prov- ing the rule, the participating parties were remarkably content with the practice, which only faded with loss of political and religious legitimacy.1 We may assume that divination or, more specifically, a visit to an oracle, satisfied a divergent set of needs. To the extent that profits were involved, participants were quite happy to reciprocate appropriately for services rendered. Already in Hesiod’s Catalogue of Women, enquirers at Dodona are said to bring along their gifts for the purpose of securing good omens;2 the votive offerings and other archaeological finds from 66 Philip Bosman the sites provide tangible evidence. The profits, of course, need not be conceived in material terms only, but also or even foremost in terms of non-tangibles like reverence and status. It seems probable, if we may continue with our market paradigm for a while, that the oracles were well aware of their standing in the religious services sector and that they jostled for market share or established niche markets. The Trophonius oracle at Lebadeia, for instance, smacks of catering for the more adventurous religious type.3 Arguing in similar economic terms, Bendlin claims that the demand for the wares offered by oracles remained relatively constant throughout antiquity, but that they had to compete ‘in the religious market-place with numerous “low” forms of mantic services’4 which ranged from local shrines to regional sanctuaries and itinerant seers. The oracles were regarded as special among such offerings but had their work cut out, so to speak, to offer a competitive product. Scholars often assume this competitive aspect among the oracular sites. Parke, for instance, assumes a bitter rivalry between the two foremost oracles Delphi and Dodona on relatively scant evidence, implying rather implausibly that the marketing desk at Dodona would leak a competing Deucalion myth to counter the Delphic version, or that the Delphi public relations officer would release a statement that Apollo acquired official family backing to speak on behalf of his father, presumably after the law boys at Dodona claimed that Zeus, and not his son, has copyright on the future.5 Nonetheless, purely on typically Greek agonistic grounds, some awareness of contending claims among the leading oracles does seem credible. My chapter’s title borrows, of course, the name of the sales tax system by which tax is calculated whenever value is added to a product/service, but I mean the term to refer to the additional value, or the draw cards that Dodona held for a visitor to the oracle over and above the obvious hope for an authentic divine response.

Dodona’s reputation: age, deity, location If we accept, then, that a visit to an oracle like Dodona entailed more than the bare exchange of information at the oracular session, and that some competition may be assumed between institutions offering such services, it seems legitimate to ask what swayed potential clients to opt for one oracle and not another, or, for that matter, another form of divination altogether. The most enticing area to explore in this regard is probably that of city states’ political manoeuvrings, which included shifting their allegiances between oracles, such as when Athens ditches Delphi for Dodona only to return to the closer site at a later juncture.6 But the fact of the matter is that formal public consultations must have constituted a minute percentage of the total number of oracular visits that took place; the vast majority were not related to city politics at all, but to various religious, social and personal issues.7 Most scholars would agree that oracular consultations were about obtaining answers on issues of real uncertainty or getting authoritative backing for a particular choice.8 Another important point is that the Greece of the archaic and classical eras – on which this essay focuses – had a good number of trans-regional oracles, among which Delphi and Dodona were often mentioned in Value-added divination at Dodona 67 tandem, presumably as indication of their particular esteem.9 Status and authority emerge as key considerations. A decision-making analogy can be drawn with the quandary of choosing schools in modern suburbia. Basically, there are two philosophies involved. The first is simple: the best school is the one closest to home. This is in fact what the evidence suggests for Dodona: the vast majority of queries came from visitors living in the vicinity.10 The second philosophy is more complex: of course good tuition is key, but various other factors come into play, such as cost and distance, tradition and reputation, quality of management, range and quality of facilities on offer, cultural activities, ease of acceptance and fit with the child’s personality. If we transfer this kind of decision-making to Dodona as a trans-regional oracle, we might be able to establish the added value such a consultation held for the visitor to the sanctuary. I would like to suggest that a visit to Dodona offered a set of associations, meanings and experiences peculiar to that oracle, and that these have not received the attention they deserve due to the fact that scholars tend to focus narrowly on what happened at the oracular session alone.11 In what follows, I briefly survey the archaeological evidence and what we know of its divination method, to clarify the different menu that Dodona offered in comparison with Delphi. I then consider the connotations that probably made up its reputation to the Greeks: its antiquity, its divinities and its location on the margin of the Greek world.

The site: archaeological evidence Dodona is situated in Epirus in north-west Greece, dominated by the lovely Mount Tomaros to the west.12 In terms of natural setting, the site impresses by its rural tranquillity, in contrast to the drama of Delphi. Since its identification in the 1830s, excavations have been conducted under various Greek archaeologists.13 The resulting architectural site map is dominated by an impressive theatre, one of the largest in Greece, a stadium, the remains of a number of ancient temples or treasuries14 and a Christian basilica, a prytaneion, a bouleuterion and the walls of an ancient acropolis. Artefacts and finds are scattered among various museums in Ioannina, Athens, Berlin, Paris and London, arguably the most interesting of which are the more than 4,200 oracular lamellae (sheets of plated lead inscribed with oracular questions), most of which are currently held in the Ioannina Archaeological Museum.15 The planished lead sheets all but settle the use of divination by lot (sortilege) at Dodona, at least (as far as I could establish) from the fourth century to 167 bc.16 Most of the buildings date from the third century bc, after the Molossian king Pyrrhus (292–272) made Dodona an administrative centre of the Epirote state.17 Evidently, Pyrrhus wished to put Dodona on a par with other integrated sanctuaries with theatres and athletic games linked to a religious festival such as at Olympia, Delphi, Epidauros and Nemea. The Naia festival in honour of the local deities Zeus Naios and Dione probably dates from this Hellenistic period.18 The site was sacked by the Romans in 167 bc, revived somewhat in the second 68 Philip Bosman century ad, and finally shut down when the Christians uprooted its famous holy oak by the end of the fourth century ad. If we reserve our attention to the archaic-classical period, the archaeological site map can be misleading. For the Dodona of earlier times, we need to remove most of the constructions from the scene, leaving only a basic hiera oikia of Zeus adjacent to the site’s renowned sacred oak tree. Dakaris’ reports claim that a new hiera oikia was built in the late third century bc. Before that, the sanctuary was much simpler: it received an enclosure wall for the first time during the fourth century. And before that, the precinct was without walls. The small original temple of Zeus beside the oak tree was only built at the start of the fourth century. What kind of structure if any preceded that is unclear, but evidently the oracle was very plain during archaic times and even through the fifth century, the oracular tree possibly only surrounded by votive offerings.19 A second interesting aspect emerging from Dodona’s archaeology is a Neolithic layer directly overlain by Hellenic geometric and artefacts datable to the eighth century bc. Nomadic habitation of the site goes back to prehistoric times, and the occasional Mycenaean artefacts (c. 1500 bc) do not point to permanent settlement either. The plausible conclusion is ‘a local perpetuation of the Neolithic style down to the time when archaic Greece began to introduce its influence into Epirus’.20 A third point of interest is that the evidence points to an older, chthonic cult below the layers belonging to the cult of Zeus. The obvious link is with Dione, Zeus’ consort at the site in historical times, as a possibly renamed earth/fertility goddess. Plausibly, the two deities had coexisted ever since the introduction of the male Zeus under Greek influence, although two Neolithic deities cannot be excluded. In brief, then, Dodona was an open-air oracle for the first period of its life, first dedicated to an earth deity or deities and then to a local Zeus with consort. Before the eighth century, the sanctuary was Neolithic. Activity picked up towards the end of the eighth century but declined again during the seventh, to reach its zenith during the sixth century. Few but high-quality dedications come from the fifth century, and even fewer from the fourth. The archaeological record does not necessarily give an accurate view of activity at the site, since subsequent plundering could have impacted on the remains found by archaeologists. For the time being, however, we may accept that Dodona remained a modest rural site during the height of its influence in the sixth and fifth centuries, and also through most of the fourth.21 Compared to Delphi with its splendid setting, temples, treasuries, theatre and stadium, Dodona was decidedly rural and low-key in appearance until the deliberate Molossian appropriation from the late fourth century to lift its status.

The divination method If not on its architectural splendour, could it be that Dodona’s esteem hinged on its particular divination method? An oracle’s main function was to provide Value-added divination at Dodona 69 trustworthy and authoritative communication with the gods.22 This Dodona could claim on the authority of those frequenting it most: Pausanias mentions that ‘to those inhabiting the mainland, to the Aetolians and their neighbours the Acarnanians and Epirotes, the doves and the oracles from the oak appeared to hold the most truth’.23 But how did the divination take place? Cicero tells us that Graeco-Roman culture distinguished between ‘natural’ and ‘technical’ forms of divination.24 The latter includes all forms of sign interpretation and is assumed to be a skill that could be acquired. The former, on the other hand, assumes possession and direct communication from the god, whether in the form of dreams/incubation or ecstatic prophecy. Both kinds claimed divine intervention but it does appear as if the most coveted form was the enthused state of inspired or ecstatic prophecy. All indications are that Delphi’s reputation rested to a large degree on the Pythia, who uttered her prophecies under the spell of Apollo. Could Dodona also claim ‘altered states of consciousness’, as Plato thought and as Sarah Iles Johnston still holds? The literature on how oracular communication took place at Dodona is baffling. Sortilege, as evidenced by a Callisthenes fragment,25 and of course by the lead plates, no doubt played a major part. Sortilege occurred at Delphi as well, but while Delphi could also boast the services of the Pythia, her counterparts’ role at Dodona is unclear. Homer’s reference to Dodona’s Selloi26 is confusing, as some scholars assume that they had some hand in the divination procedure, but since they failed to make any historical appearance, they may safely be discarded for the archaic/classical period. Herodotus claims to have spoken to the three female priestesses in his story about the oracle’s origins, but some significance should be attached to the fact that they were not known as prophetesses.27 Their name, ‘doves’ has an obvious link to the oracle emblem, the ‘high-foliaged oak of Zeus’; but does not tell us whether they were inspired or interpreting.28 Late references to sounding cauldrons and murmuring springs may be left out of the equation with relative safety unless, and this remains a possibility, Dodona always resorted to various mantic procedures. The latter, however, only supports the conclusions we may draw from the literary evidence regarding Dodona’s mantic procedure: (1) while it was not an unreliable divination method, sortilege did not enjoy the same prestige as inspired prophecy;29 (2) most Greeks’ knowledge of divination at Dodona was formed by literature more than by actual experience of how the oak and the doves communicated;30 the role of the Peleiai/Peleiades remains obscure but was probably of the ‘technical’, ‘interpretative’ kind.31

Explaining Dodona’s esteem Thus, in all probability, sixth/fifth century visitors to Dodona did not come specifically for the divination method, nor for the political clout, splendour or side attractions (sport, culture, networking) of the sanctuary. Nevertheless, it became a major oracular site and often featured in tandem with Delphi.32 So, what do we know of Dodona that could have been the reason for its great esteem? 70 Philip Bosman Antiquity The first aspect certainly counting in Dodona’s favour was its reputed great antiquity. Herodotus tells us that Dodona was not only the first oracle in the whole of Greece, but also that it was here that the Pelasgians first learned the names of the gods.33 Homer’s reference to Pelasgian Zeus of wintry Dodona gives the impression of already established venerability at the time the epic took shape. Hellenistic versions of the Deucalion myth linked to the establishment of the oracle play on existing claims to, or rather associations with, Dodona’s great antiquity.34 The tree cult further supports claims to pre-Hellenic existence. Tree worship was known to the east and north of Greece, and was a feature of Minoan cult, but scarce among the Greeks themselves.35 It may be explained as an aspect of Neolithic religion, where sacred trees developed as proxies for the life-giving forces of sun and water. Initially, they function ‘as symbols or simplified images, while at the same time they stand pars pro toto for the sacred place and for the great supernatural power which is thus made perceptible’.36 In a further step, techniques are developed to compel the indwelling gods to divulge vital information. This evolutionary process suggests a history from long before the Greeks introduced their religion to the region, and they acknowledged the fact with their reverence of Dodona’s ancient cult.

Deities In the male-dominated literature we have on ancient Dodona, the site and the tree were mainly associated with Zeus, Dione receiving scant mention. Some correc- tive to this perspective is offered by the excavated lead tablets, which fairly consistently invoke both deities.37 Dione’s general obscurity suggests her age rather than anything else. She has very little presence anywhere else in Greece, suggesting a special link to Dodona, probably from times immemorial. The Greeks offered some late aetiological myths for the name Dodona, but it is not inconceivable that the double invocation of historical times is a remnant of a memory of two ‘indwelling’ gods from the earliest layers of the site, even before the local deities were renamed when tribes of Indo-European stock came into contact with local Neolithic worshippers. Scholars interpret the meaning of Zeus’ epiclesis ναιος variously,38 and the (albeit amateur) writers on the lamellae did not consistently get the spelling right either. One contending option is that it simply derived from the verb ναίω (‘dwell, inhabit’), linking up nicely with the explanation of the tree cult as manifesting the divine force. The two deities, whether or not of both genders from the start, predated Greek settlement of the area and were, on the Greeks’ arrival, associated with the Indo-European sky god. Of their theology our sources provide little evidence, but it is certainly wrong, as Hammond assumes on the basis of votive offerings, that Zeus Naios was a war god.39 The kinds of requests on the lamellae do show great variety, but rather suggest a Zeus of communal and household affairs. Beyond the deities’ assumed concern with the fortune and general welfare Value-added divination at Dodona 71 of the enquirers, Dodona’s indwelling Zeus gained the attributes of the great deity spanning the ages, the impartial, all-knowing Zeus of Panhellenic potential that Burkert settled on.40 This idea still reverberates in the song of the Dodona priestesses (recorded by Pausanias as: ‘Zeus was, Zeus is, Zeus will be; O great Zeus’. Zeus’ reach across time is matched by Dione’s nurture from the earth when the priestesses’ song continues with: ‘Earth sends up the fruits, therefore sing the praise of Mother Earth’.41

Location The other aspect of the Dodona product relates to its location. It has been argued by Parke that the relative remoteness of Epirus was to its disadvantage: it was difficult to reach from any side, due to both distance and terrain, so that a visit would have required considerable effort. However, in regard to oracles, remoteness may have had a negative impact on visitor count but not necessarily on esteem. In fact, it seems that the more remote the oracle, the more trustworthy the ancients regarded its utterances. The extremely remote Siwah is a case in point. In this sense, Dodona profited from being removed – by distance and culture – from the factional politics and alliances that constantly threatened to compromise Delphi, and sometimes did. Remoteness, as social geographers would tell us, is very much a social construct, and this rings true in the case of Dodona, whose remoteness was relative to where in the Greek world one came from. Nor did it entail sparse population: the region appears to have been quite densely inhabited before the Roman conquest had a depopulating effect.42 Perceived remoteness probably had more to do with ambiguity in relation to Greek identity during the classical era, regarding language (northern dialect) and lifestyle (rural). Thucydides, for instance, considers the inhabitants of the region as ‘barbarians’ (2.80), but the Homeric ring to his story of Themistocles’ visit to the court of the Molossian king Admetus might well point to the region’s association with ‘olde worlde’-liness in which the customs of a pre-urban Greece still held sway.43 The Homeric connotation certainly played its part when the Molossians of Hellenistic times constituted their ethnic identity by reference to their ancestry through Neoptolemus and Achilles to Aeacus. In an analysis of the roles of Dodona and Delphi in Attic drama, Castrucci observes that both played a part in the oracular travels of the heroes, but in contrast to Delphi’s ‘tragic’ role at the start of such journeys, Dodona’s link is with the return home, ‘the recovery of the oikos, the preservation or rebuilding of family bonds and ties of genos’.44 These connotations with roots and homeland, Castrucci claims, were not mere ‘mythical literary topoi, but … historical-religious givens’, attributable to real sacred connotations of the Epirotic oracle of Zeus’.45 They are also observable in the set of myths on the region. Homer’s and Herodotus’ mention of the Pelasgians points in this direction. But Dodona is also associated with the myth of Hellos and the region of Hellopia, which again was already in antiquity seen as etymologically related to Homer’s Selloi. Significantly, according to 72 Philip Bosman legend, the Greeks (Graeci) who crossed the Adriatic to settle in Italy were from Dodona. Thus, as Kittelä has argued, both the name the Greeks called themselves – Hellenes – and the name by which the Romans referred to them – Graeci – have strong connections to Epirus in general, and to Dodona in particular.46

Conclusion It must be of some significance that, when the Epirotes started with their nation- building project from the end of the fourth century, they looked to Dodona to con- solidate their identity. To the Greeks, the sanctuary was a primordial place which predated ‘the complications of civilisation’.47 Zeus Naios and Dione, the ‘inhabiting’ deities were known for their Panhellenic, all-encompassing impartiality and Zeus the ‘god of roots’48 and of spanning the ages. The myth complex, probably not developed for, but because of such connotations, made Dodona the ‘spiritual … homeland’ of the Greeks.49 These aspects have little to do with the information transaction that took place during the oracular consultation. An enquirer at Dodona was probably not only interested in the outcome of an enquiry. Rather, in making the journey to Dodona, he or she participated in an event that had very much to do with self-discovery: a deep past with connotations of a natural divine presence, of belonging and reintegration into a wider Hellenic identity.50

Notes 1 The complexities surrounding oracular decline cannot be dealt with here, but it was by no means an even process into the Christian era and beyond. On the final years of Didyma and Delphi, see Athanassiadi (1991); on Dodona see Dieterle (2007) 23–24. For a brief recent treatment of the topic, see Stoneman (2011) 199–219. My thanks to the anonymous reviewer for this reference. 2 Hes. Cat. fr. 97, Scholiast on Soph. Trach. 1167. In terms of the transactional paradigm, such gifts were not strictly payment for services, but upfront tokens to limit unwanted outcomes. 3 Paus. 9.39. See also Ogden (2001) 80–5. 4 See Bendlin (2011) 209; Bowden (2013) 55. On cult competition, though in later times, see Chaniotis (2010) 113–38. 5 See Parke (1967) 15, 40–2, 110, 160. 6 On oracles and Greek politics, see Parker (1985); Bowden (2005). 7 Lhôte (2006); Dieterle (2007) 70–85 (with rich footnoting); Johnston (2008) 68–71. 8 Bowden (2013) 42. 9 Rosenberger’s map (2001: 214–15) indicates Didyma, Claros, Lebadeia, Oropos, Abai and Patara as further trans-regional oracles. 10 Johnston (2008) 61. 11 Johnson (2008: 72) mentions the possible sense of the gravity of the occasion associated with an oracular visit, in particular to one of the major sites. 12 The mountain now bears the name Mitsikeli. See Vandenberg (2007) 23. How the ancient Greeks approached it is not clear. Hammond (1967: 172) suggests that the ancient Sacred Way passed from the south (Ambracia) along the narrow Louros River valley and the eastern slope of the Tomaros Mountain. See also Dieterle (2007) 8–9. I would think, however, that many visitors may have taken the route through Thessaly in the east or, more probably, from the Adriatic seaboard, the mouth of the Acheron or Value-added divination at Dodona 73 the towns opposite Corcyra. Archaeological evidence suggests at least three gates, see Piccinini (2016) 163. 13 See the survey of the history of archaeological work at Dodona in Dieterle (2007) 11–15. 14 Piccinini (2016) recasts as treasuries most of the smaller structures that Dakaris identified as temples. 15 Published in Lhôte (2006). Particularly valuable is Eidinow’s (2007) categorization of the lamellae. See recent bibliographies on Epirus and Dodona in Dieterle (2007) and Meyer (2013). 16 Meyer (2013: 20 n. 33) agrees that most lamellae are fourth and third century bc but questions the use of some letter types for an early dating; Lhôte (2006: xv) believes that this form of consultation ended with the Roman sacking of the sanctuary, and Meyer (2013: 135 n. 445) concurs. Piccinini (2016) argues that the lead tablets served as records of consultation rather than playing a part in the consultations themselves. 17 For a recent revaluation and redating of Molossian history from the late fifth century bc, see Meyer (2013) 114–35, esp. 126–7. 18 Only inscriptional evidence of this festival survived. See Dieterle (2007) 42–3; Meyer (2013) 36–37, n. 82. 19 The ring of tripods suggested by Dakaris derives from a third-century bc explanation of the expression τὸ Δωδωναῖον χαλκεῖον which had already been rejected in antiquity. See Bosman (2016) 184–92. 20 Parke (1967) 99. 21 See Piccinini (2016: 152, 169), who refers to the late monumentalization of Dodona, comparable only to Samothrace, due to political and economic developments in the region after the rise of Macedonian power. 22 Bowden (2013: 42–3) questions the assumption sometimes underlying modern scholarship that oracles functioned mainly to sanction decisions already made, and argues that, in classical Greece at least, consultations were about issues of real uncertainty. 23 Paus. 7.21.2: τοῖς γὰρ τὴν ἤπειρον ταύτην οἰκοῦσι, τοῖς τε Αἰτωλοῖς καὶ τοῖς προσχώροις αὐτῶν Ἀκαρνᾶσι καὶ Ἠπειρώταις, αἱ πέλειαι καὶ τὰ ἐκ τῆς δρυὸς μαντεύματα μετέχειν μάλιστα ἐφαίνετο ἀληθείας (‘To those inhabiting this mainland, to the Aetolians and their neighbouring Acarnanians and Epirotes, the Peleiai and the oracles from the oak appeared the most truthful’). 24 Or, as in Bonnechere (2007: 150–5), between ‘inspired’ and ‘inductive’/‘interpretive’ divination. 25 Cic. Div. 1.76, quoting Callisthenes (Jacoby FGrH 124 (Callisthenes): maximum vero illud portentum isdem Spartiatis fuit, quod, cum oraclum ab Iove Dodonaeo petivis- sent de victoria sciscitantes legatique vas illud, in quo inerant sortes, collocavissent, simia, quam rex Molossorum in deliciis habebat, et sortes ipsas et cetera, quae erant ad sortem parata, disturbavit et aliud alio dissupavit. Tum ea, quae praeposita erat ora- clo, sacerdos dixisse dicitur de salute Lacedaemoniis esse, non de victoria cogitandum (‘But the greatest portent that was given to the Spartans was this: when they consulted the oracle of Jupiter at Dodona on the question of victory and their ambassadors had set up the which contained the lots, a monkey, which the king of the Molossians kept among his pets, upset the lots themselves and everything else that had been prepared for the lot-taking and scattered them in every direction. Then it is said that the priestess who was in charge of the oracle said that the Spartans should think not about victory, but about safety’). Translation in Wardle (2006) 70. 26 Hom. Il. 16.233–5; Achilles speaking): Ζεῦ ἄνα Δωδωναῖε Πελασγικὲ τηλόθι ναίων / Δωδώνης μεδέων δυσχειμέρου, ἀμφὶ δὲ Σελλοὶ / σοὶ ναίουσ᾽ ὑποφῆται ἀνιπτόποδ ες χαμαιεῦναι (‘Lord Zeus, of Dodona, Pelasgian, dwelling afar, ruling over wintry Dodona – and around you dwell the Selloi your interpreters, of unwashed feet, sleeping on the ground’). 74 Philip Bosman

27 Hdt. 2.55.1–3: ‘The following I was told by the prophetesses [προμάντιες] of the Dodonaeans: that two black doves flew from Egyptian Thebes, and the one arrived in Libya and the other with them. It sat on an oak and spoke with a human voice that it was fated for an oracle-centre of Zeus to be set up there’; Hdt. 2.54.1–3: ‘The priests of Zeus at Thebes said that two women, priestesses [ἱρείας], were carried off from Egyptian Thebes by Phoenicians, and they heard that one of them was sold to Libya and the other to the Greeks. It was these women, they said, who first founded the places of prophecy among these two peoples’ (transl. Parke 1967: 54). 28 Hom. Od. 14.327–8 (Odysseus pretending to be Eumaeus): τὸν δ᾽ ἐς Δωδώνην φάτο βήμεναι, ὄφρα θεοῖο / ἐκ δρυὸς ὑψικόμοιο Διὸς βουλὴν ἐπακούσαι (‘Odysseus, he said, was going to Dodona, so that from the god’s high-foliaged oak tree he might hear the will of Zeus’). 29 While the distinction between interpretive and inspired divination is to a certain extent artificial, both ancients and moderns seem to put greater store in those forms receiving communications directly from the gods. See Pl. Phdr. 244a–d; Cic. Div. 1.6.11–12; Parke (1985a) 200; Lewis (2014) 7–18. 30 For references to Dodona in post-classical literature see Dieterle (2007) 300–41; Parke (1967) 82–93. 31 In this instance, Burkert’s use of sources (1985: 114, 393) seems rather suspect. 32 The oracle of Zeus-Amun at Siwah in the Libyan dessert completes the trio of truly great oracles. See, for instance Aristoph. Av. 716. 33 Hdt 2.52.1–2: ἔθυον δὲ πάντα πρότερον οἱ Πελασγοὶ θεοῖσι ἐπευχόμενοι, ὡς ἐγὼ ἐν Δωδώνῃ οἶδα ἀκούσας, ἐπωνυμίην δὲ οὐδ᾽ οὔνομα ἐποιεῦντο οὐδενὶ αὐτῶν: οὐ γὰρ ἀκηκόεσάν κω. θεοὺς δὲ προσωνόμασαν σφέας ἀπὸ τοῦ τοιούτου, ὅτι κόσμῳ θέντες τὰ πάντα πρήγματα καὶ πάσας νομὰς εἶχον. ἔπειτα δὲ χρόνου πολλοῦ διεξελθόντος ἐπύθοντο ἐκ τῆς Αἰγύπτου ἀπικόμενα τὰ οὐνόματα τῶν θεῶν τῶν ἄλλων, Διονύσου δὲ ὕστερον πολλῷ ἐπύθοντο. καὶ μετὰ χρόνον ἐχρηστηριάζοντο περὶ τῶν οὐνομάτων ἐν Δωδώνῃ: τὸ γὰρ δὴ μαντήιον τοῦτο νενόμισται ἀρχαιότατον τῶν ἐν Ἕλλησι χρηστηρίων εἶναι, καὶ ἦν τὸν χρόνον τοῦτον μοῦνον (‘The Pelasgians formerly made all their sacrifices with an invocation to “the gods”, as I know through having heard it at Dodona, and they applied no title or name to any of them. For they had not heard their names. [They called them gods] for they ad placed in order all things and all activities, and maintained them. After a long time had elapsed, they heard from Egypt the names of the other gods which had come from there, but the name of Dionysus they heard much later, and after an interval they enquired of the oracle at Dodona about the names. For that oracle-centre is accepted as the oldest of the places of prophecy among the Greeks and at the time it was the only one’ (transl. Parke 1967: 53)). Nesselrath (1999: 13–14) concludes that pre-Herodotean evidence exists for links between Dodona and the sanctuaries of Thebes and Siwah, and that Herodotus’ observations on the origins of Dodona cannot simply be dismissed as ‘ein arglistig gesponnenes Netz von Publikumstäuschungen’. See also Aesch. Supp. 244–59; Ps.-Aesch. PB 829–40. 34 Parke (1967: 40) refers to versions by Thrasyboulos and Acestodorus. 35 On tree cult in ancient Greece, see Smardz (1979); on oracles and sacred trees in Asia Minor, see Graf (1993). 36 Cf. Persson (1942) 166. 37 There are a number of these: τὸν Δία τὸν Ναῖον καὶ τὰν Διώναν; τωι Δι τωι Νάωι καὶ τᾶι Διώναι; Δὶ Νάωι καὶ Διώναι; τὸν Δία τὸν Νάον καὶ τὰν Διώναν. 38 For discussion with reference to scholarly literature see Dieterle (2007: 41–2), who herself opts for ‘der zum Wasser Gehörige’. 39 See Hammond (1967) 369; Dieterle (2007) 45–7. 40 Burkert (1985) 130. 41 Paus. 10.12.10: τὰς Πελειάδας δὲ Φημονόης τε ἔτι προτέρας γενέσθαι λέγουσι καὶ ᾆσαι γυναικῶν πρώτας τάδε τὰ ἔπη˙ “Ζεὺς ἦν, Ζεὺς ἐστίν, Ζεὺς ἔσσεται˙ ὦ μεγάλε Ζεῦ. Γᾶ Value-added divination at Dodona 75

καρποὺς ἀνίει, διὸ κλῄζετε Ματέρα γαῖαν”. (‘They say the Peleiades were born even earlier than Phemonoe and the first women to sing this song: ‘Zeus was, Zeus is, Zeus will be, o great Zeus! Earth sends up fruit, therefore celebrate Mother Gaia’.) 42 As the evidence from Hesiod’s Catalogue indicates. Strabo, for instance, mentions at 7.7.5, on authority of Theopompus, that the main grouping of the Thesprotians, Molossians and Chaones consisted of fourteen tribes. 43 Thuc. 1.136–7. See also Meyer (2013) 115. 44 Castrucci (2012) 2. Castrucci is specifically discussing the Prometheus Bound, Sophocles’ Trachiniae and Odysseus Thorn-Struck, and Euripides’ Andromache and Phoenissae. 45 Castrucci (2012) 3 n. 6. 46 Kittelä (2013): 33–34. See Arist. Mete. 352b1; Strabo 7.7.10. 47 Castrucci (2012) 12. 48 Castrucci (2012) 16. 49 Mitchell (2001) 342. 50 A special word of thanks to the anonymous referees of the article who provided useful leads, which I followed up as well as I could. Unfortunately, suggestions of a comparison between Greek divination with other cultures, in China and Africa in particular, will have to be left for another study. 6 Divination and profit in the Roman world1

Federico Santangelo

The title of this volume is not just about thoughtful wordplay. The prophets and profits theme takes us right to the core of one of the major areas of tension surrounding any divinatory practice and shaping any debate on divination, not just in antiquity. The aim of this chapter is to chart some of the major developments of this issue in the Roman world; in doing so, I hope to draw attention to some wider problems. Let us start from a fairly familiar story. In 134 bc, Scipio Aemilianus was entrusted with the leadership of a new campaign at Numantia, after the debacle that Rome had suffered three years earlier and his election to a second consulship. He knew that restoring discipline within the Roman ranks was a central part of the brief. His first act as he arrived at the camp was, according to Appian, to expel all the merchants, prostitutes, diviners and fortune tellers, whom the soldiers constantly consulted in their anxiety over the prospects of the campaign. He also prevented the introduction into the camp of ‘anything superfluous’, including sacrificial victims prepared for divination.2 Sacrifices were of course followed by the division and distribution of the meat, and Aemilianus wanted to deny even that distraction to his soldiers at a crucial phase of the campaign. Extispicy – the ritual consultation of animal entrails – is likely to have been discontinued for some time. The initial targets of these draco- nian measures, however, are more instructive, and take us right into the thick of the topic of this chapter, and of the title of this collection. Aemilianus expelled individuals who had the will, skill and power to address certain needs of the soldiery: whether the desire for goods and commodities that were part of their usual supplies, sexual appetites, or the all-too-important reassurance that a diviner could offer to a man fighting a war in lands far from home. All three groups were seen as destabilizing presences, which Scipio could not tolerate. The risks the diviners presented to his authority and to the morale of the troops required swift action. In Appian’s text there is no hint of a moral judgement: the emphasis is on pragmatic considerations. The passage simultaneously gives a glimpse into a usually hidden economy around a Roman camp, populated by practitioners who offer their services to the soldiers for a fee. Many details remain unclear – including their provenance and their number – but their presence is described as a straightforward feature of the landscape that requires no justification. The presence of diviners and fortune tellers within this constituency also reminds Divination and profit in the Roman world 77 us of one of the facets of the link between divination and money, in the Roman world and beyond it. Divination was a craft that could be practised for money, and for which there was often demand, especially at times of uncertainty and instability. Recognizing this point does not necessarily amount to an uncritical acceptance of the so-called ‘market model’, which has proved so influential and so divisive in the recent historiography on Roman religion.3 It is a reminder, however, of the fact that the history of Roman divination cannot be simply the history of public divination. Aemilianus’ hostility to diviners who charge a fee for their services resonates with a concern over the link between divination and money that appears to pervade Roman culture more broadly. The foundation myth of one of the main tenets of Roman public religion, the Sibylline Books, denounces that tension in very strong terms. Several sources relate the story of the visit that King Tarquin (it is unclear whether Priscus or Superbus) received from a foreign woman, who offered to sell him nine books of prophetic material at a certain price.4 The sequence of events is well known when Tarquin refused to pay, the woman burnt three books and came back to the king with the same request; when she met with another refusal, she burnt three more books and came back to the king without lowering her price. Only at that point did the king realize that the matter required further reflection and decided to seek the advice of the augural college (another cohort of divinatory practitioners, albeit of a different kind). Their ruling was unequivocal: the books were a divine gift, and had to be purchased immediately. Once she received her money, the woman swiftly disappeared. If a lesson on divination may be drawn from this elusive story, it is that no price can actually be put on divinatory knowledge: the paradoxical behaviour of the Sibyl makes clear that it is, quite literally, priceless. Her request for payment is an exception that proves the rule: the factor that sets in motion the process leading to the acquisition of the books, and a sign that Tarquin must detect and make sense of, once he has realized that the books’ price appears to escape the usual principles that determine value formation. Perhaps revealingly, no reference is made in any source to what the Sibyl did with the fee she received. In the story of Tarquin and the Sibyl, money is most emphatically a false problem. To return to Appian’s evidence for the Numantine War, the sellers of oracles who populated Scipio’s camp upon his arrival are of course not unparalleled in the surviving evidence. Scipio’s reservations are not unfamiliar either: they may not have been led by moral considerations, but were certainly driven by deep-seated political and social concerns. Two generations before him, in a different context, the Elder Cato had voiced a similar approach in a well-known precept on the kind of religious knowledge that should not be accessible to a farmer (uilicus): haruspicem, augurem, hariolum, Chaldaeum nequem consuluisse uelit.5 It is debatable whether this passage should be read as evidence for Cato’s scepticism regarding the value of the advice provided by these experts,6 but it certainly reveals specific concerns about the nature of the divinatory advice that was accessible to an individual of lowly social status in a rural setting, and shows the intention to restrict access to it on the part of the landlord.7 At the same time, it provides us with 78 Federico Santangelo a glimpse of the array of divinatory practices that were available in the Italian countryside in the first half of the second century: their range is as wide as their exact remit is unclear. Cato lists ‘Near Eastern’ astrologers along with diviners of unclear status, practitioners of augury (no doubt ornithomancy) and haruspices, who practised the Etruscan lore based on the reading of animal entrails and the interpretation of lightning. Haruspicy is often associated with the running of the Roman government, both on a day-to-day basis and on major public occasions. There is ample evidence for its use in private settings, however, which should not be confused with the habit of some leading political figures, such as Gaius Gracchus, Sulla or Caesar, to rely on the advice of an individual haruspex.8 Cato’s cursory reference is not isolated, and does not just reflect Republican concerns: we find it echoed in Columella.9 The potential for ridiculing certain groups of practitioners of haruspicy was clearly high: the playwright L. Pomponius (flor. 89 bc) wrote a fabula Atellana entitled Haruspex uel pexor rusticus, ‘The haruspex, or country barber’ (Nonius 830 L), which probably featured the character of a charlatan of rural descent. The most instructive piece of evidence, however, is a passage in the De Divinatione in which Quintus Cicero, the character who argues the case for the reliability of divination, refers to a remarkable decision taken by the Roman government some time during the Republic, ‘when the empire was flourishing’ (tum cum florebat imperium).10 The Senate had decreed that the youngsters of ten distinguished Roman families be trained in the ancestral divinatory lore (disciplina): their education was to take place in Etruria, but ostensibly under the supervision of the Roman state.11 Haruspicy was by then already recognized in Rome as a craft that enabled the interpretation of prodigies, especially those relating to lightning. It had the great advantage of being teachable, and of being based on a set of technical principles that were codified in a written form; it was on the opposite end of the spectrum to inspired divination. Against that background, it is unsurprising that the Roman government took a direct interest in the way in which the technique was taught and transmitted. Quintus, however, voices a specific concern of the Senate: that such an important craft would be undermined by the involvement of practitioners from poor or lowly backgrounds. It was thus necessary to prevent haruspicy from becoming an object of trade and profit (ad mercedem atque quaestum ), making sure that those who practised haruspicy on behalf of the Roman government did not do it for a living, and that money would not be a factor in their interaction with the Roman government. That need was all the more acutely felt at a time when haruspicy was increasingly open to the pressure and potential of prophetic divination: from the early second century bc it becomes apparent that the haruspices were increasingly keen to produce fully fledged predictions of future events, rather than offering interpretations or rulings on specific signs and events.12 The point applies eminently to public (i.e., for the purposes of the present discussion, state-managed) divination, but has a wider import. The authority of a diviner who lived off his own trade was potentially always in question, and the risks that were associated with his work were always significant. The implication is that haruspicy practised in a public setting must be a prerogative of individuals who did not need Divination and profit in the Roman world 79 to practise it for a living; indeed, haruspicy was a valuable ground of interaction between members of the Etruscan elites and the senatorial nobility throughout the Republican period. The system for the interpretation and expiation of public prodigies, which reached far into peninsular Italy, also played an important role as an avenue of religious dialogue and integration, and in turn challenged and consolidated the fabric of Roman public religion. It was in the cycle of prodigy reporting and expiation that the haruspices played an increasingly significant role.13 To put it in the briefest and crudest terms: if a divinatory practice was to have a major place in public religion, it had to be set free from any association with money and gain, and was therefore to be entrusted to the wealthy. Stressing this point does not mean advocating a narrowly political and manipu- lative use of divination, or a sceptical or utilitarian approach to that craft. It is, on the contrary, a recognition of its significance and the importance of keeping it in safe hands. Tellingly, the point is advocated by Quintus, the supporter of divination, and not by his brother Marcus, who voices scathing views on the value of any form of prophetic divination. It is not an isolated remark, but one that impinges on Quintus’ central contention: divination is a worthwhile practice, which has a sound philosophical and theological basis, but only makes sense within a clear framework of rules and constraints, and requires a complex and teachable intellectual tradition upon which to build. Precisely because divination can be so significant, a full set of restrictions and qualifications must be in place for its practitioners to be suitably validated. Quintus’ case hinges to a considerable extent on a clear sense of the kinds of diviner that deserved to be listened to. Right at the end of his lengthy defence of divination, Quintus indulges in an excusatio non petita, in which he seeks to pre-empt some potential objections to his argument, and in which he makes clear that he has no time for certain kinds of diviner. Quaestus, profit, makes its second appearance in the dialogue, after the passage on haruspicy discussed above. Those who practise divination for money (qui quaestus causa hariolentur) are included in a list of seers who deserve condemnation, such as those who practise cleromancy or those who seek to evoke the souls of the dead.14 Such exclusionary rhetoric does not apply only to lesser forms of divination, but is also sensitive to the venues in which the craft is practised: whether obscure rural contexts, such as the land of Marsi, or the street corners where some haruspices display their lore, or in the vicinity of the circus where the astrologers are busy. Again, the polemic is not primarily about the social status of these diviners, but their lack of a respectable body of knowledge and expertise: they lack scientia and ars. These seers’ desire for remuneration for their services turns their lore, which has epistemologically doubtful foundations, into a morally questionable practice, and a potential threat.15 Quintus’ critique is then compounded by a quotation from the great Italian poet Ennius, who launched an abrasive attack on the charlatans who engage in divinatory activity for a fee. Unlike them, Quintus can build on a respectable intellectual tradition, and his quote from Ennius emphatically asserts this. Ennius’ passage – probably deriving from his tragedy Telamo – makes a pointed reference to the relationship between divination and money, which gives 80 Federico Santangelo further depth to the picture.16 The shameless diviners whom the poet (or his character) chastises pretend to offer guidance, but are both incompetent and driven by pressing material concerns (egestas). They promise wealth to those who seek their advice, and in the same breath they ask for money. It is interesting to see that Ennius uses the word drachuma, as the Greek setting of his play no doubt required; his critique may also be applied to a Greek context, but it should be referred in the first place to that of mid-Republican Rome.17 The diviners who were at work in Scipio Aemilianus’ camp advised their questioners on how they would fare in the war. Ennius claims that many of those who consulted diviners in a private capacity were driven by questions and uncertainties on economic and financial matters. Much of the background, or indeed the psychology, of such consultations eludes us, and we can only speculate on the terms of the questioners’ curiosity: for example, whether they were keen to gain insight into the overall trajectory of their fortune or into specific choices. If we are to believe the sceptical voice of the dialogue, the character of M. Tullius Cicero himself, this claim should not be accepted uncritically. Without referring to Ennius, Marcus notes that no straight- thinking individual resorts to divination – and especially to haruspicy – to seek advice on how to conduct themselves in the dispatch of their duties, whether to one’s family or to one’s community, or on how to use their money (Cic. Div. 2.11). Marcus’ polemical edge should warn (here and elsewhere) against taking his testimony at face value, on this and many other points. What matters for our purposes is the argument he puts forward, in which a clear demarcation is identified between economic decisions and the insights that divination may provide. Elsewhere in his discussion, he takes the argument to its full conclusion, and argues that there is no connection between the sudden and fortuitous increase of one’s fortune and what divination may be able to reveal (Cic. Div. 2.33). While a chance overlap between the results of the reading of the entrails and one’s sudden and unexpected gain might occur, that is not the symptom of a link between the developments in the natural world and the fate of an individual, or indeed between those developments and the ability of an individual to increase their fortune. One may fortuitously discover a treasure without there being any link with nature and its cycle. Marcus’ argument especially targets the foundations of extispicy, not just of Etruscan haruspicy: the claim that the entrails of an animal might mirror the situation in heaven is dismissed as preposterous. Significantly, the line of attack chosen here targets private consultations. Those who believe that divine signs may provide instructions on a potential small gain (quaesticulus) are singled out in Marcus’ polemic: another indication of the relevance of divination to the economic sphere. Since Marcus is so fundamentally hostile to divination, the argument deployed by Quintus on the tension between divination and profit plays a far less important role in his discussion. Although Marcus recognizes the political significance of some forms of divination, he is not interested in defending them from the pressure of incompetent and misguided practitioners. The link between divination and the possibility of unfair profit is evoked on only one occasion in his discussion, and is turned into a polemical weapon: his discussion of cleromancy – lot divination – is unreservedly critical, and one of the opening arguments Divination and profit in the Roman world 81 used against it is that it is intended to generate gain, as well as fostering superstition and leading people into error, and can only be inspired by a fraudulent intention on the part of the diviner.18 The fact that cleromancy has no status in Roman public divination, despite the high distinction of the cleromantic sanctuary of Fortuna Primigenia at Praeneste, makes his task easier, and enables him to criticize in equal measure the institutions where cleromancy is practised and the diviners who offer it in private settings – the sortilegi about whom Quintus also expresses reservations.19 If one takes the specific standpoint on the link between divination and money, Marcus could be accused of not engaging with Quintus’ argument as fully as he could have done. Quintus does not claim that divination can enable one to make a profit or increase one’s wealth, and indeed he is at pains to separate the domain of money from that of divination and prophecy. Moreover, he states unequivo- cally that there is a discernible difference between divination and other forms of prediction, among which he singles out the ability of some individuals to make the most of their intellectual mind and raise themselves to understand divine matters. Far from being inspired by a divine impulse, these individuals base their predictions on careful observation of natural phenomena. This craft, which is not merely about theoretical knowledge, enables them to make eminently practical decisions, and indeed to increase their wealth: the signal example being that of Thales of Miletus, who proved to his sceptical contemporaries that even philosophers can make money and made a major investment in olive trees before they started flourishing (Cic. Div. 1.111–12). His natural observation enabled him to make an advance judgement of the potential of those plants which no one else was in a position to make. Quintus also notes that Thales was the first among his contemporaries to predict a solar eclipse – and makes no reference to the fact that eclipses were often regarded as prodigies that required ritual action, not as natural phenomena. Marcus does not take up this example in his riposte. Thales is briefly mentioned, along with Anaxagoras, as an example of intellectually aware individuals who would have easily seen the natural and easily explainable causes behind certain alleged prodigies (Cic. Div. 2.58). Much of Roman public religion hinges on the premise that involvement in ritual is a collective process, which entails a considerable degree of collegiality. Hardly any major religious action can be carried out in isolation, away from the scrutiny of a cluster of peers or of an interested audience. This is very clear under the Republic, but remains an important feature in the imperial period too. It is unclear how the haruspices were organized in the Roman res publica: they were routinely consulted by the Senate and the magistrates, but there is no evidence for their organization in one body in the whole of the Republican period. While there is evidence that the Roman government had made arrangements to control their training, there is no reason to believe that there was a set pool or a list of haruspices from whom the experts consulted by the government were drawn. It is only with Claudius that we see an initiative, led by the emperor himself, to create an ordo of sixty haruspices: in 47 ce he put forward a proposal to the Senate, which was then enforced by the pontiffs.20 The princeps was, as is well known, 82 Federico Santangelo a connaisseur of things Etruscan, and his interest certainly played a part in the decision. On the other hand, there was a longer-term need to define the status of the haruspices, especially vis-à-vis the power of the emperor and the position of the Senate. Even after Claudius, however, we are none the wiser on the form of remuneration they received – if any. There appears to be a substantial change under the Severans, when a large body of evidence for the presence of haruspices in the army emerges, and points to the stable and diffuse presence of haruspices in the military context, down to the legions. The set of concerns that had led Scipio Aemilianus to expel the diviners from his camp received a very different response in that case. The diviners were embedded in the ranks, and their training and organization were tightly managed by the government. For the first time, there is evidence to suggest that at least some of them received payment: L. Fonteius Flavianus proudly displays the title of haruspex Augg(ustorum) ducenarius.21 Concerns regarding the profit that may be drawn from divination reflect a wider concern about the risks that divination entails when undertaken outside an institutional context and without the checks and balances that the membership of a college or a fraternity may bring with it. There is a sizeable body of evidence for these preoccupations, well beyond the philosophical discussion of Cicero. Livy, for instance, on a few occasions singles out the sacrificuli uatesque who were in action in Rome, right in the Forum, and offered their expertise to the gullible populace; during the Hannibalic War, their main target was the rural populace that had come to town to escape the devastation and misery brought about by the fighting in the countryside. On two occasions, he makes clear that quaestus, profit, was the chief aim of these individuals, and that their activity is a craft that involves error. In 213 their presence first became a matter of muted concern for the boni, and then prompted official action on the Senate’s part.22 The terminology used to refer to these diviners is rather unspecific: sacrificuli were no doubt engaged in extispicy, while the uates surely practised inspired divination.23 The differentiation between these two groups is not fully apparent, and it is not clear whether they would have been happy to be referred to with the labels that Livy uses. Their presence in Rome should not be associated exclusively with moments of crisis: they should be understood as part of a landscape in which different and – at least at times – competing sources of divinatory and prophetic knowledge were in operation. In some moments of grave crisis some of those voices had to be muted, and government action was required. Drawing attention to the fact that these diviners drew material gain from their lore, unlike the religious experts who operated as part of the cultus publicus, was an argument that could credibly be used to undermine their authority. Livy, as we have just seen, resorted to it. Such diversity becomes even more striking, and more significant, if one turns away from Rome and looks at the complexity of the empire and the variety of its provincial and local contexts, and if one goes beyond the literary evidence and brings into the discussion some documentary material. As pointed out above, Roman divination should not be understood simply as the range of divinatory practices that were carried out in the city of Rome and were part of the remit of the government. The range of options available becomes noteworthy as soon as Divination and profit in the Roman world 83 one starts listing certain relevant instances, which expose us to a series of questions and preoccupations that were shared by those who resorted to divination, and drastically change the outlook from the elite standpoint of the De Divinatione. Two examples, both from the Greek-speaking provinces of the empire, are especially instructive, and have received recent discussions in Toner’s book on Roman popular culture and Beard’s study of risk in Roman culture.24 One of the most peculiar sources on ancient divination is the Greek text known as the Sortes Astrampsychi, which survives in two different editions, and purports to derive from the work of the Egyptian magician Astrampsychus, but is in fact a practitioner’s handbook, and should probably be dated to the second century ce.25 This is just the sort of material that would have prompted the concerns of those who placed public religion at the centre of the debate, and who would no doubt have regarded with hostility written material that eluded public scrutiny and mediation; it is conceivable, however, that on most occasions it will have been tolerated. The text is opened by a series of ninety-two questions, which offer an invaluable insight into the concerns that prompted those who resorted to the advice of a diviner; they reflect the viewpoints of a diverse set of people, from reasonably well-off businessmen to slaves. A series of answers, which are linked back to the questions through an elaborate system of numerical correspondences, also survives.26 While this could not be regarded as a safe statistical sample, it is certainly to be taken seriously. Within that range, the weight of the questions that pertain to the economic domain is considerable: I have counted thirty-one. A strong emphasis in the questions impinges on debt and investment, as well as anxiety over the outcome of business transactions and questions over inheritance matters.27 Remarkably, only a small range of questions (seven, on my calculation) explicitly ask the oracle for instructions on how to steer certain economic decisions, such as ‘If I lend money will I not lose it?’ and ‘Will I profit from the undertaking?’28 Most questions appear to revolve around scenarios that are fairly distant in the future, or upon which no decisions have to be made imminently. Economic matters must have also featured prominently in the set of questions that were asked of the so-called ‘dice oracles’ that are attested in a number of cities of south-western Anatolia in the second century ce, and of which several epigraphical copies are extant on large stone pillars, typically in prominent public locations: the best preserved texts survive from Kremna, Perge, and Termessos, among which there is a considerable degree of overlap.29 Given the nature of these divinatory consultations, this material does not record the questions that were asked of the oracle, but yields a wide variety of answers, each corresponding to a different numerical combination. The questioner would throw five four- sided sheep knucklebones (astragaloi); the sides were assigned the values 1, 3, 4 and 6, and the resulting outcomes (e.g. 1 + 3 + 3 + 3 + 4) were added up. In this instance, since several combinations could give a result of fourteen, the exact one would have to be looked up among the fifty-six possible responses that are listed on the pillar. In this case, the answer is: ‘You kick against the goad, you struggle against the waves, you search for a fish in the sea: do not hasten to do business. 84 Federico Santangelo It does not help you to force the gods at the wrong time’.30 At least twenty-six answers include clear instructions on how to conduct oneself in a business transaction or on other economic choices.31 The advice they provide is usually firm and clear. Although establishing a tight link between the percentage of unfavourable responses and the margins of risk of commercial activity seems overly determin- istic, there is no doubt that these sets of responses are strongly preoccupied with the economic sphere and its challenges:32 not just with profit, but also with the accidents of survival and the hazards entailed by trade. The responses set out to offer firm guidance on how to control the future by drastically curtailing the range of options available to the questioner.33 This highly selective discussion opened with the claim that the title of this volume – Prophets and Profits – is not just a matter of a pleasing alliteration, and has been trying to take that callida iunctura seriously.34 Much of the best work that has been done on Greek and Roman religion over the last half-century or so has been driven by the intention to bring back to the ancients their alterity, their otherness, and to explore the terms of what separates them from us, rather than pursuing the tempting, and often facile, route of analogy. Yet exploring the tension between these two different, but closely related areas – prophecy and profit – in antiquity can provide a formidable opportunity to reflect critically on our own time, and on our own role and responsibility in a context within which we go about teaching and researching the past. Exploring the boundaries and the areas of overlap between prophecy and profit is ultimately about asking questions of agency, and debating human freedom.

Notes 1 I am very grateful to my colleagues at UNISA for their invitation to Pretoria and to the participants in the Prophets and Profits conference for their reactions to a preliminary version of this chapter. I should also like to thank the members of a seminar audience at Rome La Sapienza for their comments and questions on aspects of the argument presented here. The feedback of Katherine East, Micaela Langellotti, Marco Maiuro, Chris Mowat and two anonymous referees on later drafts has been crucial. John Thornton has made an insightful suggestion on an important point. Elio Lo Cascio and Eleonora Tagliaferro have offered invaluable advice and encouragement at various stages. 2 App. Iber. 85: ἐλθὼν δὲ ἐμπόρους τε πάντας ἐξήλαυνε καὶ ἑταίρας καὶ μάντεις καὶ θύτας, οἷς διὰ τὰς δυσπραξίας οἱ στρατιῶται περιδεεῖς γεγονότες ἐχρῶντο συνεχῶς· ἔς τε τὸ μέλλον ἀπεῖπε μηδὲν ἐσφέρεσθαι τῶν περισσῶν, μηδὲ ἱερεῖον ἐς μαντείαν πεποιημένον (‘When he arrived he expelled all traders and prostitutes; also the seers and diviners, whom the soldiers were continually consulting because they were demoralized by defeat. For the future he forbade the bringing in of anything unnecessary, or any victims for purposes of divination’). 3 Bendlin (2000) is its most effective proponent for the Republican period. For a vigorous critique see Scheid (2013) 43–64; Scheid (2016) 17–31, 149–52. See also Beck (2006). 4 Dion. Hal. Ant. Rom. 4.62; Plin. Nat. 13.27.88; Lact. Inst. 1.6.10–11; Serv. Aen. 6.72. 5 Cato Agr. 5: ‘Let him [sc. the farm steward] not consult any haruspex, augur, seer, or Chaldaean [astrologer]’. 6 Horster (2007) 338. 7 North (1990) 58–60. 8 Torelli (2011: 142–4) gives references to previous bibliography. Divination and profit in the Roman world 85 9 Cf. Colum. 1.8.6: haruspices sagasque, quae utraque genera uana superstitione rudis animos ad impensas ac deinceps ad flagitia compellunt, ne admiserit (‘He must not admit haruspices and witches, two sorts of people who incite ignorant minds through false superstition to spending and then to shameful practices’). Translation by H.B. Ash, modified. 10 Cic. Div. 1.92: Etruria autem de caelo tacta scientissume animaduertit, eademque interpretatur quid quibusque ostendatur monstris atque portentis. quocirca bene apud maiores nostros senatus, tum cum florebat imperium, decreuit ut de principum filiis x singulis Etruriae populis in disciplinam traderentur, ne ars tanta propter tenuitatem hominum a religionis auctoritate abduceretur ad mercedem atque quaestum. (‘Etruria has the greatest knowledge of things struck by lightning and also interprets what is signified by each prodigy and portent. For this reason, in the time of our forebears, the Senate, at a time when our empire was thriving, decreed that of the sons of leading citizens groups of ten should be handed over to the individual Etruscan peoples to be instructed in the discipline so that an art of such great importance should not, because of a lack of manpower, lose its religious authority to become an object of commerce and profit’). Translation by D. Wardle. 11 I am following the text printed by W. Ax, R. Giomini and D. Wardle, who amended the manuscript’s sex to X. See also Val. Max. 1.1.1: decem principum filii senatus consulto singulis Etruriae populis percipiendae sacrorum disciplinae gratia traderentur (‘the sons of ten leading citizens were entrusted to individual peoples of Etruria in order to learn the lore of the sacred rites’). A. S. Pease and S. Timpanaro printed ex; the resulting text should be translated as ‘decreed that ten sons of the leading citizens of each of the Etruscan peoples should be handed over’). For a full discussion of the textual and historical issues raised by this passage see Timpanaro (1988) 301; Wardle (2006) 326. 12 Santangelo (2013) 84–98. 13 MacBain (1982); Rosenberger (2005); Santangelo (2013) 85–7, 98–114. 14 Cic. Div. 1.132. For a close reading of this text on the tension between divination and monetary gain see Nice (2001) esp. 163–6; Nice (2003) 412–13. See also Wardle (2006) 422. 15 Isnardi 1955 remains essential reading on the close integration between technē and ēthos in Hellenistic culture, and its far-reaching political implications (esp. page 107 on the link between the notion of ‘êthos deteriore’ and that of ochlos). 16 A similar point was made in Accius’ Astyanax 169 R.: nil credo auguribus, qui auris uerbis diuitant/alienas, suas ut auro locupletent domos (I do not trust at all the augurs, who enrich the ears of others with words, so that they may fill their own houses with gold’). See the discussion of both fragments in Rüpke (2012) 57–8. See also the list of female diviners in Plaut. Mil. 692–4, which makes reference to the remuneration they receive: da quod dem quinquatribus/ praecantrici, coniectrici, hariolae atque haruspi- cae;/ flagitiumst, si nil mittetur quae supercilio spicit (‘give me something to give on the Quinquatria to the spellcaster, the dream interpreter, the diviner and the haruspica; and it would be calamitous if nothing were handed to the woman who reads the eye- brows’). 17 Allegations of greed against diviners are common in Greek literature, and should not be read as evidence for scepticism about divination. See Flower (2008) 135, 146–7. 18 Cic. Div. 2.85–6, esp. 85: tota res est inuenta fallaciis aut ad quaestum aut ad superstitionem aut ad errorem (‘The whole scheme was fraudulently contrived for the sake of profit, or as a means of encouraging superstition and error’). 19 Grottanelli (2005) 131–8; Klingshirn (2006); Santangelo (2013) 73–83. 20 Tac. Ann. 11.15.1; cf. Haack (2003) 85–98. On this critical phase in the history of haruspicy, within the wider context of a discussion of the character of Umbricius in Juv. Sat. 3, see also Nice (2003) 404–6. 21. CIL 6.2161, from Rome: ducenarius is abbreviated as CC. See Haack (2003) 137–76, esp. 144–6; on ducenarius see OLD s.u.: ‘paid a salary of 200,000 sesterces’. 86 Federico Santangelo

22 Livy (25.1.6–8. 39.16.8) suggests that the events of 213 bc were not an isolated occurrence: the expulsion of uates and sacrificuli was deemed necessary on other occasions. I do not think we can be certain that Livy got ‘carried away by his own rhetoric’, as suggested by Briscoe (2008) 278. 23 It is tempting to see a close correspondence between Livy’s uates/sacrificuli and μάντεις καὶ θύτες in App. Iber. 85 (see above n. 1). 24 Toner (2009) 44–52; Beard (2013). 25 The reference edition of this text is Naether (2010). See also Browne (1983) for a shorter version, and Stewart (2001) a longer version. For an English translation of the latter and a useful introductory discussion see Hansen (1998) 291–324. See also Beard (2013) 101–4. 26 A brief summary is in order. The questioner would pick one of the ninety-two questions (which are numbered from twelve to 103); he would be asked to think of a number between one and ten, which the diviner would add to the number of the question. That figure would then be converted into a different one on the basis of a table of correspondences. The new number would direct the diviner to a numbered group of ten answers; the final step involves looking up the answer that corresponds to the number the questioner was asked to come up with at the beginning of the consultation. See Hansen (1998) 288; Beard (2013) 102. Beard rightly glosses that ‘if it seems baffling, that is exactly what it was meant to seem’. 27 Toner (2009) 48–9. 28 See questions 58 (νη εἰ δανείσας οὐκ ἀπολέσω) and 81 (πα εἰ κερδαίνω ἀπὸ τοῦ πράγματος). See also the three typical questions singled out in Plut. De Pyth or. 28.408c: ‘εἰ γαμητέον ‘εἰ πλευστέον ‘εἰ δανειστέον’ (‘Should I marry, should I sail, should I lend money?’). Parker (2016: 82–3) discusses the evidence from Dodona for questions on loans and debts, and notes that issues of financial probity were often in the background of such consultations. 29 Graf (2005a: 82–4) lists fourteen texts in total. The article as a whole is the best discussion of this material. See also Beard (2013) 99–101. Greaves (2012: 183–91) offers an excellent overview of the use of astragaloi in cleromantic contexts, with a strong emphasis on the archaic and classical periods; and makes an attractive case for their use at Branchidai-Didyma (191–203). 30 See the text in Horsley and Mitchell (2000) no. 5, from Kremna: λακτίζεις πρὸς κέντρα, πρὸς ἀντία κύματα μοχθεῖς | ἰχθὺν ἐν πελάγει ζητεῖς, μὴ σπεῦδ’ ἐπὶ πρᾶξιν· | [οὔσοι χρήσι]μ̣όν ἐστι θ̣[εοὺς βιάσασ]θ̣α̣ι̣ ἀκαίρως. Translation from Graf (2005a) 87. 31 The interplay between numbers, letters and words has a role of wider significance in Greek divination, especially in the Sibylline tradition: see the valuable discussion in Tagliaferro (2000) 149–52. 32 Cf. Toner (2009) 48. Parker (2016: 81–2) discusses comparable queries from Dodona as evidence for ancient attitudes to mobility. 33 On the limits of the notion of ‘control’ in oracular responses, see Eidinow (2013: 31–2), which is now essential reading on the construction of the self in oracular consultations in the Greek world and on the degree of interdependence with fellow human beings and the divine that presupposes them. Greaves (2012: 196) issues an important caveat on the potential of cleromancy to produce long ‘narrative’ responses, which is often under- estimated in modern discussions: this general assumption seems to be corroborated by the material of Kremna, Perge and Termessos. 34 Herman (2006) sets out to do that in a provocative and rewarding study of religious education in contemporary South Africa: a valuable read for anyone who cares about the link between access to knowledge and social justice. 7 Profiting from prophecy Q. Marcius Rex and the construction of the Aqua Marcia1

Alex Nice

Prelude: the Aqua Marcia

In ad 60 the Emperor Nero’s desire for a luxurious lifestyle brought infamy and danger to the princeps because he swam in the source of the Aqua Marcia. He had polluted the sanctity of the sacred waters. Nero’s subsequent sickness confirmed the ira deum (Tac. Ann. 14.22.6). Tacitus’ comments are not mere rhetorical exaggeration. Elsewhere, the terminus of the River Clitumnus was marked by a temple to the god and by several other shrines. No swimming was permitted in its sacred area (Pliny, Ep. 8.8). According to Seneca, the sources of great rivers, the springs of hot waters and pools of opacitas or immense depth were ven- erated, cultivated and sanctified (Sen. Ep. 41.3, cf. Front. Aq. 4.3.1). Frontinus refers to the outstanding sanctity (sanctitas) and worship (colitur) of the fontes Camenae, Apollinaris and Iuturnae because of their healing qualities (Front. Aq. 4.2). Servius (Ad Aen. 7.84) is emphatic: nullus … fons non sacer (‘there was no spring that was not sacred’). The source of the Aqua Marcia stood at the thirty-eighth milestone on the Valerian Way.2 For most of its course, it ran underground on the right bank of the River Anio. About 10 km from Rome, it rose on substructures and arches, entering the city at Porta Maggiore. From a holding tank on the Viminal Hill, its waters were distributed to various sites around Rome, including the Capitoline. Part of the water was deposited into the so-called rivus Herculaneus, which ran past the Gardens of Pallas and the Caelian Hill, before ending at the Porta Capena (Front. Aq. 19). The Porta Capena was also an important junction point for religion and divination. Close by were the watery shrines of Egeria and the Camenae.3 Their fountain was praised by Vitruvius, alongside the Marcian aqueduct, for the purity of its water (Vitr. De Arch. 8.3.1). Here the Vestal Virgins drew their sacred water for cleansing and Numa was rumoured to have met the water nymph Egeria. Through hydromantia he learnt the secrets contained in the pontifical books.4 The association with divination is further emphasized through the most celebrated of the Camenae, Carmenta (or Carmentis), whose festival fell in January each year.5 She was styled by the Augustan poets as felix vates and vates fatidica, foretelling the future of Aeneas’ sons (Ov. Fast. 1.585–6; Verg. Aen. 8.340). With quasi-religious awe, the ancient sources exalted the source of the Marcia for its splendor, 88 Alex Nice the waters for their clarity, frigor (‘coldness’) and salubritas (‘healthiness’), and its importance in supplying pure drinking water to the city (Front. Aq. 89; 91; 92; Vitr. Arch. 8.3.1; Mart. 6.42.18–21; Pliny, NH 31.41). The waters were, Pliny said (NH 31.41), ‘a gift from the gods’.

The construction of the Aqua Marcia

The cultivation of this gift of the gods was made possible in 144 bc. In that year the Roman Senate instructed the praetor urbanus, Q. Marcius Rex, to reclaim and repair the conduits of the Appian and Anian aqueducts which had sprung leaks because of their old age and from which citizens were illegally diverting the water supply.6 Due to an incrementum Urbis (‘an enlargement [of the population] of the city’), Marcius was further instructed to create wider access to water.7 In addition to the repair work, during the next four years (144–140 bc) he brought in a third and more salubrious supply, the so-called Aqua Marcia. Fenestella takes up the story:

legimus apud Fenestellam, in haec opera Marcio decretum sestertium milies octingenties, et quoniam ad consummandum negotium non sufficiebat spa- tium praeturae, in annum alterum est prorogatum. eo tempore decemviri, dum aliis ex causis libros Sibyllinos inspiciunt, invenisse dicuntur, non esse fas aquam Marciam seu potius Anionem – de hoc enim constantius traditur – in Capitolium perduci, deque ea re in senatu M. Lepido pro collegio verba faciente actum Appio Claudio Q. Caecilio consulibus, eandemque post annum tertium a Lucio Lentulo retractatam C. Laelio Q. Servilio consulibus, sed utroque tempore vicisse gratiam Marci Regis: atque ita in Capitolium esse aquam perductam.8

We read in Fenestella that 180,000,000 sesterces were decreed to Marcius for this work, and since there was not sufficient time during his praetorship to complete the task, that it [the praetorship] was prorogued for another year. At that time, the decemviri, while they were inspecting the Sibylline books for other reasons, are said to have discovered that it was not right for the Aqua Marcia or rather the Anio – for regarding this, there is more certainty – to be brought to the Capitol. And when Appius Claudius and Quintus Caecilius were consuls, there was a debate about this matter in the Senate with Marcus Lepidus speaking on behalf of the College, and likewise three years later, when Gaius Laelius and Quintus Servilius were consuls, the matter was brought up again by Lucius Lentulus, but on both occasions the influence [gratia] of Marcius Rex had won the day, and in this way the water was brought to the Capitol.

The costs involved were immense: 180,000,000 sesterces and when Marcius’ praetorship was prorogued, the decemviri sacris faciundis claimed to have found objections to the building of the aqueduct: dum aliis ex causis libros Sibyllinos inspiciunt, invenisse dicuntur, non esse fas aquam Marciam seu potius Anionem – de Q. Marcius Rex 89 hoc enim constantius traditur – in Capitolium perduci. Perhaps this a precise rec- ollection of the oracle which caused them concern and a reference to the river from which the Marcia drew its water.9 But there is no mention in Livy’s capri- cious excerptor, Julius Obsequens, neither in 143 nor in 140. Nonetheless, his text preserves the aliis ex causis:10

[Amiterni puer tribus pedibus natus. †Caurae sanguinis rivi e terra fluxerunt.] cum a Salassis illata clades esset Romanis, decemviri pronuntiaverunt se invenisse in Sibyllinis, quotiens bellum Gallis illaturi essent sacrificari in eorum finibus oportere. (Obs. 21)

At Amiternum, a boy was born with three feet. At †Caura, rivers of blood flowed from the earth. When the Salassians inflicted defeat upon the Romans, the decemviri pronounced that they had found in the Sibylline Books that, whenever they were going to wage war against the Gauls, there ought to be sacrifices within their borders.

The perfect infinitive invenisse for a discovery in the Sibylline Books is used in this sense only here and at Frontinus.11 Coupled with the future infinitive, it marks a rare irruption of the prophetic into the ordinarily prescriptive responses of the Sibylline Books.12 The story is confused but the essential detail is confirmed by Livy’s Oxyrhynchus epitome: Marcius’ aqueduct was constructed contra Sibyllae carmina and the waters of the River Anio brought to the Capitol.13 Twice the matter was debated in the Senate and twice the gratia of Marcius Rex prevailed.14

The gratia of Q. Marcius Rex Frontinus (or, perhaps, Fenestella) is judicious in his choice of words: gratia, not auctoritas or potestas, carried the day.15 Although gratia could be used almost synonymously with those and other political terms, it denoted a particular type of political credit which could influence the outcome of elections or votes.16 Gratia, said Cicero, was the memory and reciprocation of duties (officia), honour (honor) and friendships (amicitiae);17 along with beneficence, gratia was necessary for concordia (political concord).18 Gratia was an important feature of the recipro- cal nature of Roman social relations: ‘Through gratia a person was able to create and maintain social networks, that – apart from being instrumental in achieving political or other goals – had the potential of enhancing one’s social status’.19 It was, therefore, a form of influence allied to notions of fides, amicitia and patronage. Gratia persuaded the Senate to grant triumphs to L. Furius Purpurio (200 bc; RE 86) and Q. Fulvius Flaccus (179 bc; RE 61; Liv. 31.48–9.4; 40.59.1). In the former case, objections to the celebration of the triumph were raised on grounds of propriety: in using an army entrusted to another commander, Furius seemed to have contravened the chain of command and, in his eagerness to celebrate a triumph, had left his province prematurely and contrary to precedent. The consulars 90 Alex Nice in the Senate argued that Furius should have waited for the consul before engaging the enemy; the opposition that only the praetor’s achievements should be considered, and whether the victory had been achieved under his own auspicia. Ultimately, the vote took place in the consul’s absence and the praetor’s gratia won the day.20 In another instance, the gratia of the absent praetors was not enough to persuade the Senate that they should bring their armies back to Rome. Livy records that the senatorial debate was keenly contested between the friends of the new praetors and those of the old (Liv. 39.38.10–11). Such examples suggest that, in senatorial debates at least, gratia was vital where personal prestige and opportunities to advance one’s own dignitas were at stake. It was a test of character, popularity, powers of persuasion and of achievements. Such gratia could be accumulated through the generations and exist within families. For example, M. Popillius avoided prosecution thanks to the gratia of his brother, the consul of 172 bc, despite his absence from Rome (Liv. 42.22.7), in 80 bc Metellus Pius canvassed for Q. Calidius because as tribune he had passed a law restoring Metellus’ father to citizenship (Val. Max. 5.2.7); and, through his own efforts, Cluentius was able to preserve his ancestral gratia (Cic. Pro Cluent. 196). Gratia was, therefore, allied to a politician’s social capital. It was associated with the ‘accumulated symbolic capital’ derived from his maiores. Ambitious politicians took every opportunity to publicize and to emphasize that ‘symbolic capital’.21 There was a ‘constant renewal’ of familial achievements when a magis- trate entered office (Cic. De Leg. 2.1; Plut. Aem. 11.1), to bolster their arguments in public contiones (Livy, 45.24.12),22 or at funerals (Polyb. 6. 53.1–54.1–2).23 The ways in which an orator constructed a speech in a senatorial meeting in the curia or, for a more religious meeting on the Capitol, could lean on historical antecedent and an appeal to the maiores.24 More generally, the topographical surroundings of the rostra, the Capitol, and the Curia were a locus for an array of monuments tes- tifying to the res gestae of individuals and their families, including those specific to the gens Marcia, all of which could be deployed as oratorical weapons.25 One such example was the statue of Q. Marcius Tremulus which was ‘embedded in the complex symbolic web of the politico-religious topography of the urbs Roma’.26 On one interpretation, the Philippi, Figuli and the Reges were his descendants.27 It was probably in 145 bc, the year before the new water course was proposed, that the tribune C. Licinius Crassus had first led the people from the comitium into the open area of the forum in front of the temple of Castor: evidence perhaps of the incrementum urbis.28 There, during his tenure as praetor urbanus, the public contiones would have provided Q. Marcius Rex with plentiful opportunities to remind the Roman public of his illustrious ancestral heritage.29 For, historically, in a world where much was predicated on being ‘first, best, greatest’, the Marcii were a stand-out clan.30 Prior to 144 bc they could enumerate the first plebeian dictator, censor, pontifex and augur, in addition to thirteen consulships, six censorships and a triumph.31 In common with other leading gentes, they had developed a kind of ‘corporate identity’ based around mythological descent and ‘Republican’ merit.32 The origins of the gens Marcia were traced to Numa’s daughter Pompilia, who had married the first pontifex maximus, Numa Marcius. Their progeny was the Q. Marcius Rex 91 fourth king of Rome, Ancus Marcius.33 Special relationships with the Tiber and the port at Ostia were claimed for Ancus Marcius.34 In one tradition, he had built the Aqua Marcia (Pliny, NH 31.41). Furthermore, the regal lineage of Q. Marcius Rex emphasized the development of the pontificate. The pontiffs were entrusted with oversight of the sacra publica et privata, the maintenance of ancestral rites and the adoption of foreign practices. Additionally, they were to teach which prodigies were to be recognized and expiated (idem pontifex edoceret, quaeque prodigia fulminibus alioue quo uisu missa susciperentur atque curarentur).35 It was Numa Marcius who had first received those guidelines from Numa and, as the first act of his reign, Ancus Marcius had ordered the pontifex maximus to copy out the sacra publica in Numa’s commentaries on to a white board, publicly displayed (Liv. 1.32.2). The stories probably derived from the reputation of the Marcii as Republican priests. C. Marcius Rutilus Censorinus (cos. 310; RE 98) was not only the first Roman to hold both pontificate and augurate simultaneously but one of the original five plebeian holders of those priesthoods.36 Moreover the cognomen Rex, apart from an obvious allusion to their regal ancestry, seems to derive from M. Marcius (RE 20), the first plebeian rex sacrorum.37 He had lived in the Regia, Numa’s traditional home, and, in addition to certain sacrificial duties, he had responsibility for the calendar.38 Within living memory – and the most recent memories were the most important – M. Marcius Ralla had been a duovir aedis dedicandae for the temple of Fortuna Primigenia on the Quirinal and also dedicated a temple of Veiovis, a god sometimes associated with Apollo, on the Capitol.39 In 180 bc, Q. Marcius Philippus (cos. 186, 169; RE 79), who had overseen the prosecution of the Bacchanalia affair, was co-opted to the decemviri.40 Not only were the Marcii famous priests, they were also famous prophets. In 213 bc, at the same time as Marcius’ ancestor was rex sacrorum, ‘those Marcian brothers, men of noble birth’ or perhaps only one Cn. Marcius, a vates inlustris had warned the Romans to beware the River Canna (the Aufidus).41 The warning was taken seriously. As a result, the oracles were included in the Sibylline collection and the ludi Apollinares were instituted to preserve the favour of Apollo and to drive the Carthaginians from Italy (Serv. Aen. 6.70, 72). Their jurisdiction fell annually every 6 July to the praetor urbanus: the ludi Apollinares of 144 bc must have been an ideal opportunity for Q. Marcius Rex to emphasize his familial expertise in religion and divination. In the ensuing conflict between the Sibylline Books and the Aqua Marcia, claims to priestly expertise and a privileged relationship with Apollo and the libri Sibyllini were powerful arguments in support of Marcius’ case. This sense of ‘Republican’ merit is also apparent in the way in which the Marcii appear to have allied themselves with plebeian or popularis causes, not least the Aqua Marcia. Not only had they paved the way for the plebeians to share in the governance of the Roman state but they had also granted them greater protections. C. Marcius Rutilus, four times consul and the first plebeian dictator and censor, may have carried the lex Marcia adversus faeneratores.42 Certainly, Livy recounts that the consuls made the discharge of debts a state matter (Gaius, Inst. 4.23; Liv. 7.21.5). Rutilus’ popularity among the plebeians facilitated his election 92 Alex Nice to the censorship and opened the way for the plebeians to have a greater share in the governance of the Republic.43 Other incidental protections had also been acquired through the censorship. For, in addition to conducting the census and revising the rolls of the senatorial and equestrian orders, the censors organized the budget of the Roman state, regulated property taxes, managed the vectigalia, contracted new public works and ensured the maintenance of older buildings.44 The value of this office to the plebeians is implied in Livy’s account of Marcius Rutilus’ election to that office and in his descendant’s self-imposed regulation against anyone holding the office more than once.45 Members of the Marcii had also passed laws granting the people the right to bear arms and relief from the stipendium (war tax).46 The social, political and financial importance of the Aqua Marcia make it likely that the senatorial discussions of 143 and 140 took place in the temple of Jupiter on the Capitoline.47 Dedicated in the very first year of the Republic, it ritually symbolized the shaking off of the burden of kingship and libertas for the people (Liv. 1.55.2–4). There, alongside the prophecies of Apollo and Marcius, hung the ‘Marcian shield’, a silver aegis emblazoned with the image of Hasdrubal, commemorating the victory of L. Marcius over the Carthaginians in 212 bc (Liv. 25.39.12–13; 16–17). The shield could be interpreted as a reminder of a ‘morally uplifting Roman past’.48 It was also a reminder that the Marcii had the favour of the gods on their side. Prior to the battle, a ring of fire had enveloped Marcius’ head, a miraculum eius verae gloriae.49 The oratorical value of such symbols is well-illustrated by the examples of Marcus and Titus Manlius. In 385 bc, the trial of Marcus Manlius eventually had to be moved to the Peteline wood to avoid his effective invocation of the sacred hill while decrying the abuses of the patricians and the feneratores, and the concomitant misery of the plebeians (Liv. 6.11–20). In 340 bc, the consul T. Manlius had effectively appealed to the statue of Jupiter to dismiss the claims of Annius and the Latins to the consulship and membership of the Roman Senate, a gesture later repeated by Cicero in 63 bc in his denunciation of Catiline.50 The Capitoline, moreover, was associated with the Marcii’s own mythological past as the seat of the Sabine kings. The sanctity of the area was further reinforced through the auguraculum on the arx which represented one of the Marcii’s memorable early priestly offices, Marcius Ralla’s temple to Vediovis and a shrine in honour of Concordia. Within Marcius’ lifetime, in 164 bc, the censor Q. Marcius Philippus had erected not only a sundial but also a statue of Concordia in a public location.51 The principle of concordia (political harmony) at Rome could be traced back to Numa and his legendary association with Pythagoras. Storchi Marino has ingeniously argued that this legend, which developed in the fourth century bc, is attributable to the Marcii.52 Mythology and historical record coincide in one last Marcian monument: the statue of Marsyas in the forum.53 Coinage of L. Marcius Censorinus (moneyer 82 bc) depicted the silenus walking, his right arm raised to the sky, carrying an askos; the obverse portrayed his nemesis, Apollo.54 Q. Marcius Rex, could comfortably conflate his own religious conflict with that of Marsyas and Apollo, and draw on Q. Marcius Rex 93 the statue’s polyvalent symbolism as symbol of liberty, champion of the poor, prophet and river deity. A skilled aulos player, this Marsyas, Cybele’s favourite, had foolishly challenged the god Apollo to a musical contest. When he eventually lost, the god is said to have flayed the silenus alive and the blood that flowed created the River Marsyas in Phrygia.55 Once in Italy, however, Marsyas was no longer the foolish challenger of Apollo but the agent of Pater Liber and, perhaps, a symbol of liberty itself.56 His statue could be seen as indicative of the freedom that the plebs had progressively acquired throughout the centuries and in which the Marcii themselves had been active participants, taking the lead in the assumption of the censorship, pontificate and augurate.57 Already in Greece, the musical conflict between Apollo and Marsyas could be politicized as a struggle between rich and poor, between refinement and popular appeal, between the old and the new.58 The askos carried by the forum’s Marsyas may have been a continual reminder of the burdens carried by the poorer classes (and an allusion to the Trojan origins of Rome through Aeneas’ son, Ascanius).59 Later sources record that the statue of Marsyas was unable to bear the face of the younger of the Novii, infamous feneratores (moneylenders).60 Torelli has plausibly argued that the statue was erected by Marcius’ famous censor ancestor, L. Marcius Censorinus, in 294 bc.61 Nexum had recently been abolished and the curule aediles had prosecuted several feneratores. The money from the fines was used for public building projects (Liv. 10.23.11–13). It is possible, as David (followed by Kondratieff) has argued, that the statue’s position close to the comitium and the praetor’s tribunal meant that the praetor urbanus, Quintus Marcius Rex, could be regarded ‘as a champion of debtors and a crusader against usurers’.62 Marsyas, like the Marcii, was also a famous prophet who had introduced augury to Italy.63 In Italy he had given the Marsi their name (Sil. Ital. Pun. 8.502–4) and, on the shores of the Fucine Lake, he had founded his own, now submerged, city, Archippe.64 These stories were being concocted in the 140s bc by the historian Cn. Gellius. The surviving fragments of history reveal an interest in the development of the pontificate, the Sabine origins of Rome, the fusion of Greek and Italian mythology with specific references to Marsyas and the Marsi, and the regal lineage of the gens Marcia.65 Later, Pliny ascribed the origin of the Aqua Marcia to Ancus Marcius, a story that no doubt owed its genesis to the Marcii themselves, but he also traced the Marcia to its original water sources in the Paelignian Mountains. Once called the Aufeia, it passed from its ultimate source, the Pitonia, through the territory of the Marsi and across the Fucine Lake (Pliny, NH 31.41). The origins of that narrative derive from Lycophron’s Alexandra, which dates to the early part of the second century bc.66 There the author refers to the ‘Marsionic waters of Lake Phorkys’, perhaps a wordplay on Marsi/Marsyas, and the underground stream of the Titonia (Pliny’s Pitonia).67 In Phrygia, the River Marsyas too ran underground and had a reputation for limpidity (Phrygiae liquidissimus amnis).68 Like the Aqua Marcia, it was sacred at its source (the fons ejected stones).69 The location of the Marsyas statue in the Roman forum close to the three Sibyls, the statue of Attus Navius and the ficus further served to remind the 94 Alex Nice Roman audience of Marsyas’ conflict with Apollo, the origins of augury, and the importance of prophecy in the Roman Republic. Contained within these con- flicting images of a mythological past was the constant reminder of the necessity for concordia, for a balance between Apollonian reason and Dionysian excess, and the importance of the people. In the spirit of concordia, Marcius’ new and fresh water supply would improve sanitation and healthy living conditions within the city and create extra housing opportunities for the people on the Capitol, thus helping to relieve the pressures on Rome’s infrastructures caused by the incrementum urbis.

Political alliances? With all of this in mind, is it possible to identify the sources of Marcius’ support, those whom his gratia had influenced? Following Münzer, Stuart argued for a senatorial contest pitching an Aemilian-Marcian-Claudian faction against Scipio and his coterie.70 Based on the law passed after Manlius’ conviction in 384 bc, ne quis patricius in arce aut Capitolio habitaret,71 Morgan suggested that Marcius might have garnered the support of plebeian senators who hoped to influence decisions made on the Capitol.72 While the old plebeian-patrician rivalries had been replaced by a new political establishment, patricians still dominated magisterial elections.73 Indeed, attempts to pass populist legislation in the second century bc suggests that an ‘us and them’ divide continued to pervade Roman politics. This might naturally pitch families such as the Aemilii and Cornelii Scipiones against the proposed project but, as we shall see below, at least one major patrician family had good reason to support Q. Marcius Rex. Rodgers objected to this schematic notion of political factions in the Senate:74

[The] entire fabric of political intrigue disintegrates for lack of evidence. In its place we have the senior decemviri Lepidus and Lentulus opposed to Marcius, acting probably in response to the practical interests of the college. On Marcius’ side we have no one in particular, ‘only enormously successful gratia’.75

Of one supporter, however, we can be sure. Quintus was assisted by his relative Publius Marcius (RE 89), almost certainly his brother and the tribune of 171 bc.76 In other cases where gratia played a role in the outcome of an election or senatorial debate, brothers could be influential, even over-zealous, supporters. For example in 193 bc the contest for the consulship was enlivened by the canvass- ing powers of P. Scipio Africanus and T. Quinctius Flamininus (duae clarissimi aetatis suae imperatores or ‘the two most famous commanders of their age’), this former in support of his brother Asiaticus and the latter his brother Lucius. In 184 bc, Ap. Claudius Pulcher acted with undue enthusiasm (effuso studio) in promot- ing the candidacy of his brother Publius, and in 183 Cn. Baebius left his province to hold the elections because his brother was a candidate.77 Along with M. Fulvius and M. Cornelius Cethegus, Publius Marcius had been appointed by the praetor, Q. Marcius Rex 95 C. Sulpicius Galba, to rein in the consul Cassius (Liv. 43.1.12). Patron-client relationships were at the heart of such appointments. Presumably, Publius could use those connections to encourage other senators to join Marcius’ side, not least Ser. Sulpicius Galba (RE 58), probably the praetor’s cousin, who was consul between 144–140 bc and an augur.78 No doubt other members of the gens could be persuaded to support Marcius’ cause. To emphasize: this was the most expensive and grandest building project by far in the second century bc and those associated with the project were set to gain considerably in terms of honour and prestige. It could only add to the social and symbolic capital of the gens Marcia. Marcius could presumably, therefore, lean on the consulares from his own gens: C. Marcius Figulus (cos. 162 and 156; RE 61) and L. Marcius Censorinus (cos. 149, cens. 147 bc; RE 46).79 Different branches of the family were not averse to capitalizing on each other’s successes, as demonstrated by the coinage minted by L. Marcius Philippus (cos. suff. 38 bc; RE 77), moneyer in 56 bc. The obverse of the coins carries a portrait of Ancus Marcius, and the reverse presents an equestrian statue, either representing that of Tremulus in the forum or that of Q. Marcius Rex, proudly prancing atop the Aqua Marcia.80 Furthermore, the construction of the Aqua Marcia was affiliated to the reno- vations to the Anio Vetus and the Appia. The names of those responsible for the construction and renovation of aqueducts were passed down monumentally and in literature.81 Restoration and renewal, as Augustus later demonstrated, could be as important as novel construction (Res Gest. 20). Responsibility and the gloria for the construction of the Anio Vetus belonged ultimately, after the untimely death of the other duumvir aquae perducendae, to M. (or Q.) Fulvius Flaccus.82 The Fulvii Flacci had not had a consul since 179 bc but soon after the construction and renewal of the aqueducts, Servius and Gaius Fulvius Flaccus were elected consul, for 135 and 134 bc respectively.83 Casting the net more broadly, the Fulvii Nobiliores would also have had a vested interest in the construction project. Without accepting Stuart’s far-fetched arguments regarding the proposed aqueduct construction of 179 bc and its relationship to the Aqua Marcia, it is sufficient to note that, as censor, M. Fulvius Nobilior (cos. 189; RE 91) had awarded contracts for a new aqueduct which was never fin- ished.84 Moreover, the Fulvii Nobiliores had their own special association with Numa, Egeria and the Camenae. In addition to her nightly tutorials in divination, Egeria had allegedly first instructed Numa to reform the Roman calendar (Liv. 1.19.5–7). Historically, in 189 bc, M. Fulvius Nobilior had first placed a monumental calendar in his temple of Hercules Musarum as well as moving there the shrine of the Camenae which had originally stood next to the grove where Numa had had his secret meetings with Egeria. Much later, in the first century bc, it was L. Marcius Philippus who had restored Nobilior’s temple, caring for both his calendar and Numa’s shrine to the Camenae.85 More generally, the development and management of the calendar was a pontifical concern.86 The Fulvii and Marcii, therefore, had shared ancestral, antiquarian and religious concerns celebrated in the intersection of the aqueducts at the Porta Capena and the Grove of the Camenae. 96 Alex Nice The renown and monumental memory (handed down to his descendants) of Rome’s earliest aqueduct, the Aqua Appia, fell to its eponymous constructor Ap. Claudius Caecus (cos. 307; 296; cens. 312 bc; RE 91).87 One of the consuls for 143 was an Ap. Claudius Pulcher (RE 295), who was also likely to have been a colleague of Ser. Sulpicius Galba in the augurate.88 In his account of events in 143, Dio Cassius adds the detail that the necessity for the sacrifice in Gaul mentioned by Obsequens had been caused by an argument between neighbouring tribes over the supply of water for the gold mines. Two of the decemviri sacris faciundis were sent to the general Ap. Claudius Pulcher, to make the necessary sacrifices.89 As Coarelli observed, the decemviral objections to Marcius and the religious events in Gaul are united in their references to water; the name Appius; the threat of the Gauls (who had once overrun the Capitoline); and the intervention of the priests.90 Claudius eventually prevailed in the war against the Salassi but was denied the privilege of a triumph: he celebrated it at private expense. The triumph was marked by a disturbance, driven by political jealousy (inimicitia), in which Claudius was only saved by his sister (or daughter), the Vestal Claudia.91 It is not implausible that the attempts to delay Ap. Claudius’ actions in Gaul and to limit his right to triumph (and greater dignitas) were connected to the decemviral objections to the Aqua Marcia. Furthermore, Ap. Claudius had good reason to be associated with the aqueduct: the project would enhance his own claims to be a suitable candidate for the censorship of 142 bc (Plut. Aem. 38.2–5). As Beck has recently observed, drawing connections between aristocratic families through intermarriage is suspect because of ‘the inherent risk that the subject raises suspicions of parties and factions’ or, more bluntly, accusations of ascribing to Münzer’s overly schematic concept of Adelsparteien.92 Nonetheless, one further ally may be mooted. Sex. Iulius Caesar (RE 148, 149) was consul in 157. Either his son or his nephew married the daughter of Q. Marcius Rex.93 A quid pro quo might be inferred. In the 140s bc, the standing of the gens Marcia was at its peak, that of the Iulii Caesares somewhat diminished. That marriage alliance, as Julius Caesar observed in 68 bc, was to prove of enormous benefit to both gentes (Suet. Caes. 6.1). Members of the Marcii, Fulvii, Claudii and Iulii Caesares would have had extensive networks of clients, marriage alliances and senatorial connections. It would have been in their interests to promote the building of the aqueduct, not least for the financial rewards but also for the dignitas, honos and gloria that would accrue from association with the project. There may not be a visible ‘fabric of political intrigue’,94 nor a Marcian ‘faction’, but Q. Marcius Rex was surely not without powerful allies and associates in the Senate.

Religion and the 140s bc So far, we have focused on the political sources of Marcius’ gratia and his potential support in the Senate. Like the Aqua Marcia itself, neither can be divorced from the religious and prophetic reputation of the gens Marcia. Morgan raised the possibility that ‘religious reasons had something to do with the opposition on both occasions’.95 In particular, he sought to explain that religious opposition by Q. Marcius Rex 97 the decemviri’s concerns about encroachment on their lands by squatters.96 The answer to the episode, I think, lies in a more general analysis of religious trends in the late third to early second century bc. This period witnessed a growing interest in foreign religious practices and a tendency towards prophecy. The cult of the Persian mother goddess Cybele had been introduced to Rome in 204 bc and the mysteries of Dionysus had grown to such an extent by 186 bc that the Roman Senate was forced to take radical measures to suppress its worship. Both were forms of worship that appealed to a broader cross-section of society, notably the urban and rustic poor. Astrology and astrologers made their presence felt at Rome. Individuals such as the statesman and scholar C. Sulpicius Galus (cos. 166; RE 66) wholeheartedly embraced Greek astronomical theory and, allegedly, demonstrated its practical utility in affairs of war and politics (Cic. Rep. 1.21–4). His ability to explain an eclipse to the common soldiery was both a matter for awe and suspicion, perhaps because an understanding of lunar and solar cycles was associated with manipulation of the calendar. According to Valerius Maximus, the poorer and less well-educated classes came under the sway of horary astrologers who made deceitful predictions. For this reason, in 139 bc a praetorian edict expelled the Chaldaeans from Rome.97 Finally, responses to prodigia, particularly by the haruspices, show a growing tendency to prophetic rather than ritual prescription throughout the second century bc, although the circumstances surrounding the discovery of the carmina Marciana in 213 demonstrate that there was, potentially, always a richer and less prescriptive aspect to Roman religion.98 Not surprisingly, as Rome’s empire expanded and the opportunities for personal wealth and prestige increased dramatically, political and religious conflicts became increasingly intertwined. Beard, North and Price have demonstrated that those religious conflicts from the late third and early second century which are known to us turned on points of religious law.99 Inevitably, the loser was the mag- istrate who would have to cede to the decision of, normally, the pontifex maximus. In the case of the Sibylline Books, there are few glimpses of similar disputes. However, one example, which also relates to the gens Marcia, suggests that they were already exploring ways to promote their family fortunes via divination. In 169 bc, T. Marcius Figulus (RE 65) had reported that a palm tree had sprung up in his house. The prodigy was rejected because it had occurred on private ground (Liv. 43.13.6). Titus was surely the brother of the praetor C. Marcius Figulus (cos. 162, 156; RE 61) and the cousin of Q. Marcius Philippus, then consul for the second time. With war against Perseus on the horizon, the political and military ambitions of both were evident and it may be argued that T. Marcius was attempting to influence the allocation of the military commands on behalf of his relatives. It appears that the decemviri found a diplomatic solution without offending a senior senator: another palm tree which had sprouted in Marcius Ralla’s temple of Fortuna Primigenia was recognized (Liv. 43.13.1–8). Caius was given command of the fleet and Philippus command in Macedonia (Liv. 43.15.2–3, 44.1–2). The Senate and its priests were ever wary of families laying claim to too much kudos for themselves.100 98 Alex Nice It is perhaps no surprise, therefore, to see that the decemviri were able to find an oracle in the books which prophesied that the construction of an aqueduct to the Capitoline was nefas. The decemviri, however, could act only on the information passed to them. Prodigies were reported to the Senate, which would decide whether they were pertinent, and then refer them to the appropriate authority: pon- tifices, decemviri or haruspices.101 It seems unlikely, therefore, that the decemviri were ‘already intent on opposing the aqueduct’.102 Nor is it correct to argue that this was an example of a ‘non-expiatory’ pronouncement since, at the end of the day, the prodigy was disregarded.103 As we noted above, the problem of the water supply in Gaul and the threat this implied to Rome was satisfied by a sacrifice conducted by the decemviri. In part, the episode can be explained by a prevailing atmosphere of religious and political conservatism, perhaps under the influence of Scipio Aemilianus and his supporters.104 In 149 or 146 bc, the ludi saeculares in honour of Dis and Proserpina were held, perhaps, to coincide with the outbreak of the Third Punic War.105 During the course of the Third Punic War, Scipio Aemilianus ‘devoted’ Carthage to the gods of the underworld and ‘evoked’ Juno of Carthage (Macr. Sat. 3.9.6–11). In 145 bc, when C. Licinius Crassus, proposed that priesthoods should be filled by popular election rather than the current means of co-optation by existing members of the priesthood, C. Laelius (later Sapiens) successfully opposed this measure, drawing attention to the arcane vocabulary and traditions of the priestly colleges.106 The reasons for the proposal are not immediately clear, although it seems that it might have been made at a time when ‘membership of the colleges and their activities could be perceived as standing in the way of popular legislation’.107 The time was not yet right for a more democratic process in respect of Rome’s sacred institutions: that would have to wait until the lex Domitia of 104 bc (Cic. De Leg. Agr. 2.18; ad Brut. 1.5; Vell. 2.12). To this same period belong the leges Aeliae Fufiae, which sought to regulate the procedures for obnuntiatio, perhaps to confirm the unwritten aspects of its practice. These laws may have confirmed that augurs and magistrates could use obnuntiatio to prevent or delay the passage of controversial measures at legislative assemblies.108 The construction of the Marcian aqueduct would have allowed settlement on the Capitol in greater numbers than had previously been allowed. This might have given Q. Marcius, and his supporters, undue influence over the legislative and judicial meetings held on this spot. In view of the possible benefits to the plebeian classes, the populist concerns of the Marcii and the prospects of enhancing his own gratia, it is not surprising that the opposition to Marcius came not only from two Xviri but also two members of the political establishment: M. Aemilius Lepidus and L. Cornelius Lentulus, which sought to limit the extent that one individual could court favour among his peers and the populace. This appears to have been a genuine concern in the 140s. To 143 also belongs the lex Didia, which extended the sumptuary legislation of the lex Fannia to the whole of Italy and made it an offence not only for individuals to give lavish dinner parties but also to attend them.109 In the same period too, Laelius was forced to drop his proposed land bill (Plut. Ti. Gracch. 8.5). Both are arguably connected to the problems Q. Marcius Rex 99 caused by an influx of wealth and manpower during the 140s and both can be viewed as limiting the extent to which individuals could extend their client base. In short, the unwritten rules of political conduct may have motivated the decemviral opposition to the aqueduct. No doubt they were afraid of how much influence would accrue to Marcius should his aqueduct be successful.110 By the same token, however, Marcius had right and enormous gratia on his side. The pronouncement by the decemviri was evidently false and may not have appealed to those senators of a more conservative disposition, despite their misgivings about the extent to which Marcius’ might win popular favour. As time passed, it appears that senators were more willing to place religious objections in the way of political proposals. Not many years later, when Gaius Gracchus wanted to build his colony, Junonia, on site of ancient Carthage, his opponents said that wolves had torn up the boundary markers. Despite Gracchus’ objections (there were, of course, no wolves in Africa), senatorial opposition prevailed, and the lex Rubria was repealed.111 In the first century bc, other patently false oracles swayed the Senate.112 A particular reading of one oracle in 87 bc allowed the consul Octavius to expel Cinna and six tribuni plebis to guarantee ‘peace, tranquillity and security’ for Rome (Gran. Lic. 35.1–2); in 56 bc, the Books tactfully thwarted Pompey’s attempt to restore Ptolemy Auletes. They encouraged a warm reception but predicted danger to the state should the Egyptian king be assisted with a multitude (Cassius Dio, 39.15.2–16.1). Others failed to gain traction at all: in 63 bc, P. Cornelius Lentulus, the Catilinarian conspirator, had laid claim to a Sibylline oracle predicting power in Rome for three Cornelii (Cic. Cat. 3.9; Sall. Cat. 47.2; Plut. Cic. 17.4); in 44 bc, the quindecemviri were alleged to have found a verse in the Sibylline Books which said that only a king could defeat the Parthians (Cic. Div. 2.110; Suet. Caes. 79; Plut. Caes. 60, 64; Cassius Dio, 44.15.3). Cicero offers the oracle as a wordplay on the noun rex. In another situation, on his way from Alba to Rome, Caesar had allegedly proclaimed that he was a Caesar not a rex.113 Oracles which might be directly relevant to the Marcii Reges were then to be found in the Sibylline collection. Such prophecies, Cicero says, were kept deliberately obscure so that they might be adapted to different situations and different times.114 No doubt this was also true in the affair of the Aqua Marcia and Q. Marcius Rex. Where in previous times the conflicts were primarily about matters of religious propriety between noble equals, they were beginning to take on a greater significance: to expose the rifts which existed between the concerns of the people and the ruling establishment.115 In the case of the Aqua Marcia, contracts had already been settled and millions of sestertii guaranteed to the interested parties.116 Much more than reputation was at stake.

Conclusion The early history of the Aqua Marcia cannot be divorced from the sacral importance attached to water and its associations with divination. In this respect, the very construction of the Marcia connected the Latin world with that of Rome. Pliny the Elder extended the Marcia’s course chronologically (back to Ancus Marcius) 100 Alex Nice and geographically (to the Fucine Lake).117 The origins of his mythography can be traced back to Lykophron’s Alexandra and the histories of Cn. Gellius, who had composed his monumental work at a moment contemporary with Marcius’ aqueduct project. The Marcia thus traversed (mythologically, at any rate) not only the countryside of Italy but also the history of Roman divination, from the Marsian snake charmers, herbal healers and augural experts, past the monuments sacred to Numa and the Muses, before arriving at the Capitol, the home of the Sibylline Books and Roman augury. In those surroundings, Q. Marcius Rex was able to make the most of his family’s ‘accumulated symbolic capital’. His gratia could lean on the legendary and physical reminders of his ancestors’ achievements, their plebeian concerns, political successes and their roles as priests and prophets. Essential to Marcius’ gratia must have been the statue and presence of the silenus Marsyas in the Roman forum. Inextricably linked to notions of popular power (his aulos/ tibia was the sound of the lower classes) and associated with the urban praetor’s platform, Marsyas was also a symbol of debt relief and a champion against the feneratores, and a divinity with a special expertise in augury. This skill was further emphasized by his position in the forum, close to reminders of the origins of Roman divination. In the context of the Aqua Marcia these aspects of his character are incidental: Marsyas the river god takes on greater significance. The watery connotations were enhanced by his Greek myth, which emphasized his transformation into a river god, and the limpid waters of the Phrygian River Marsyas, along with the prodigial connotations of its source. Religion at Rome was often used to examine issues of political concern, particularly if those issues were likely to enhance the auctoritas of an individual beyond what was socially acceptable. It provided an important mechanism to delay action where appropriate and to offer an alternative system of checks and balances which could not be directly attributed to political opposition but rather the will of the gods. The prodigies reported in 143 bc not only implicated the Aqua Marcia but involved another watery problem on Rome’s borders. There, an ally of Marcius was involved, Ap. Claudius Pulcher. His ancestor had constructed an aqueduct whose conduits Marcius had repaired. In a previous generation, the carmina Marciana had warned the Romans to beware a river, to make sacrifices and to vow games to drive the enemy from their borders. Plausibly, the decemviri might have argued that they had found an oracle relevant to the Marcian aqueduct. The popularis motivations of Marcius may have put him at odds with the establishment concerns of the Aemilii Lepidi and Cornelii Lentuli, worried that one of their number might gain disproportionate influence. The feigned discovery of oracles prohibiting Marcius’ building project foreshadow the political contests of the later second and first century bc. The immediate profit for Q. Marcius Rex was the rare honour of a statue prominently placed on the Capitol behind the temple of Jupiter.118 The real story of Marcius’ victory, however, is revealed in the succeeding generations. His son and grandson were respectively consuls in 118 bc and 68 bc. In the same year, at his aunt’s funeral, Julius Caesar could claim descent from the Marcii Reges. In 56 bc, Q. Marcius Rex 101 the moneyer L. Marcius Philippus issued coinage displaying, on the obverse, the head of Ancus Marcius with lituus behind and, on the reverse, a togate equestrian statue above five arches inscribed AQUA MARC – references to the legendary origins, res gestae and historical monumenta of the Marcii.119 Commemorated on the Capitol, the seat of the Sabine kings and the home of Roman divination, Q. Marcius Rex had symbolically transcended his own ancestors. Plutarch might rightly praise the success of the aqueduct builders. Q. Marcius Rex’s contribution to the collective memory and symbolic capital of the gens Marcia helped to embed the family in the foundation story of the principate (Ov. Fast. 6.693–710). Ultimately, it was Rome that profited most economically and socially from Marcius’ defeat of the decemviri. For throughout the empire, Marcius’ aqueduct continued to supply fresh, clean and clear drinking water to the city and its citizens.

Notes 1 I am grateful to Federico Santangelo, Bill Dominik and the anonymous referees for various suggestions and improvements to this paper. 2 Front. Aq. 7.6–8; cf. Strabo, 5.3.13; Pliny, NH 31.41. On the archaeology of the Aqua Marcia see Ashby (1935); Sartorio (1986) 42–56; Aicher (1995) 36–7, 56–8, 159–62; Rodgers (2004) 158–68. 3 Rodriguez Almeida, LTUR 1.216, s.v. Camenae, Camenarum fons et lucus; Richardson (2016). 4 On Numa’s relationship with Egeria see Varr. Ant. Rer. Div. 1, app. 4 = Cardauns (1976) i.46; Liv. 1.19.5; 1.21.3; Dion. Hal. 2.61; Plut. Numa 4.1, 13.1, 15.5. Porta Capena: Juv. Sat. 3.10–20; Mart. 3.47.1. 5 Scullard (1981) 62–4. 6 Front. Aq. 7.1–5: post annos centum viginti septem, id est anno ab urbe condita sex- centesimo octavo, Ser. Sulpicio Galba Lucio Aurelio Cotta consulibus cum Appiae Anionisque ductus vetustate quassati privatorum etiam fraudibus interciperentur, datum est a senatu negotium Marcio, qui tum praetor inter cives his dicebat, eorum ductuum reficiendorum ac vindicandorum. et quoniam incrementum urbis exigere videbatur ampliorem modum aquae, eidem mandatum a senatu est, ut curaret, quatenus alias aquas posset in urbem perducere. * priores ductus restituit et tertiam illis salubri- orem * duxit, cui ab auctore Marciae nomen est. 7 For the meaning of incrementum in this context see Morgan (1978) 28, n. 14. 8 Front. Aq. 7.1.4–5 = Fenestella: F12 Cornell (= F10, Peter). Cornell only attributes 7.4 to Fenestella. 9 Rodgers (1982) esp. 177. 10 Wissowa (1912) 538. 11 Inventum (est) is occasionally used for decemviral discoveries in the Sibylline Books: Liv. 5.14.4; 10.47.7; 29.10.6. It is used elsewhere for the discovery of the prodigy or ritual fault at Liv. Per. 14.3; 5.17.2; 27.26.14; 41.14.7; 39.22.5; Obs. 17; 55. 12 It is possible that the historical record has ‘ironed out’ traces of the prophetic. See, for example, Mazurek (2004) on the prophetic nature of the Sibylline Books. He discusses the case of the Marcia at pages 159–60. 13 POxy. 188–90 (142–139 bc): ---]nae devota est aqua annio aqua/Marcia in Capi]tol- ium/contra Sibyllae carmina./perducta]. 14 The emendation pro collega to pro collegio is preferred, see Rodgers (2004) 165. 15 On the notion of gratia see: TLL 6.2.2205.80; 2.2206; 2.2211.10; 2.2212. Schertz (2005: 157–8, cf. 119) tentatively suggested that ‘The tradition of the Marcii as religious leaders and seers perhaps lent auctoritas and credence to Marcius Rex’s arguments’. 102 Alex Nice 16 Moussy (1966) 375–91. 17 Cic. De Inv. 2.66: gratiam, quae in memoria et remuneratione officiorum et honoris et amicitiarum observantiam teneat; cf. 2.161. 18 Cic. Fin. 2, 117: tollitur beneficium, tollitur gratia, quae sunt vincla concordiae. Cf. Ov. Ars Am. 2.463–4. 19 verboven (2007) 866–7. 20 For other examples of gratia within the Senate see: Liv. 2.44.3; 4.48.15; 4.57.4; 4.57.6; 5.8.13; 5.29.7; 6.37.7; 22.26.4; 33.22.6; 33.46.7; 39.38.10. 21 Hölkeskamp (2010) 107–24; (2015) 182. 22 Pina Polo (1995). 23 Hölkeskamp (2016) 181–2. 24 For example, see Cic. De Agr. 2.1. In general, see Morstein-Marx (2004) 77–104. For the location of meetings see Taylor and Scott (1969) esp. 559–61; cf. Weigel (1986) 333–40. 25 See Morstein-Marx (2004) 177–207; Hölkeskamp (2016) 169–213. 26 Hölkeskamp (2016) 177 and, more generally, 175–81. See also Manuwald (2007) 791–4. 27 Itgenshorst (2005) 254; cf. Hölkeskamp (2016) 183–6. 28 Cic. Amic. 96; Varr. Rust. 1.2.9; cf. Plutarch, who ascribes the innovation to Gaius Gracchus (Gracch. 5.3). 29 On Tremulus and his triumph see Liv. 9.42.10–43.24, cf. Diod. 20.80.1–4; Münzer (1930) RE 14.2, 1596–7, s.v. Marcius, 106; MRR 1.165–6; van Ooteghem (1961) 41–5; Oakley (2007) 3.555–68; Statue: Liv. 9.42.10–11; 9.43.22; Pliny, NH 34.23; Bergemann (1990) 156, cat. L10; Richardson (1992) 145–6; Papi, s.v. Equus Q. Marcius Tremuli in LTUR II, 229–30; Sehlmeyer (1999) 57–60. 30 Wiseman (1985). 31 MRR 2.587–9. 32 Hölkeskamp (2010) 81–2. 33 Liv. 1.18.6–10; Dion. Hal. 2.76.5 = Cn. Gell. Fr. 22 Cornell; Liv. 1.32.1; Plut. Numa 21.4–6; Cor. 1. 34 Dion. Hal. 3.44–5; cf. Liv. 1.33.6 and 9; Camous (2004). 35 Liv. 1.20.6–7. On the pontificate and its functions see Beard, North and Price (1998) 1.24–6; cf. 55–8; Van Haeperen (2002); Berthelet (2011). 36 Consul (Liv. 9.33.1; Diod. 20.27.1); censor 294 (Liv. 10.47.2; Val. Max. 4.1.3; Plut. Cor. 1.13; cf. Jerome, Chr. p. 128) and 265 bc (Fast. Cap.; Val. Max. 4.1.3; Plut. Cor. 1.1); pontiff (Liv. 10.9.2; 10.10.6–9; ILS 9338); augur (Liv. 10.9.2; Val. Max. 4.1.3; Plut. Cor. 1; ILS 9338). See Rüpke (2008) 790, s.v. 2384, C. Marcius C.f. L.n. Rutilus Censorinus. 37 Liv. 27.6.15–16. Rüpke (2008) 2.1135; MRR 1.283; Szemler (1972) 174–5; Bianchi (2010) 136–7. 38 On the duties of the rex sacrorum see Beard, North and Price (1998) 1.55–8; Bianchi (2010). 39 Liv. 34.53.5–6; Alföldi (1972). 40 Philippus’ decemvirate: Liv. 40.42.12; see also Liv. 41.21.10–11; cf. Obs. 10. See Engels (2007) 512–13. As consul, Philippus had overseen the prosecution of the Bacchanalian conspiracy: Liv. 39.8–19; CIL i2.2, 581. See also Briscoe (2008) 230–90, esp. 230–49. 41 Liv. 25.1.6–13. On this event and the carmina Marciana see Bouché Leclerq (1882) 4.128–30; Gagé (1955) 270–9; North (2000a) esp. 100–2; Russo (2005); Engels (2007) 766–8. Marcii: Cic. De Div. 1.89; 2.113; Marcii vates; Serv. Aen. 6.70, 72. On Marcius vates: Cic. De Div. 1.115; Liv. 25.12.2; Amm. Marc. 14.1.7; Zon. 9.1 (417a–b); Fest. 165L (Cn. Marcius); Pliny, NH 7.119; Arn. 1.62; Porph. ad Hor. Sat. 2.1.26; Symm Ep. 4.34; Isid. 6.8.12; Macr. Sat. 1.17.25. 42 Consul 357 (Liv. 7.16.1; Diod. 16.28.1), 352 (Liv. 7.21.4; Diod. 16.52.1), 344 (Liv. 7.28.6; Diod. 16.74.1) and 342 bc (Liv. 7.38.8; Diod. 16.82.1; Dion. Hal. 15.3.1); dictator 356 bc (Liv. 7.22.6); censor 351 bc (Liv. 10.8.8; cf. FGH 2 B.1153, no. 255. Degrassi 105, 404f.). Q. Marcius Rex 103 43 Liv. 7.22.9–10. For an analysis of this period, including Rutilus’ censorship and patrician opposition see Hölkeskamp (1987) 62–113. 44 Lintott (1999) 115–20. 45 Election: Liv. 7.22.9–10; C. Marcius Rutilus Censorinus (cens. 294 and 265): Val. Max. 4.1.3; Plut. Cor. 1. 46 Arming the proletariat: Hemina, F24 Cornell = Non. 67M = 93–4L. Enn. Ann. 170–2 (Skutsch); Oros. 4.1.3; Aug. CD 3.17. See Cornell (2013) 3.172–3; stipendium: Pliny, NH 34.23. 47 Taylor and Scott (1969) 559–61; cf. Weigel (1986) 333–40. The Oxyrhynchus papyrus appears to preserve the phrase in aede vota est, potentially a reference to the place of the meeting. 48 Jaeger (1997) 122–31. 49 Liv. 25.39.16. For the full account see Liv. 25.37.1–39.18; App. Hisp. 7; Val. Max. 2.7.15; 8.15.11. 50 Liv. 8.5.1–8.6.7, esp. 8.5.8–10. Cf. Cic. Cat. 1 and 3 and his references to Jupiter. Vasaly (1993) esp. 40–87; Morstein-Marx (2004) 92–107; Hölkeskamp (2016) 170–213. 51 Cic. Dom. 30; Pliny, NH 7.214; Cens. De Die Nat. 23.7. 52 Storchi Marino (1992); Humm (2014). 53 On this statue generally see Small (1982) 68–92; Torelli (1982) 98–105; Rawson (1987); Coarelli (1985) esp. 91–123; Denti (1991); Schertz (2005); Guzzo (2015); Santangelo (2016). 54 RRC 1.363. 1a–d. 55 On the myth of Marsyas see Weis (1992). 56 Recently disputed by Santangelo (2016), although this reading was already implicit in Greek versions of the myth. 57 Torelli (1982) 105; Coarelli (1985) 111–19. 58 Wilson (2003) esp. 185–7. 59 Reinach (1914) 321–37. 60 Hor. Sat. 1.6.119–21. See also Acron ad Hor. Sat. 1.6.120–1; Porph. ad Hor. Sat. 1.6.120–1; cf. Mart. Ep. 2.64. Coarelli (1985) 101–11. 61 Torelli (1982). 62 Kondratieff (2010) 97; David (1995) 375. 63 Cn. Gell. Fr. 17 Cornell (= Solin. 1.8–9); Serv. Ad Aen. 3.359. 64 Cn. Gell. Fr. 16 Cornell (= Pliny, NH 3.108); Sil. Ital. Pun. 8.502–4; cf. Solin. 2.6. 65 Cornell (2013) 2.362–83. 66 Hornblower (2015), 36–9; 168 bc: White (1997) 49–51. 67 Lyc. Alex. 1270–80; Letta (1972) 59; D’Amato and Letta (1975) 206–15: ‘La leggenda dell’aqua Marcia potrebbe essere la riedizione in funzione romana di tale primitiva leggenda e la principale responsabile della sua caduta in oblio’. For the ‘Lake of Marsyas’ see Wiseman (2004) 21–2; Hornblower (2015) 451. 68 Ov. Met. 6.392–400; cf. Curt. 3.1.1–5. Pliny, NH 31.41. 69 Celaenae: Herod. 7.26; fons: Pliny, NH 31.19. 70 Stuart (1943) 443 n. 18; (1944) 42; (1945) 248–51. 71 Liv. 5.50.4; Ogilvie (1965) 740. 72 Morgan (1978) 49–52. 73 Hölkeskamp (2010) 84. 74 Morgan (1978) 26, 35–43; Rodgers (2004: 165–6) overemphasizes the extent to which Stuart (1943, 1944, 1945) ‘constructed an elaborate nexus of political alliances’. 75 Rodgers (2004) 165. 76 Plut. Cor. 1; Liv. 43.1.12. Münzer (1930) RE 14.2, 1582, s.v. 90. 77 Liv. 35.10.1–11 (Scipio and Quinctius); Liv. 39.32.5–15 (Claudius); Liv. 40.17.8 (Baebius); cf. Liv. 39.42.7; 42.10.10. 78 Rüpke (2008) 909 no. 3189. 79 MRR 1.447 (Figulus), 1.458 (Censorinus). 104 Alex Nice 80 On the coinage and its interpretation see RRC 1.448–9; Stuart (1945); Geschke (1968) 25–42; Morgan (1978) 38–41; Evans (1992) 139–40. 81 Cf. Wiseman (2014) 48. 82 Front. Aq. 6.4. See Rodgers (2004) 151–8; 153 s.v. 6.3 on Fulvius. 83 MRR 1.488–90. 84 Liv. 40.51.7; Stuart (1943) and (1945). 85 Richardson (1992) 187. 86 See p. 91 above. 87 Liv. 9.29.5–6: et censura clara eo anno Ap. Claudi et C. Plauti fuit; memoriae tamen felicioris ad posteros nomen Appi, quod viam munivit et aquam in urbem duxit; [6] eaque unus perfecit; Front. Aq. 5.3: nomen aquae ad Appii tantum honorem perinuit. 88 MRR 1.471–2. Augur: Rüpke (2008) 619–20 no. 1225. 89 Cassius Dio, 22 Fr. 74.1; cf. Oros. 5.4.7. Strabo (4.6.7) relates that the Salassi were depriving tribes in the valley below of the water necessary for irrigation. 90 For this see Coarelli (1988) 315–23. 91 Cic. Cael. 34; Val. Max. 5.4.6; Suet. Tib. 2.4–5; Dio Cass. Fr. 74.2; Oros. 5.4.7; Macr. 3.14.14. 92 Beck (2015) 61. 93 Suet. DJul. 6.1. Münzer (1930) RE 14.2, 1601, s.v. 113 ‘Marcia’. 94 Rodgers (2004) 165–6. 95 Morgan (1978) 48. 96 Morgan (1978) 48–9, based on Oros. 5.18.27. 97 val. Max. 1.3.3. Morgan (1978: 52–3) offers an unconvincing explanation of this episode as an attempt to thwart the lex Gabinia. 98 North (2000b). See for example Liv. 38.45.3 (187 bc); Obs. 18 (152 bc), 27a (133 bc), 28 (130 bc), 28a (129 bc), 29 (126 bc), 37 (114 bc), 44 (102 bc). 99 Beard, North and Price (1998) 1.105. 100 On this episode see Engels (2007) 517–21 esp. 521; Briscoe (2012) 428–32. 101 On the discovery, referral and interpretation of prodigia see the overviews in Rasmussen (2003) 47–52; Engels (2007) 750–9; Orlin (2010) 111–36. 102 Morgan (1978) 48. 103 Mazurek (2004) 159–60. 104 On prophecy, for example, see North (2000b); on conservatism and the Scipiones, see Zetzel (1972); Rawson (1973); Beard, North and Price (1998) 1.109–10. 105 Cens. De Die Nat. 17.11. On the Secular Games in general see Forsythe (2012) 50–76. 106 Cic. Lael. 96; MRR 1.470. 107 Beard, North and Price (1998) 1.109. 108 See Astin (1964); Sumner (1963); Beard, North and Price (1998) 1.109–10. 109 Macr. 3.17.6. Cf. Kübler RE 4a, 295; 905; Weiss RE 12, 2353. 110 Rodgers (2004) 163–4. 111 App. BC 1.3.24; Rawson (1974) 197–8. 112 The situation was made more complex after the destruction of the Books in the fire which engulfed the temple of Jupiter in 83 bc: see Dion. Hal. 4.62. The collection was reconstituted in 76 bc: Varro Fr. 60 Cardauns (Dion. Hal. 4.62.5); Varro Fr. 56a Cardauns (Lact. Inst. 1.6.7–12); Fenestella Fr. 18 Peter (Lact. Inst. 1.6.14); Tac. Ann. 6, 12; Santangelo (2013) 147–8; Keskiaho (2013) 165–9. 113 Cic. Div. 2.110; Suet. Iul. 79.2; Deutsch (1928), 394–8. 114 Cic. Div. 2.111. This is, of course, Cicero’s sceptical persona arguing against the value of divination. 115 Beard, North and Price (1998) 105. 116 On contracts see Front. Aq. 95–6. 117 Pliny, NH 31.41; cf. Lyc. Alex. 1270–80. 118 CIL 16.5; cf. RRC 1.449, but Front. Aq. 1.7 mentions the statue nowhere. Q. Marcius Rex 105 119 On the coinage and its interpretation see RRC 1.448–9; Stuart (1945); Geschke (1968) 25–42; Morgan (1978) 38–41; Evans (1992) 139–40. On the statue see above notes 26, and 29. Hölkeskamp (2016: 189–91) argues that the togate statue must represent that of Tremulus. A togate equestrian statue would also be suitable for Q. Marcius Rex, who had not won a military victory. There were precedents: a togate equestrian statue of Cloelia on the Via Sacra celebrated another watery feat – her swim across the Tiber. See Liv. 2.13.11; Pliny, NH 34.28–9. 8 Valerius Maximus and the language of stars

Jeffrey Murray

In 139 bc, by order of a Praetorian edict, astrologers were expelled from Rome and Italy. Our principal source for this event is the Tiberian author Valerius Maximus, whose Facta et Dicta Memorabilia, one of the few extant collections of Latin exempla from the imperial period, is at this very point plagued by a major lacuna in all extant manuscripts.1 The chapter on superstitions (1.3), however, in which this exemplum falls, has been preserved in the two late antique epitomes of Valerius’ work by Julius Paris and Januarius Nepotianus. This expulsion of astrologers – the first of its kind on record – was brought about during the consulship of Cn. Calpurnius Piso and M. Popillius Laenas by the praetor peregrinus, Cn. Cornelius Hispalus.2 In Paris’ account, a double motivation is given as the justification for this expulsion: not only were the astrologers’ predictions false, but, second, they also gained profit by them (‘For they spread profitable darkness with their lies over frivolous and foolish minds by fallacious interpretation of the stars’; levibus et ineptis ingeniis fallaci siderum interpretatione quaestuosam mendaciis suis caliginem inicientes). The financial advantage is also made clear by Nepotianus’ use of the verb venditarent (‘they tout’).3 In his discussion of both passages, Frederick Cramer gives primary importance to the illegitimacy of astrology as an acceptable means of divination, but as David Wardle has pointed out, the Latin more likely indicates that only fallacious interpretations by astrologers were condemned; furthermore, he notes that in all subsequent expulsions we know of in imperial Rome, nowhere is material profit given as a motive.4 Eric Orlin recently argued that instead of focusing on the groups being expelled – in this case, astrologers and Jews – it is perhaps more worthwhile to focus on the very act of expulsion itself, and the possible reasons behind why the Romans would have employed such a tactic.5 This approach has merit, especially considering the fact that these expulsions were never entirely successful, as both astrologers and Jews are found at later dates in Rome. Orlin argues, therefore, that this expulsion and others like it were merely symbolic, an ‘attempt to draw boundaries between Roman and non-Roman’, thereby highlighting religion as the principal site for defining Romanness.6 While useful in understanding potential ideological reasons for expulsions more generally, Orlin’s approach fails to explain the given motivation of profit. Valerius Maximus and the language of stars 107 Geraldine Herbert-Brown, in turn, has suggested that Valerius’ inclusion of the expulsion of astrologers in 139 bc may have been influenced by reflection on contemporary events. Astrologers, for instance, appear to have been behind Libo Drusus’ plotting in ad 16, which resulted in Libo’s suicide, and possibly their subsequent expulsion by Tiberius the following year (cf. Tac. Ann. 2.32.3; Dio 57.15.8–9).7 While this may provide some of the rationale behind Valerius’ choice to include this specific example, the material motive may simply have been imported from his source. Although our sources as they stand do not allow us complete certainty, Livy is Valerius’ probable model for this exemplum, given not only Valerius’ reading pattern within this chapter specifically and also more generally across his work, but also given his choice of language; for the Oxyrhynchus Epitome also designates the astrologers as Chaldaeans (P.Oxy. 668 II. 191–2).8 In his history, Livy presents negative appraisals of a number of greedy charlatans as well as foreign superstitio from Rome’s Republican past: the pretend seers, for example, who, in the aftermath of a drought in 430 bc, introduced new rites – a group of people, Livy tells us, who made their profit by captivating men’s minds with superstitions (4.30.9); or the priests and prophets (sacrificuli ac vates), in another example, who preyed on people’s anxieties during the Second Punic War (25.1.8); or even in his account of the Bacchanalia scandal of 186 bc (39.8.1–8). It is surely likely, therefore, given this general pattern in Livy’s work, that Valerius not only imported the substance of the exemplum from him, including the financial motivation, but also his morally condemnatory tone.9 Pauline Ripat has recently provided a useful reassessment of what our evidence actually tells us about the various Roman expulsions of astrologers. She also, furthermore, cautions us to examine more carefully the judgements we make about the place of astrology within Roman society more generally.10 In Valerius’ presentation of the expulsion of 139 in particular, she rightly, I think, sees an ‘exercise in undercutting non-elite divinatory activity’, and argues that the ‘paternalistic concern’ which Valerius offers towards those ‘foolish minds’ is false, used only to portray these prophets in a negative light; in her words, ‘the spending habits of the poor rarely concerned the rich’.11 The generally negative attitude towards astrologers and astrology among Romans of the earlier Republic soon gave way to an increasing popularity at Rome, which reached its zenith under Tiberius.12 Even so, expulsions of astrologers continued (as seen in the example above in the aftermath of the Libo Drusus affair). While astrology was never incorporated into the structures of state religion as a recognized form of divination, the emperor’s personal beliefs appear somewhat different. From the time of his self-imposed retirement on the island of Rhodes, Tiberius showed himself to be a keen amateur in astral matters, with the astrologer Thrasyllus as his guide and friend, as both Suetonius (Tib. 14.4) and Tacitus (Ann. 6.20–1) tell us. Cassius Dio, whose generally positive account of astrologers at Rome is likely informed by his Severan context, confirms this portrait (cf. 55.11.1, 57.19.3–4).13 Written during the reign of Tiberius, to whom the work is dedicated, the Facta et Dicta Memorabilia, is arranged into various chapters illustrating a range of 108 Jeffrey Murray topics and themes. Much of the substance of many of the exempla, however, comes from earlier authors, and Valerius’ role was largely that of editor. Given Tiberius’ own enthusiasm for astrology, therefore, the presence or lack of astrological elements in the work becomes pertinent. One scholar, for example, has claimed that ‘The antiquarianism of Facta et Dicta Memorabilium [sic] seems to be at odds with the changing political, social and religious circumstances of the Empire’ and continues to claim that ‘Valerius Maximus does not exhibit the enthusiasm of his contemporaries for astrology and other forms of divination. Like Livy and his predecessors, Valerius’ work is backward-looking and anachronistic. It celebrates “traditional” religion which was believed to have characterized the Republic’ and does ‘not adequately reflect his own time’.14 In direct contrast to this, turning to the only monograph dealing with aspects of Roman religion in Valerius Maximus, one is confronted by the following explanation by Hans-Friedrich Mueller, ‘Valerius’ enthusiasm for stars’ he says, ‘may be contrasted with the dearth of astrological interest’ found, for example in an author such as Pliny, and he goes on to claim that ‘Valerius’ belief in stars would … put him in the mainstream of imperial thought’.15 Mueller then goes on to cite Franz Cumont for further discussion of astrology under the empire, and qualifies what he has been saying by concluding that ‘all star systems (as all gods) are not equal. Foreign interpretations, as Valerius 1.3.2 [sic; Val. Max. 1.3.3] (regarding the expulsion of Chaldaeans and Jews) points out, must sometimes be expelled’. The fact, therefore, that there is no chapter De Astrologia in the first book, which takes as its focus Roman religion generally, appears at first sight a glaring omission, as well as confirmation of Valerius’ supposed ‘traditional’ and ‘anachronistic’ bent posited above.16 While it is true that, despite Tiberius’ personal interest, the subject finds no formal treatment within Valerius’ work, this does not mean that Valerius was unaware of, or even unconvinced by, contemporary fashions. Sidereal matters do on occasion find expression within his work, and indeed what is most striking about these instances is, perhaps, where they occur. For the majority of references either occur in crucial passages that comment specifically on contemporary events, or within Valerius’ own authorial comments on the material he excerpts – passages where he does not rely on the words of his sources. While my deliberate slippage here between astrological and sidereal elements might conflate matters that should be kept separate – that is, astrology proper and the language of the stars associated with imperial cult – it is useful, I argue, to consider them alongside one another, as indeed is often done by both the sources themselves and the secondary literature.17 The tradition of scientific astrology that was gaining ground at Rome during the early empire is perhaps best seen in the positive portrayal that Valerius affords exemplars of its practice. At 3.3.ext.4, for example, Valerius recounts the torture of Anaxarchus, Greek philosopher and friend of Alexander the Great, by Nicocreon, the tyrant of Cyprus. At the threat of having his tongue cut out, Anaxarchus bites it off, chews it up and spits it into the tyrant’s mouth. In praising the philosopher as a model of endurance (patientia), Valerius refers to his tongue which so eloquently expounded the movement of the stars (siderum motus). Similarly, at 8.7.ext.2 Valerius highlights Pythagoras’ diligence (industria) in learning astronomy Valerius Maximus and the language of stars 109 from the Persians (‘He went to Persia … His docile mind absorbed what they ungrudgingly displayed to him: the motions of the stars, the courses of the planets, the force, individuality and effect of each one’; inde ad Persas profectus … a qui- bus siderum motus cursusque stellarum et uniuscuiusque vim proprietatem effec- tum benignissime demonstratum docili animo sorpsit). A final example, also in Book 8, illustrates just how important was Sulpicius Galus’ knowledge of celestial phenomena in bringing about the Roman victory at the Battle of Pydna in 168 bc (Val. Max. 8.11.1). In the examples discussed thus far, Valerius could be seen to be negative, neutral or positive regarding prediction by the stars – in each case, his attitude is dependent on other variables. In this way, through his didactic purpose of educating his readers in ethical thinking, Valerius enables them to judge each specific situation or context on its own merits – what Rebecca Langlands has termed a ‘situational ethics’.18 Elsewhere, Valerius uses sidereal imagery to highlight the virtue of individuals. For example, in an exemplum of Scipio Africanus’ majesty (2.10.2b), Valerius relates an anecdote from during Scipio’s retirement in Liternum, when he was visited at his country villa by pirates who revered him as a god.19 In his concluding comment to this example, Valerius underlines the adulation of the pirates through a hyperbolic reference to what could be described as a reverse catasterism – here stars descend, placing themselves among human beings (‘If stars were to come down from the sky and offer themselves to men, they would receive no greater reverence’; delapsa caelo sidera hominibus si se offerant, venerationis amplius non recipient). As Scipio Africanus takes pride of place as one of Valerius’ most favoured exemplars, the implied comparison could even suggest that Scipio himself is deserving of catasterism – an idea already signalled in the Somnium Scipionis of Cicero’s De Re Publica.20 Similarly, in an exemplum from Book 8 which details the Roman peoples’ support of Q. Lutatius Catulus, the substance of which is taken from Cicero, Valerius’ own introductory segue into the example is to comment that Catulus was ‘almost borne to the stars by the voice of the Roman people’.21 In both instances, Valerius displays beliefs consistent with contemporary thought about astral immortality. But the most significant places within Valerius’ work where his language celebrates sidereal elements are passages which deal with the imperial family.22 Valerius’ ‘sacralizing’ language, in these instances, should be read in relation to the growth of the imperial cult during the later years of Augustus’ reign and subsequently also under Tiberius, and reflects a shift in what could be said about the emperor in prose from the earlier Augustan period, reflecting not only the development of a ruler cult, but also how the role of the emperor was perceived. From as early on as the preface, Valerius refers to Caesar as a star, along with Augustus, intimating also Tiberius’ own present and future celestial divinity.23 Later, at 2.1.10, Valerius alludes more generally to the Caesars as divi – the brightest part of the heavens (caeli clarissima pars). At 3.2.19, in a chapter on bravery (forti- tudo), Valerius refers specifically to Caesar as a man once ‘superior in military and civilian life, but now the bright splendour of the stars, the god Julius’ (sed ut armorum et togae superius, nunc etiam siderum clarum decus, divum Iulium). Wardle is overly cautious about whether this description reveals Valerius’ belief 110 Jeffrey Murray in Caesar’s catasterism; he notes that the reference ‘need mean no more than that Caesar was in heaven after his death’.24 The comet that had appeared in the sky in July 44 bc during his funeral games was exploited by Octavian as proof of Caesar’s apotheosis – a belief taken up by many of Valerius’ contemporaries.25 In the climactic exemplum within a chapter illustrating changes in fortune, Valerius refers again specifically to Caesar who, through his virtues, we are told, secured his path to heaven (6.9.15). The exemplum itself recounts an incident from Caesar’s youth in which, having been captured by pirates, he ransomed himself for fifty talents. Capturing the pirates in turn, he had them crucified. Valerius’ explanation – that if Fortune did not spare even the brightest star in the universe, whose divinity was the same as her own, mere mortals should hardly hope for more – continues the sidereal vocabulary where Caesar is concerned, concluding that he vindicated himself by his ‘celestial power’ (caeleste numen). Augustus receives less frequent stellar treatment from Valerius.26 One possible reason for this is that opportunistic use of celestial phenomena, such as that which Augustus made in 44 bc in order to link himself to his adopted father, was no longer necessary – nor, indeed, could he guarantee a repeat performance – and hence his own astrological propaganda was not as pronounced.27 Augustus was formally deified by the Senate in ad 14, and both Suetonius (Aug. 100.4) and Cassius Dio (56.46.2) relate that a former praetor had been found to attest to his ascent to heaven. In Valerius, outside of the direct reference in the preface itself, only oblique allusions are made to Augustus’ celestial position (e.g. 1.7.1; 7.7.4), usually only to indicate his posthumous divinity.28 Tiberius, in turn, as we have already seen in the preface, is said by Valerius to equal the stars of his father and grandfather in their divinity. In his typical grandi- ose rhetoric, Valerius refers to Tiberius as the ‘surest salvation of the fatherland’ (certissima salus patriae), whose ‘celestial providence’ (caelesti providentia) enables him to act as arbiter over morality and the virtues and vices that Valerius will go on to discuss. Elsewhere, in an elaborate exemplum, Valerius recounts Tiberius’ dash to his dying brother’s side (5.5.3). In concluding remarks to this example of fraternal piety, Valerius compares Tiberius and Drusus to Castor and Pollux – the Greek mythological brothers who were turned by Zeus into the con- stellation Gemini, and were also later widely celebrated at Rome.29 Here Valerius is very much cued into contemporary associations between the heavenly twins and the imperial brothers. Tiberius’ shrewd politicking, in associating himself and his brother with the pair, has been carefully mapped out in a recent article by Edward Champlin.30 It is surely not coincidental, then, that elsewhere (4.6.ext.3) Valerius again refers to the ‘noble fame’ of Castor and Pollux, the ‘pair of brothers destined for the stars’ (nobilis famae … destinatum sideribus par fratrum). A final example on Tiberius comes from 9.11.ext.4, the climactic exemplum in a chapter illustrating outrageous words and criminal deeds (Dicta Improba aut Facta Scelerata). The passage itself is a central piece of the internal evidence in debates regarding the date of the work. If the unnamed villain of the exemplum is indeed Lucius Aelius Sejanus, as the current consensus holds, and not M. Scribonius Libo Drusus, or some other unknown conspirator, as some have Valerius Maximus and the language of stars 111 argued, then a date after ad 31 is necessary for the completion of the work, and not the early years of Tiberius’ reign (ad 14–16).31 In language reminiscent of the preface, the exemplum celebrates Tiberius’ leadership and preservation of the commonwealth in the face of a potential conspiracy. The Roman Empire, Valerius tells his readers, was safeguarded by Tiberius’ present divine power (praesenti numine). Although numen could be understood more generally as the deity worshipped at each site (e.g. at the altars, the sacred couches or the temples), here it is more likely a reference to Tiberius, as scholars have noted, given the exemplum’s parallels with the preface, where Tiberius’ divinity is afforded by praesens fides (‘present certainty’).32 In conjunction with this, Valerius maintains that the eyes of the gods (oculi deorum) were vigilant and that the stars retained their strength. Oculi deorum, given its use elsewhere in the work, specifically at 4.3.3, where Valerius refers directly to Augustus and Tiberius as the ‘two divine eyes of the commonwealth’ (duobus rei publicae divinis oculis), is likely to have wider application to Divus Julius and Divus Augustus here rather than to the state only. What remains to be explained is Valerius’ phrase sidera suum vigorem obtinuerunt – ‘the stars maintained their potency’ in Shackleton Bailey’s translation. Elsewhere Valerius refers to Caesar and Augustus specifically as stars, and they appear to be the most natural interpretation here. But the possibility remains open that Valerius could be referring more broadly to astrology – that is, Tiberius’ nativity stood, and he was not destined to die in ad 31. Jane Bellemore, although incorrect in her identification of the conspirator as Libo Drusus, certainly suggests that astrology more generally could be meant; similarly, although he admits that this sort of speculation ‘belongs rather to a historical novel’, Anthony Birley suggests the possibility that Thrasyllus was behind Tiberius’ change of mind regarding Sejanus.33 Valerius, then, although he says nothing that supports the predictive role of celestial bodies, does indeed demonstrate an enthusiasm for the language of stars – albeit confined mainly to passages of his work that deal with the imperial family and contemporary events. In this way, he is seen to be an exemplar of his period.34 While he nowhere extends the scope of accepted divination, he celebrates the extended pantheon of Julio-Claudian Rome, and his work, although ‘backward looking’ and ‘anachronistic’ in parts, does reflect current concerns. The Facta et Dicta Memorabilia thus provides a niche vantage point from which to view these somewhat controversial aspects of Roman religion and politics in the early empire.

Notes 1 val. Max. 1.3.3. For text and translation of Valerius Maximus throughout I have used Shackleton Bailey’s Loeb edition of 2000, occasionally modified. The lacuna occurs from 1.1ext.5 until 1.4.ext.1. See Wardle (1998) 18. 2 For detailed discussion of this exemplum, see Wardle (1998) 148–51. An epitaph in Rome gives the name of the praetor as Cn. Cornelius Scipio Hispanus. 3 On Nepotinaus’ peregrina scientia, cf. Cic. Balb. 55.8. 4 Cramer (1954) 233–5; Wardle (1998) 149. Divination for the sake of financial gain, however, is presented pejoratively: Cic. Div. 1.132 with comments by Nice (2001) 165 and Wardle (2006) 421. 112 Jeffrey Murray 5 Orlin (2010) 182–4. 6 Orlin (2010) 184. 7 Herbert-Brown (2002) 120–1; Green (2014) 66 n. 4; cf. Ripat (2011) 118–20. Wardle (1998: 149) also comments on the exemplum’s possible contemporary reference. 8 Livy is one of Valerius’ major sources (along with Cicero, Varro and Pompeius Trogus). See Bloomer (1992) 59–146; Wardle (1998) 15–18. Elsewhere, Valerius varies his vocabulary for astrologers; at 5.7.ext.1 he labels a certain Leptines as mathematicus. Valerius is the only ancient author to mention this figure (see RE XII.2407), and it is likely that he has taken over the choice of term from his source, possibly Pompeius Trogus in this instance. 9 On the thorny issue of Livy’s religious belief and scepticism, see Levene (1993). 10 Ripat (2011) 115–54. 11 Ripat (2011) 126. 12 Accounts of the rise of astrology at Rome are found in Cramer (1954); Volk (2009) 127–37; Green (2014). 13 For example, see Swan (2004) 280–1. 14 Nice (1999) 101. 15 Mueller (2002) 223–4 n. 178. 16 valerius’ treatment of Roman religion in Book 1 is in the first instance determined by the preface he adapts from Cicero (Har. Resp. 18): 1.1.1a is, as Ryan (unpublished) now rightly emphasizes, the preface to the following analysis, although Cicero does not credit astrology anywhere in his written work. While Valerius does include a section on dreams (notably after those on the types of divination accepted within Roman state religion), it is noticeably thin on Roman examples that would have been accepted. 17 See e.g. Volk (2009) 14–15: ‘Astronomy and astrology were often practised by the same person … As a matter of fact, our terminological distinction between “astronomy” and “astrology” is largely absent in antiquity. The Greeks and Romans for the most part used both astronomia and astrologia to refer to the “study of the stars” in general, without regard to whether astrological (in the modern sense) presuppositions were involved. If anything, astrologia was the more general and widespread term, while astronomia, favoured by the Platonic school, may on occasion have had somewhat more esoteric connotations’. See also Cramer (1954) 3–4. 18 Langlands (2011) 100–12. 19 See Mueller (2002: 104–5) for further discussion of this exemplum. 20 Cumont (1912: 180) writes: ‘Another doctrine was also taught, that the divine souls of sovereigns come from a higher place than those of other men, that the greater a man’s dignity, the greater is the dower he gets from heaven. But, in a general way, the rites employed to ensure immortality to kings by putting them on a level with the gods, were by degrees extended to important members of their entourage. This was a sort of privilege, of posthumous nobility, which was conferred on great ministers of state, or which they usurped, long before the common crowd of the dead attained it’. For the special place afforded to Scipio Africanus the Elder in Valerius, see Bloomer (1992) 150. 21 val. Max. 8.15.9: Q. etiam Catulum populus Romanus voce sua tantum non ad sidera usque evexit: nam cum ab eo pro rostris interrogaretur, si uno Pompeio Magno omnia reponere perseverasset, absumpto illo subiti casus incursu in quo spem esset habiturus, summo consensu adclamavit ‘in te’. vim horati iudicii admirabilem, si quidem Magnum Pompeium cum omnibus ornamentis, quae rettuli duarum syllabarum spatio inclusum Catulo aequavit! Cf. Cic. Leg. Man. 59: Reliquum est ut de Q. Catuli auctoritate et sententia dicendum esse videatur. Qui cum ex vobis quaereret, si in uno Cn. Pompeio omnia poneretis, si quid eo factum esset, in quo spem essetis habituri, cepit magnum suae virtutis fructum ac dignitatis, cum omnes una prope voce in [eo] ipso vos spem habituros esse dixistis. See also, Sall. Hist. 5.24M [5.20McG]; Vell. 2.32.1; Plut. Pomp. 25.5; Dio 36.36a. Valerius Maximus and the language of stars 113 22 Wardle (2000: 479) makes a similar point, noting that although the imperial family do not feature frequently within the Facta (in comparison, for example, with an exemplar such as Marius), nonetheless the passages in which they do appear are significant quali- tatively, as ‘direct or indirect panegyric of the imperial house appears at key structural points in the work, which gives it a greater prominence’. 23 1.praef.: ‘My petty self shall betake me to your goodwill all the more properly in that other divinity is inferred by opinion whereas yours is seen by present certainty as equal to the star of your father and grandfather, through whose peerless radiance much far-famed lustre has accrued to our ceremonies. For other gods we have received, the Caesars we have bestowed’; mea parvitas eo iustius ad favorem tuum decucurrerit, quo cetera divinitas opinione colligitur, tua praesenti fide paterno avitoque sideri par videtur, quorum eximio fulgore multum caerimoniis nostris inclutae claritatis accessit: reliquos enim deos accepimus, Caesares dedimus. 24 Wardle (1997) 342 n. 95. 25 See Ramsey and Licht (1997); Pandey (2013) 405–49. 26 On Augustus’ treatment more generally, see Wardle (2000) 483–9. 27 I owe this point to David Wardle. 28 Cf. 9.15.2. 29 See Wardle (2002: 433–40) for detailed discussion of this exemplum. 30 For the mythological pair’s popularity at Rome, as well as their use in Tiberian propaganda, see Champlin (2011) 73–99. 31 On the dating of the Facta et Dicta Memorabilia, see Carter (1975) 30–3; Bellemore (1989) 67–80; Briscoe (1993) 398–404; Combès (1995) 8–11; Wardle (1998) 1–6; Shackleton Bailey (2000) 1–3; Themann-Steinke (2008) 17–28; Briscoe (2010) 380–1. 32 Wardle (2000) 491. 33 Bellemore (1989) 79; Birley (2007) 147–8. 34 valerius could be included among the ‘shared intellectual space’ of the Roman authors of the period such as Manilius, Germanicus and Ovid, who all seek to find an ‘appropriate Roman means of celebrating the power of stars’. See Green (2014) 190. 9 ‘Arrows fletched from our own wings’ Discovering a ‘Delphi of the mind’ in the writings of the Early Church Fathers1

Daniel Crosby

Introduction Theodoret of Cyrus, a Church historian of the fifth century,2 provides a provocative historical anecdote from the previous century:

And first Emperor Julian forbade the sons of the Galileans (for that is what he named the worshippers of the Savior) to take part in the words of the poets, and rhetors, and philosophers.3 For he said, according to the proverb, ‘We are shot by arrows fletched from our own wings.’ For arming themselves from our own writings, they make war against us.4

The statement indicates that some pagans of the fourth century believed that the Christian apologists implemented the arguments of certain earlier poets, rhetors, and philosophers, who were sometimes skeptical of the traditional Graeco-Roman religion and at other times unaware of the potential implication of their words, as weapons against paganism. The sentiment rings true even if Julian never spoke these words. In the mid-fourth century, paganism was under attack and the Early Church Fathers were leading the intellectual front of the assault. One important maneuver in Christian polemics was to discredit the authority of the Delphic Oracle.5 In this contribution, I first discuss five particularly salient arguments that the Early Church Fathers leveled against the Oracle: the ambiguity of the oracles, the daemonic source of its inspiration, the gender of the Pythia, the Christian overtones of certain oracles of Delphi, and the clear signs of the serious decline (even the end) of its importance as a functioning oracle. I then show that each of these arguments has clear antecedents among skeptical philosophers, thus demonstrating that the Early Church Fathers are working with a received tradition, just as Julian’s complaint suggests. Finally, drawing on social constructionist theory, I suggest that their use of Delphic tradition to discredit the Delphic Oracle, an institution they claim is no longer operating, signals the existence of two different realities of the Oracle of Delphi: a ‘Delphi of the mind’ and the ‘Delphi of fact.’6 The course of scholarship on the Delphic Oracle has swung rather pendulum-like over the last century. During the excavations begun by the French Archaeological School in 1893, it was revealed that no chasm existed beneath ‘Arrows fletched from our own wings’ 115 the temple of Apollo, contrary to what some of our ancient literary sources say.7 This dramatic refutation of an established tradition began the swing of the Delphic pendulum between trends of harmonization, which endeavored to correlate the ever-growing wealth of archaeological evidence with the literary sources, and skepticism, which sought to disregard any textual witness that did not agree with the archaeological evidence.8 For example, one faction of scholars suggested that the frequent earthquakes, to which Delphi is prone, were responsible for erasing all trace of this chasm, fissure, or slight depression of the pavement of the temple.9 Another faction preferred to scrutinize the literary sources rather than the expertise of the French archaeologists, and in the process, they discredited the chasm as an imaginative embellishment or a traditional misunderstanding.10 With the chasm all but closed, natural, geological explanations of the Pythia’s supposed prophetic frenzy declined in favor of more synthetic, psychedelic, and psychological ones.11 More recently, an interdisciplinary team reexamined the geomor- phology of the area around Delphi and determined that the temple of Apollo may sit atop an intersection of two faults, which reopened the cleft, however far, to the possibility of noxious subterranean vapors in the form of ethylene gas and offered some explanation for the ancient belief in a chasm below the temple.12 However, the most recent scholarship has demonstrated the serious flaws in that theory, from the evidence for the location of the faults, to the volume of ethylene needed to induce the Pythia’s supposed ‘trance-like state,’ to the methodology involved in the usage of historical sources.13 The pendulum continues to swing.14 At their core, all of these studies share a positivist approach to the study of the Delphic Oracle in which archaeological and textual evidence are used, whether affirmed or dismissed, to support related to the existence of certain physical features and the historicity of certain practices and events. In this contribution, I will explore evidence that may not reflect the historical reality but, nonetheless, reflects a historical reality, namely the historical perspective of the Early Church Fathers. With regard to scholarship on the Delphic Oracle, the most often dismissed class of evidence is the writings of the Early Church Fathers, whose obvious bias has hampered their credibility and, as a consequence, their evidentiary weight and reception.15 Although perhaps factually inaccurate on a number of counts, their writings can still be important sources for how certain people thought about the Delphic Oracle, and to that effect, they provide good evidence. The apologetics of the Early Church Fathers argued not only in favor of Christianity, but also against many traditional forms of Graeco-Roman religious experience and expression. Among these institutions, the Delphic Oracle was of particular interest to them. The Early Church Fathers singled out the Oracle for fierce attacks; at the same time, they proclaimed that the issuance of oracles had been discontinued. To quote Parke and Wormell, although the apologists understood that the Oracle was no longer performing the function for which it was best known,

The tradition of [the oracles’] prophetic powers continued to remain as an influence in the minds of men. The Christian apologists think it worth their 116 Daniel Crosby trouble to devote much space in their writings to refuting the idea that the oracular powers of Delphi were in any way a confirmation of pagan belief.16

Their polemics against the Oracle reveal a disconnection between a realistic appreciation of the threat posed by the Oracle of the third and fourth centuries and the profound thoroughness of their response. The Early Church Fathers seem to have felt that the oracle of Delphi was still a threat, despite the fact that they thought it was no longer issuing oracles. For this reason, the attention they paid to Delphi raises a question about what the Delphic Oracle was and what it meant to them. Was the Oracle that the Early Church Fathers railed against different from the institution that existed in antiquity? I propose to answer this question by defining reality differently from traditional scholarship on the Delphic Oracle with its emphasis on strict historicity. Instead, I adopt a social-constructionist definition of reality, by which reality is understood as a ‘taken-for-granted’ body of knowledge, ‘regardless of the ultimate validity or invalidity (by whatever criteria) of such “knowledge,”’ that is ‘developed, transmitted and maintained in social situations.’17 This reality is one that comes into being through the social interaction involved in communication, particularly through language.18 So, while the positivist and empiricist approach studies the reality of objects, the social constructionist approach studies the reality of how objects are thought about in a society and described in language. The theory has produced influential and paradigm-shifting scholarship since the 1960s, such as Benedict Anderson’s classic study on the concept of nationalism and Edward Said’s critique of the Western construction of ‘Orientalism,’19 but only relatively recently have ideas of social constructionism found their way into classical scholarship. It has become more acceptable to study the realities which the ancients constructed for themselves and their underlying motivations for doing so. Zeitlin, for example, states:

In proposing that there is some conceptual category in the Athenian theater named ‘Thebes’ and that some underlying ‘unity of place’ organizes these disparate stories and their treatment in the work of all three Athenian tragic poets, I am, in effect, suggesting that we look at Thebes as a topos in both senses of the word: as a designated place, a geographical locale, and figura- tively, as a recurrent concept or formula, or what we call a ‘commonplace.’20

Here, Zeitlin parses the reality of Thebes the polis which, in one reality, stood about fifty kilometers distant from Athens and the ‘Thebes’ of the stage which the tragedians constructed as an ‘anti-Athens’ or Athens’ ‘mirror opposite.’21 Also, Berman has shown in a recently published volume that the Theban landscape was constructed from elements of the Boeotian polis’ physical topography and buildings, as well as from elements found in the myth and literature of the late-archaic and classical periods that do not reflect the historical polis.22 Thebes, then, is a composite with at least two facets to its reality in the ancient world: one grounded in a geographical location and another that was imagined. ‘Arrows fletched from our own wings’ 117 I contend that the lens of social construction can be a useful one for studying the Delphic Oracle, an institution and place that, like Thebes, is prominent in both myth and history and often stood at a remove from those who wrote about it. Through this lens, the misunderstandings, fabrications, and falsehoods that the Early Church Fathers wrote about the Delphic Oracle may fairly be considered reality, provided that it can be shown that their statements were socially transmitted and maintained as part of a tradition and not merely their own fictions. First, therefore, it will be necessary to elucidate and comment on the assault made by the Early Church Fathers against the Oracle of Delphi, showing not only that they felt compelled to change the perception of the most honored institution of the Graeco-Roman world, but also that they borrowed extensively from the pagan tradition of philosophical skepticism to effect that result. Second, the fact that the Early Church Fathers admit that Delphi no longer existed as an oracular institution, and yet wage a war against it, indicates that they were concerned not with the institution, but with its reception among their contemporaries. This evidence signals their recognition of two different, but real, Oracles of Delphi: the one being the declined institution, the ‘Delphi of fact,’ and the other, the constructed reality, which continued to exercise a great amount of influence on pagan religious thought during the third and fourth centuries, a ‘Delphi of the mind.’

The debts of our fathers The Early Church Fathers were certainly not alone in their stance against traditional Graeco-Roman religion. That ground had already been well trodden by earlier philosophers. Thus, in many instances, the apologists did not need to invent new controversies, but simply to dredge up old ones and occasionally to present them in a new package wrapped in Christian morals and teachings. In the case of the Oracle of Delphi, the debts of the Fathers are clear. In what follows, I discuss four particular arguments of the five mentioned above as representative of that debt – these are the ambiguity of the oracles, the daemonic source of their inspiration, the gender of the Pythia, and the Christian message of certain oracles that Delphi was thought to have issued – and I demonstrate the similarities they show to earlier literary tradition in each case. The ambiguity of its oracles was the most popular argument against Delphi’s authority. Eusebius, the fourth-century bishop of Caesarea, believed that the oracles of the Greeks were ‘extremely well designed for deceit, and being composed in an equivocal and ambiguous manner, they are fit, not without skill, for either of the outcomes expected from the event.’23 To demonstrate his argument, Eusebius claimed that the Delphic Oracle was responsible for the downfall of its own faith- ful patrons, offering as an example the well-known tale of Croesus. In this story, Croesus, who had lavished Pythian Apollo with his most exquisite dedications, was told by the Oracle only that a great empire would fall if he should cross the River Halys. Believing the great empire to be that of the Persians and not his own, he attacked.24 It was his that fell. Clement of Alexandria in the late second century was also aware of this ironic and unfortunate episode: ‘He [Apollo] betrayed his friend 118 Daniel Crosby Croesus, and having forgotten the reward he had received (for thus was he a lover of deceit), led Croesus across the River Halys to the funeral pyre.’25 For the Early Church Fathers, Pythian Apollo was a traitor even to his most beneficent patrons. The argument made by both Clement and Eusebius was definitely not original. Oracular institutions, and the Delphic Oracle in particular, seem to have had a reputation for ambiguity going back perhaps as far as the philosopher Heraclitus, who famously said: ‘The lord whose oracle is in Delphi neither tells, nor conceals, but gives a sign.’26 Cicero also records the example of Croesus and adds about the oracles in general that ‘Some were so intricate and obscure that their interpreter would need an interpreter and the oracles themselves must be referred back to the oracle; and some are so equivocal that they require a dialectician to construe them.’27 According to Dio Chrysostom, writing in the early second century, Diogenes the Cynic warned a man traveling to Delphi to be cautious of the response the Pythia would give him, lest he unwittingly cross his own Halys.28 Delphic ambiguity was such a well-known topic in the Greek world that Lucian could even mock Apollo, whom he portrays as hedging his bets in his prophecies, for the amusement of his audience.29 The Early Church Fathers also challenged the nature of the source of the Pythia’s oracular inspiration. In other words, the apologists were able to cast a measure of suspicion on the Oracle by redefining the nature of and relationship between Apollo and the daemons. The Greek understanding of the cosmological order of δαίμονες seems to have evolved much over time, but we need only treat one tenet here. That pertinent element of daemonology stems from the classical period in Plato’s supposition that daemons were intermediaries through which the gods and mortals interacted, bridging the gap between the opposing realms of mortal and immortal, which could not intermingle without pollution.30 Therefore, daemons could be made responsible for relaying prophecy from Apollo, the prophetic god, to the Pythia, his oracular priestess. Xenocrates, the successor of the Academy after Speusippus, seems to have been the first to open the door to what would become the more common interpretation of the daemonic in Christian thought: malevolent spirits. He submitted that daemons, being between the realms of gods and mortals, shared in both divine power and human passions, and because of these passions, they varied in virtue and vice.31 Pushing his theories further, Porphyry and Theophrastus, one of the immediate successors of Aristotle whom Porphyry quotes extensively, believed that all sacrifice was defilement, an unholy and shameful act, in the sight of the gods and worthy only of the daemons: ‘The one who ponders reverence toward the gods knows that no animate thing is sacrificed to the gods, but to the daemons either good or bad.’32 Some Greek philosophers, then, not only argued that there was a distinction between good and evil daemons, but also that all sacrifice was an element of daemonic worship. The Early Church Fathers furthered the argument of these philosophers, taking it to its ultimate conclusion. Eusebius says:

Porphyry, relying on Theophrastus as his witness, says that animal sacrifice is not fitting for the gods, but only for daemons, such that according to the argument of himself and Theophrastus, Apollo is a daemon but not a god.33 ‘Arrows fletched from our own wings’ 119 Eusebius indicts Apollo especially, whose oracular pronouncements endorsed sacrifice to the gods, and only subsequently extends his argument to the whole pantheon.34 However, Clement of Alexandria and Eusebius were not content to prove that the Graeco-Roman gods were merely daemons; they desired to prove that the gods, and Apollo in particular, were deceitful, misanthropic daemons, in order to discredit the Delphic Oracle. They accomplished this task through various means, but the most powerful argument used was an appeal to commonplace ethics:35

For if the philosophers considered the sacrifice of irrational animals to be accursed and sacrilegious, abominable, defiled, unjust, unholy, not harmless to those who are sacrificing, and on account of all these reasons unworthy of the gods, what ought to be thought about the sacrifice of people?36

Both Clement and Eusebius go through a litany of examples of this practice, and thus the Graeco-Roman gods are consigned to the ranks of the evil daemons by the common perception that human sacrifice is immoral:

But if anyone should say that the custom of human sacrifice is not evil, but was most properly practiced by the ancients, he must condemn all those of the present day, because no one worships like their fathers.37

Apollo’s Oracle at Delphi could not have been truthful because its source was evil. In this instance again, we see that the Early Church Fathers employed a pagan tradition, specifically the theology that developed out of the pagan philosophical schools of the classical and Hellenistic periods, to assault the basis for the Oracle’s authority. Another method for discrediting the Delphic Oracle was through an appeal to overt sexism in their contemporary culture. The common thread of Greek male chauvinism is clearly seen in Plato and in the works of other philosophers: women are, by their very nature, inferior to men according to the Greeks.38 The argument appears most clearly in the writings of the third-century theologian Origen:

If the Delphic Apollo were a god, as the Greeks think, whom is it more necessary for him to choose as a prophet than a wise man, or, if one could not be found, at least one who would be inclined in that direction? How would he not wish that a man prophesy rather than a woman?39

Origen plays to the sexist mindset that was prevalent in Greek society by questioning the tradition of Apollo’s priestess. If we can agree that even a wisdom-inclined man is superior to a wise woman, then why is the Pythia a she and not a he? The answer adduced by the Church Fathers would have appeared, at first glance, to be a shock to the reverent Greek, for they unabashedly attributed to the mantic session the sensational and the burlesque. Origen believed that Apollo’s choice of prophet was dictated by his sexual appetites: ‘sitting at the mouth of the Castalian 120 Daniel Crosby cave, the prophetess receives a spirit through her genitals; being filled with this, she utters the holy statutes and divine oracles,’ and ‘the pure mantic spirit, Apollo, slips from the body of earth into the so-called prophetess through her genitals as she sits at the Pythian cave.’40 In one sense, his description may conjure up images of the treatment of certain gynecological diseases through vaginal fumigation (θυμίασις / ὑποθυμίασις);41 in another sense, Origen is representing the inspiration of the Pythia as the result of sexual congress between the god and his priestess. Origen was not alone in this understanding. John Chrysostom paints a similar picture, adding that the effect of the prophetic session produced an odd symptom:

This very Pythia, being a woman, is said to sit at times with her legs spread upon the tripod of Apollo. Thus, the evil spirit rising up from below and issuing itself through her genitals fills the woman with madness, and she, letting down her hair further, then raves like a bacchant and spews foam from her mouth, and being thus in a drunken state, she utters the words of her madness.42

Even Clement of Alexandria may have had this understanding of the origin of the prophetic power when he wrote of seers in general that ‘some were stirred by demons, or were disturbed by waters, and fumigations [θυμιαμάτων], and air of a certain sort.’43 In addition to the ability to pronounce oracles, the spirit made her wild and crazed.44 Thus, the Church Fathers could draw a clear dichotomy: the Hebrew prophets were superior not just because they were men, but also because they prophesied in a sober state of mind and under the inspiration and power of God.45 The story that an exhalation rose up from below the Pythia was not a new one. Strabo, Plutarch, and Pausanias were aware of the tradition of a prophetic πνεῦμα at Delphi.46 There is even some indication that the obscene imagery of the Pythia employed by the Christian apologists was not without precedent. Certainly, the position of the tripod over the supposed source of the πνεῦμα is suggestive of vaginal fumigation, in and of itself, not to mention the fact that the symptoms of certain conditions in women, for which the prescribed treatment might be fumigation, included bacchant-like behavior popular in later characterization of the Pythia.47 Plutarch seems to be the first to give us a more base account of Delphic consultation in a treatise called On the Delay of Divine Vengeance: ‘At the same time, he, going forward, tried to show him the light that, as he said, came from the tripod, and passing through Themis’ vagina [διὰ τῶν κόλπων], came to rest upon Parnassus.’48 As we have seen in Eusebius above, the word κόλπος can also mean vagina or womb, so it is clear that whatever Plutarch actually meant, the text could be read with an eye for scandal.49 Additionally, Pseudo-Longinus says:

The account holds that the Pythia approaches the tripod where there is, as they say, a chasm in the earth that emits an inspired vapor, and sitting in that place and being impregnated by the daemonic power, she immediately chants oracles according to her inspiration.50 ‘Arrows fletched from our own wings’ 121 The Pythia is impregnated by the prophetic force, and their offspring is the oracular pronouncement. Although the evidence is slight, the fact that any evidence exists at all demonstrates that this idea was at least extant before the time of the Early Church Fathers.51 Further, that the Christian Fathers were among our earliest sources should not surprise us; they could hardly have kept themselves from pointing out such a salacious fact, while the reverent or squeamish might have preferred cleaner versions. The Christian apologists were also able to demonstrate the superiority of Christianity over paganism by calling upon the Delphic Oracle as a witness. Tradition held that the Pythia undermined her own authority with her responses on occasion. We are told by Eusebius that when the Nicaeans went to inquire at the Oracle about a certain sacrifice to Apollo, the Pythia said:

It is not possible to restore the babbling Pythian voice, for already having become weak by great lengths of time, it has turned the bolt of un-prophetic silence. But pay to Phoebus the prophetic offerings as is the custom.52

A similar declaration is made in an oracle reported by Philostorgius, an Arrian church historian writing in the late fourth and early fifth century. The Emperor Julian’s emissary to the Oracle was told:

Tell the emperor that the cunningly wrought hall has fallen to the ground. No longer does Phoebus possess his pool, nor his mantic laurel, nor the talking spring. Even the babbling water has vanished.53

It is clear from the Pythia’s own admission that there is no oracular power left at Delphi! The decline of the Oracle, however, had begun much earlier, and the Pythia had foreseen that the end would come. According to Eusebius, Augustus asked the Oracle of Delphi who would rule after him. The response he received was a ‘Hebrew boy, a god who rules among the blessed, commands me leave this house and go back to Hades. Now then, leave our altar in silence!’54 In the same words, the Pythia proclaims the eventual victory of Christianity and consigns herself to damnation! This was not the only oracle to speak of Christ; writing in the eighth century, John of Euboea provides another example:

After a long time one will come to this much-divided earth, and he will become flesh separated from the Fall. He will free it from the corruption of incurable desires by the infinite limits of his divinity. He will become hated by an unbelieving people. He will be hanged on high as one condemned to death. He will willingly submit to suffering these things, but having died, he will rise to eternal life.55

That these oracles are forgeries should at least be suspected, but it would be an excessively uncharitable reading to suggest that the Church Fathers fabricated them themselves. Rather, it seems more likely that some of these oracles, perhaps 122 Daniel Crosby produced and fathered upon Delphi by zealous Christians and gathered into collections by oracle-mongers, came to be considered part of the genuine tradition of Delphic oracles, such that the apologists had no reason to scrutinize them and every reason to make use of them.56 Julian’s lament that early Christian apologists shot ‘arrows fletched from our own wings’ was, indeed, an accurate metaphor. We can clearly observe how significant those borrowings were in this case study on the Delphic Oracle, and we ought not to be surprised. Although the use of Delphic literature to discredit Delphic literature has an inherent irony, it was through this literature that the argument could be made most convincingly. It is as Faustus the Manichean of Augustine’s epoch suggested: a pagan was more inclined to be persuaded by arguments that were working within a tradition with which he was familiar (like oracles, oracular literature, and philosophy) than one with which he was not (like the Hebrew prophets).57

The ‘Delphi of the mind’ As I argued in the introduction, trying to reconcile or differentiate the archaeological and the literary evidence is not the only means of investigating what Delphi was in reality. One scholar who acknowledges this fact has attempted to view the issue from a different angle. Roger Lipsey proposed the existence of two Oracles of Delphi:

From the beginning there were two Delphis: the Delphi of fact and the Delphi of the mind. Throughout ancient times, the two followed generally parallel courses, linked to one another by pathways however slim, yet likely to diverge because Delphi lived richly in the minds of men and women, most of whom would never visit the holy city or enjoy the privilege of inquiring there. Delphi drew to itself, as if magnetically, both the refined inventions of poets, orators, and playwrights and the rough storytelling of the common people. Even when historical figures about whom we know a great deal visited Delphi – for example, the model of Roman virtue, Cicero – the recorded results seem to swerve gaily from the verifiable realm of history to the imaginative realm of story, which may or may not be wholly ‘true.’58

The point of Lipsey’s book is to revive the philosophical ‘Delphi of the mind,’ and to that effect, he provides commentary on a number of popular and thought- provoking Delphic tales.59 The lore and commentary he presents, however, do not adequately show the ‘Delphi of the mind,’ as an abstracted, constructed reality existed in the minds of the ancients, only that its elements may at times be observed by the modern mind in these stories. It seems a far more important argument to Delphic scholarship to demonstrate that the ‘Delphi of the mind’ was just as real as the ‘Delphi of fact’ in ancient times. The best place to find the effect of the ‘Delphi of the mind’ is where the two Oracles of Delphi diverge most significantly: during the decline in its operations ‘Arrows fletched from our own wings’ 123 and significance. The Early Church Fathers were quite cognizant of a decline in the Oracle at Delphi. Clement of Alexandria, Eusebius, and Gregory of Nazianzus, the fourth-century bishop of Constantinople, all declared that the prophetic spring was dried up.60 Parke, Wormell, and Scott are quick to point out that the celebra- tions of Christians like Clement of Alexandria were ‘premature,’61 but it was a fairly well-acknowledged fact among pagan authors from the first century bce onward that the Oracle of Delphi had undergone a serious decline that required explanation.62 Plutarch, one of the priests of Apollo at Delphi, could discuss a decline in the importance of the Delphic Oracle in a treatise that came to be called On the Obsolescence of Oracles. This decline was evident even earlier, in Cicero’s time, who wrote: ‘Now, therefore, its glory is diminished …’ and, more damn- ingly, ‘Why are Delphic oracles not issued in this very way at the present time and have not been for a long time, such that nothing now is viewed with more disdain?’63 It is clear that by the first century bce the effects of numerous destructions by earthquakes and enemies, the depopulation of Greece, the decentralization of Delphi in the Hellenistic period and under the Roman Empire, and the developing trend of skepticism in philosophy, to name but a few reasons, left behind a ‘Delphi of fact’ that was noticeably less significant, despite several attempts to revitalize the institution.64 Further evidence of decline is found in the corpus of Delphic oracles. In Fontenrose’s catalogue of Delphic responses, the 343 responses include only seven that he classifies as historical or quasi-historical and which are datable to a time after 200, and he only considers one of these, an oracle from the mid-third century, to be genuine.65 The issuance of oracles at Delphi might have continued now and then, but the fact that only three genuine oracles, even by the most liberal count,66 survive from a period spanning two centuries is itself suggestive of the status of Delphic oracles in literature in the late imperial period and, by extension, the perception of the Delphic Oracle in the same epoch. Thus, the decline of the Delphic Oracle was a well-established fact by the time of the Early Church Fathers, but the fact of its decline did not stop them from casting a dark shadow on the legacy of the Oracle or from stripping the Pythia naked. Why would the Early Church Fathers fire a full quiver of shafts at the Delphic Oracle when, by their own admission, the prophetic institution had no further function in pagan religion during the empire? The evidence indicates that the Early Church Fathers felt compelled to respond in their apologetics not to the declined institution at Delphi, the ‘Delphi of fact,’ but to the collective, literary tradition of Delphi, the ‘Delphi of the mind,’ which had set roots so deep in the Graeco-Roman mind that they outlasted the tree. The struggle of the apologists was truly ‘not against flesh and blood’; it was against a construct that was every bit as real in their contemporary culture. Nevertheless, their claim of the status of the ‘Delphi of fact’ demonstrates that the Early Church Fathers recognized a distinction between the physical reality and the constructed reality. Only the latter merited serious concern and attention in their writings. Given the shortage of evidence on which to base conclusions about the practices of the Oracle of Delphi in antiquity and the necessarily speculative character of the research that investigates the Oracle from a positivist angle,67 there are at 124 Daniel Crosby least two benefits that recommend a social constructionist approach in this sub- field. The first is that it opens the conversation of scholarship to considerations of the processes by which the construct of the Delphic Oracle came to be formed and the importance of that construct to ancient perceptions of the reality of the Oracle.68 The parameters of this contribution prohibit more than a suggestion of the manner in which this construct may have come about generally. I therefore submit that as the romance of the Oracle captured the hearts and minds of the ancients, material accreted to its tradition, spoken or written. Sometimes the material was factually accurate, sometimes it was false, but it was the full scope of that tradition, a taken-for-granted body of knowledge that formed a certain reality of the Oracle in the minds of the ancients. Many Greeks and Romans in late antiquity would not have had the opportunity or even the desire to visit Delphi, and thus, the most common way to access the Oracle was through the more readily available literature and storytelling relating to it. The second benefit of the social constructionist approach is that it can embrace all the extant literature related to the Delphic Oracle, including the often-ignored Early Church Fathers, since it regards each text as important evidence for the way that the ancients thought about the Oracle and constituted some of its reality. This facet of reality is also part of Delphi’s reception.

Notes 1 A version of this contribution first appeared in a festschrift titled A Dangerous Mind: The Ideas and Influence of Delbert L. Wiens and has been used here with the permission of Wipf & Stock Publishers. See Crosby (2015). The present contribution has had the benefit of two years of reflection and further research as well as the comments of a number of helpful advisors, reviewers, colleagues, and friends who deserve my sincerest gratitude. 2 All dates are given in years ce unless otherwise specified. 3 Theodoret’s complaint pertains to Julian’s educational policies. Julian issued a rescript in 362 barring Christians from teaching as part of his larger program of reform aimed at reviving the cults of the gods that ensured the survival of the state (Jul. Ep. 61c). For discussion of Julian’s educational laws, see Downey (1957) 97–103; Banchich (1993) 5–14. The specific charge that Julian forbade Christian children from attending lectures in the schools is neither corroborated by any non-Christian source nor by Julian’s own κοινὸς νόμος in the rescript above. Rather, the last lines of the rescript specifically state that none were to be excluded from receiving the healing benefits of proper education as a remedy for their mindlessness. For this reason, Hardy (1968: 132 n. 6) challenges the historicity of Theodoret’s claim. 4 Theodoret, HE 185.9–13: καὶ πρῶτον μὲν ἀπηγόρευσε τῶν Γαλιλαίων τοὺς παῖδας οὕτω γὰρ τοῦ σωτῆρος ἡμῶν τοὺς θιασώτας ὠνόμαζε) ποιητικῶν καὶ ῥητορικῶν) καὶ φιλοσόφων μεταλαγχάνειν λόγων. “τοῖς οἰκείοις γάρ”, φησί, “πτεροῖς κατὰ τὴν παροιμίαν βαλλόμεθα ἐκ γὰρ τῶν ἡμετέρων συγγραμμάτων καθοπλιζόμενοι τὸν καθ’ ἡμῶν ἀναδέχονται πόλεμον”. All translations are mine. There is a fable about an eagle being shot with an arrow made from his own feathers in the Aesopic corpus: Aesop. 273.3, 1b. As Wright (1923: 299 n. 4) and Sommerstein (2008: 148) point out, the imagery of this story was striking enough in antiquity for it to be used by Aeschylus (A. Fr. 139), who is quoted by Aristophanes (Ar. Av. 807–8) and alluded to by many other authors even far later. ‘Arrows fletched from our own wings’ 125 5 I use Oracle with a capital O to distinguish the institution of the Delphic Oracle from its individual pronouncements, the Delphic oracles. 6 Ironically, it was the Early Church Fathers’ reliance upon the pagan literary tradition both for their polemics and apologetics that ensured the survival of much of that tradition. For discussion, see Fuhrmann (1990). I have borrowed the terms ‘Delphi of the mind’ and ‘Delphi of fact’ from Lipsey (2001), whose work I discuss at greater length in the conclusion. 7 Strabo 9.3.5; Diod. 16.26. For the findings of the French archaeological team relating to the chasm, see Courby (1927) 64–6. 8 A noteworthy exception is Maurizio (1995: 69–86), whose work analyses the depictions of the Pythia’s activities in light of anthropological evidence and theory. 9 Flacelière (1965) 48; Dempsey (1918) 59; Rous (1976) 110–17. Rous bases his theory on the statement that ‘rien ne laisse moins de ruines qu’un trou!’ (1976: 110). Similarly, Bourguet (1914: 250) suggests a systematic demolition of the site as a possible solution to the problem of the missing chasm. 10 Oppé (1904); Will (1942); Amandry (1950) 215–30; Fontenrose (1978) 197–203. Although Parke and Wormell do not support the historicity of the Delphic chasm, they feel forced to concede that the possibility cannot be excluded (1956: 1.21). 11 For psychedelic theories, see Holland (1933) 201–14; Nilsson (1941) 160; Littleton (1986) 76–91. For psychological theories, see Dodds (1951) 70–4; Parke & Wormell (1956) 1.37–41; Flacelière (1965) 50–1. 12 De Boer, Hale and Chanton (2001); Spiller, Hale and De Boer (2002). 13 Etiope et al. (2006); Foster and Lehoux (2007); Lehoux (2007). 14 The revival of the noxious gas theory in the early 2000s was very well publicized and popularized, appearing even in Scientific American. See Hale et al. (2003). The theory continues to attract the notice of scholars despite work demonstrating its flaws. See Clay (2009) 11–12; Scott (2014) 23–4. Those who have recently demonstrated the flaws of the theory have yet to receive much attention. 15 Although some scholars enjoy mentioning and even quoting the writings of the Early Church Fathers, most dismiss them as unreliable evidence for Delphic oracular practice. See Oppé (1904) 218; Fontenrose (1978) 210; Price (1985) 136. Amandry (1950: 23 n. 1) is more even-handed in his assessment of the value of the writings of the Church Fathers but, except in making passing comments on their thoughts, he only very seldom relies on their authority without additional support from earlier writers. Others such as Will (1942); Flacelière (1965), and Rous (1976) do not cite them at all. 16 Parke and Wormell (1956) 1.288. 17 Berger and Luckmann (1966) 3. These two scholars are responsible for coining the term ‘social construction.’ 18 For a brief discussion of the importance of language in the theory of social constructionism, see Burr (2003) 7–8. 19 Anderson (1983); Said (1978). I should point out the importance that Anderson places on language and printing in the construction of the European concept of a nation and nationality. Said’s work on ‘Orientalism’ is somewhat analogous to what I am suggesting happened in the case of the Oracle in the writings of the Early Church Fathers. Said shows that a society that imagined itself superior perpetuated traditional ‘knowledge’ as commonly accepted reality which did not accurately reflect the reality but supported the society’s claim of superiority. 20 zeitlin (1990) 131. 21 zeitlin (1990) 144. 22 Berman (2015). 23 Eus. PE 4.1.8: εὖ μάλα δὲ πρὸς ἀπάτην ἐσκευωρημένων πλάσματα τυγχάνειν, μέσῳ καὶ ἀμφιβόλῳ συγκείμενα τρόπῳ πρὸς ἑκάτερά τε τῶν ἀπὸ τῆς ἐκβάσεως προσδοκωμένων οὐκ ἀφυῶς ἐφαρμόζειν (Mras, Akademie Verlag, 1956). See Ciholas 2003:101; Lipsey (2001) 218. 126 Daniel Crosby 24 Eus. PE 5.20–1. The foundation of Eusebius’ critique comes from the work of Oenomaus of Gadara, the second-century Cynic who wrote a work called On the Detection of Impostors in which he ridiculed credulousness, particularly the belief in prophetic oracles. The book became a happy hunting ground for Christian apologists eager to tear down the foundations of pagan beliefs and institutions. For the Croesus episode, see Hdt. 1.53. 25 Clem. Al. Protr. 43.3.14–16: προὔδωκε τὸν Κροῖσον τὸν φίλον καὶ τοῦ μισθοῦ ἐκλαθ .όμενος (οὕτω φιλόδολος ἦν) ἀνήγαγε τὸν Κροῖσον διὰ τοῦ Ἅλυος ἐπὶ τὴν πυράν 26 DK B93. Diels and Kranz (1951), Plut. Mor. 404D: ὁ ἄναξ, οὗ τὸ μαντεῖόν ἐστι τὸ ἐν Δελφοῖς, οὔτε λέγει οὔτε κρύπτει ἀλλὰ σημαίνει. Fontenrose (1978: 236–8) argues that the Delphic Oracle did not have a reputation for ambiguity in antiquity, claiming further that this is a ‘wholly modern’ idea. His argument is essentially a dismissal of certain evidence for oracular ambiguity outside of the oracles themselves. About the fragment of Heraclitus, he says that scholars who use the fragment in support of equivocalness ignore the context of the passage within Plutarch’s work. The context, however, cannot tell us anything about the meaning that Heraclitus intended; it only tells us how Plutarch and his interlocutors understood his meaning. The ambiguity of the Oracle is certainly a part of Delphi’s constructed reality in antiquity, as the examples will show. 27 Cic. Div. 2.115: partim flexiloquis et obscuris, ut interpres egeat interprete et sors ipsa ad sortes referenda sit, partim ambiguis et quae ad dialecticum deferendae sint. 28 D. Chr. 10.23–6. 29 Lucian, DDeor. 18(16).1; JTr. 28. 30 Pl. Sym. 202E. Exactly what the δαίμονες were to the Greeks was apparently a matter of debate even among the Greeks. See Burkert (1991) 331–2. These δαίμονες were known in Greek religion since at least the time of Homer (e.g., Il. 1.222). 31 Plut. Mor. 360D–E (Fr. 145), 361B (Fr. 146), 416C–D (Fr. 142), 417B (Fr. 147). The fragment numbers are those assigned by Isnardi Parente and Torandi (2012). See discussion in Schibli (1993) 144–149. This theory of daemonic powers allowed Plutarch’s Cleombrotos to believe that the decline in the importance of the Delphic Oracle that he perceived in his own time and had read about in the past was due to the defection of the daemons that were responsible for relaying Apollo’s prophecy. See Plut. Mor. 418C–E. It is to Xenocrates, perhaps, that our culture owes its conception of the evil demon. Dillon (1996) 31–32; Burkert (1991) 332; cf. Plut. Mor. 419A. 32 Eus. PE 4.15.1: οἶδεν δὲ ὁ τῆς εὐσεβείας φροντίζων, ὡς θεοῖς μὲν οὐ θύεται ἔμψυχον οὐδέν, δαίμοσιν δέ, ἀλλ’ ἤτοι ἀγαθοῖς ἢ καὶ φαύλοις. Cf. Plut. Mor. 361B (F 146), 417C–D. 33 Eus. PE 4.10.3: ὁ δὲ τὸν Θεόφραστον μαρτυρόμενος θεοῖς μὲν οὔ φησιν ἁρμόζειν τὴν διὰ ζῴων θυσίαν, δαίμοσιν δὲ μόνοις, ὥστε κατὰ τὸν αὐτοῦ καὶ Θεοφράστου λόγον δαίμονα εἶναι, ἀλλ’ οὐ θεὸν τὸν Ἀπόλλωνα. Origen appears to be alluding to a similar line of reasoning, citing an uncertain Pythagorean commentary on the Iliad. See Origen, C. Cels. 3.28; 7.6, 35; 8.62. 34 Eus. PE 4.9. 35 Clement of Alexandria relies mostly upon mythology, stories of the gods being on earth, living human-like lives, and dying, as a means to discredit the ‘deathless’ ones (Clem. Al. Protr. 2.24P–31P; cf. Origen, C. Cels. 6.2). Eusebius points out the disparity between mythology and the opinions of the philosophers, both of which were supported by oracular pronouncements at different times (Eus. PE 3.14–15). Additionally, Apollo, who was supposed to be the sun, could not descend in order to bring inspiration (Eus. PE 3.16). Finally, Apollo recommended an inquirer to sacrifice to an evil daemon in an oracle (Eus. PE 4.20). 36 Eus. PE 4.15.5: εἰ γὰρ ἡ διὰ ζῴων ἀλόγων θυσία ἐπάρατος καὶ κακόθυτος πρὸς τῶν φιλοσόφων ἐλέχθη μυσαρά τε καὶ ἄδικος καὶ ἀνόσιος καὶ οὐκ ἀβλαβὴς τοῖς θύουσιν καὶ διά γε ταῦτα πάντα θεῶν ἀναξία, τί χρὴ νομίζειν τὴν δι’ ἀνθρώπων σφαγῆς; Cf. Clem. Al. Protr. 3.42.1–9. See also Ciholas (2003) 52–3. ‘Arrows fletched from our own wings’ 127

37 Eus. PE 4.20: εἰ δὲ λέγοι τις μὴ φαῦλον εἶναι τὸν τῆς ἀνθρωποθυσίας τρόπον, ὀρθότατα δ’ ὑπὸ τῶν παλαιῶν τελεῖσθαι, ὥρα τοῖς νῦν καταμέμφεσθαι πᾶσιν, ὅτι μηδεὶς ὁμοίως τοῖς πατράσιν εὐσεβεῖ. 38 When, and only when, his partners in dialogue are willing to suspend their objections based upon societal preconceptions, Plato’s Socrates lays something of a foundation for a more positive view of the female sex in Greek society, granting that they are at least capable of preforming the same functions as males, although to an inferior level of ability. This train of thought is an appreciable departure from the Greek societal views concerning the accepted role of women and their abilities. See Pl. Rep. 455B–C; cf. Arist. Pol. 1.1254B10–14, NE 8.1158B11–14. However, to Plato’s Socrates, the infe- riority of women to men is a fact even at the ontological level, as in his discussion of metempsychosis at Pl. Tim. 41D–42D, 90E–91B. 39 Origen, C. Cels. 7.5.27–31: εἴπερ δὲ θεὸς ἦν, ὡς Ἕλληνες οἴονται, ὁ ἐν Δελφοῖς Ἀπόλλων, τίνα μᾶλλον ἐχρῆν αὐτὸν ἐκλέξασθαι προφήτην ἢ τὸν σοφὸν ἢ μὴ εὑρισκομένου τοῦ τοιούτου κἂν τὸν προκόπτοντα; πῶς δ’ ἂν οὐκ ἄνδρα μᾶλλον προφητεύειν ἐβούλετο ἤπερ γυναῖκα; See also Ciholas (2003) 79–81; Lipsey (2001) 215. 40 Origen, C. Cels. 7.3.25–30: περικαθεζομένη τὸ τῆς Κασταλίας στόμιον ἡ τοῦ Ἀπόλλωνος προφῆτις δέχεται πνεῦμα διὰ τῶν γυναικείων κόλπων·οὗ πληρωθεῖσα ἀποφθέγγεται τὰ νομιζόμενα εἶναι σεμνὰ καὶ θεῖα μαντεύματα. Origen C. Cels. 3.25.31–4: διὰ τοῦ Πυθίου στομίου περικαθεζομένῃ τῇ καλουμένῃ προφήτιδι πνεῦμα διὰ τῶν γυναικείων ὑπεισέρχεται τὸ μαντικόν, ὁ Ἀπόλλων, τὸ καθαρὸν ἀπὸ γηΐνου σώματος. See also Ciholas (2003) 82–3; Lipsey (2001) 214. 41 The process of vaginal fumigation calls for a woman to be seated over a cauldron of smoldering spices in order to adjust the position of the uterus within her body, which was thought to inform the overall health of a woman. See Littre (1853) 444. See also the discussion in Sissa (1990) 44–9. John Chys. Ep. I Cor. Homil. 29.1 (242.11–19): λέγεται τοίνυν αὕτη ἡ Πυθία γυνή 42 τις οὖσα ἐπικαθῆσθαι τῷ τρίποδί ποτε τοῦ Ἀπόλλωνος, διαιροῦσα τὰ σκέλη εἶθ’ οὕτω πνεῦμα πονηρὸν κάτωθεν ἀναδιδόμενον, καὶ διὰ τῶν γεννητικῶν αὐτῆς δια δυόμενον μορίων πληροῦν τὴν γυναῖκα τῆς μανίας, καὶ ταύτην τὰς τρίχας λύουσαν λοιπὸν ἐκβακχεύεσθαί τε, καὶ ἀφρὸν ἐκ τοῦ στόματος ἀφιέναι, καὶ οὕτως ἐν παροινίᾳ γενομένην τὰ τῆς μανίας φθέγγεσθαι ῥήματα. 43 Clem. Al. Strom. 1.21.135.2: οἳ δὲ καὶ ὑπὸ δαιμόνων κινηθέντες ἢ ὑδάτων καὶ θυμιαμάτων καὶ ἀέρος ποιοῦ ἐκταραχθέντες. 44 If we are to imagine the inspiration of the Pythia as a sexual act between god and priestess, might we also see the result of the inspiration, the uncontrollable madness, as orgasmic? 45 Origen, C. Cels. 7.3–4; Clem. Alex. Strom. 1.21.135.3; John Chys. Ep. I Cor. Homil. 29.1. 46 Str. Geo. 9.3.5; Paus. 10.5.7. Plut. Mor. 414E1–6: εὔηθες γάρ ἐστι καὶ παιδικὸν κομιδῇ τὸ οἴεσθαι τὸν θεὸν αὐτὸν ὥσπερ τοὺς ἐγγαστριμύθους Εὐρυκλέας πάλαι νυνὶ δὲ Πύθωνας προσαγορευομένους ἐνδυόμενον εἰς τὰ σώματα τῶν προφητῶν ὑποφθέγγεσθαι τοῖς ἐκείνων στόμασι καὶ φωναῖς χρώμενον ὀργάνοις (‘It is silly and childish to suppose that the god himself just like ventriloquists, who long ago were called “Eurycleis” but now “Pythones,” enters into the bodies of his prophets and utters making use of their mouths and voices as instruments’). Ammonius’ objection is that it is beneath the dignity of Apollo to enter the Pythia for himself, not that this was not a conception of the mantic mechanism. The agent responsible was a daemon. See Plut. Mor. 417A. Πνεῦμα has a broad range of meanings in Greek literature, from ‘wind’ or ‘breath’ to discarnate ‘spirit.’ See LSJ, s.v., ‘πνεῦμα.’ 47 One gynecological treatise describes such symptoms as eyes rolling back, grinding of the teeth, and flowing of saliva. See Littre (1853) 32. Sissa remarks that in the case of the Pythia, ‘it is as if a well-known image of a traditional therapy had been distorted for the purpose of representing the disease that it was intended to cure’ (1990: 50). Sissa also believes that the obscene nature of the process of oracular consultation at Delphi 128 Daniel Crosby is the reason for much of the silence that we encounter in the primary sources of the classical period on mantic session at Delphi: ‘It evokes what ought not to be seen: an inspired pregnant woman in a temple – a woman who simultaneously opens her mouth and her vagina’ (1990: 52). This is a possible interpretation of the evidence. On the other hand, Amandry (1950: 47–8) proposes that the madness of the Pythia was a counterfactual tradition that sprung from a misunderstanding of the word μηνία and its derivatives at Pl. Phdr. 244A–B. 48 Plut. Mor. 566D: ἅμα δ᾽ἐπειρᾶτο προσάγων ἐπιδεικνύειν αὐτῷ τὸ φῶς ἐκ τοῦ τρίποδος, ὡς ἔλεγεν, διὰ τῶν κόλπων τῆς Θέμιδος ἀπερειδόμενον εἰς τὸν Παρνασόν. Cf. Iambl. Myst. 3.11.126. Themis, in certain versions of the Delphic Succession Myth, was said to have held the means of prophecy at Delphi. See for example E. IT 1259–70. 49 LSJ, s.v., ‘κόλπος.’ 50 [Longinus], Subl. 13.2.5–9: τὴν Πυθίαν λόγος ἔχει τρίποδι πλησιάζουσαν, ἔνθα ῥῆγμά ἐστι γῆς ἀναπνέον, ὥς φασιν, ἀτμὸν ἔνθεον, αὐτόθεν ἐγκύμονα τῆς δαιμονίου καθισταμένην δυνάμεως παραυτίκα χρησμῳδεῖν κατ’ ἐπίπνοιαν. Amandry says: ‘Le mot ἐγκύμων employé dans ce texte prouve qu’Origène et saint Jean Chrysostome trou- vaient déjà chez les auteurs païens la notion d’une fécondation physique de la Pythie par l’esprit apollinien, dont ils devaient tirer le thème de railleries obscenes’ (1950: 53 n. 1). 51 The date of the treatise On the Sublime, which is commonly attributed to a certain Longinus, is debatable. However, current theories place its authorship sometime between the first and the third century, which would have given time enough for both Eusebius and John Chrysostom to have read the work. For a summary of the debate on the date of the treatise, see Russell (1964) xxii–xxx. 52 Eus. PE 5.16.12–15: Πυθῷον δ᾽οὐκ ἔστιν ἀναρρῶσαι λάλον ὀμφήν, ἤδη γὰρ δολιχοῖσιν ἀμαυρωθεῖσα χρόνοισιν, βέβληται κληῖδας ἀμαντεύτοιο σιωπῆς, ῥέξατε δ᾽ὡς ἔθος ἐστὶ θεόπροπα θύματα Φοίβῳ. See also Parke and Wormell (1956) 2.194 (475). Parke and Wormell note that sixteen of the twenty-four ‘significant words’ are hapax legomena in the corpus of Delphic oracles. Fontenrose’s case (Didyma 41) that this is a pronouncement of the Oracle at Didyma is based on the observation that the Pythia’s silence would be claimed by a speaking Pythia if it were genuine and that the oracle immediately preceding this in Eusebius’ work is attributed to Didyma (1978: 427). 53 Parke and Wormell (1956) 2.194 (476): εἴπατε τῷ βασιλῆϊ, χαμαὶ πέσε δαίδαλος αὐλά. οὐκέτι Φοῖβος ἔχει καλύμβαν, οὐ μάντιδα δάφνην, οὐ παγὰν λαλέουσαν, ἀπέσβετο καὶ -λάλον ὕδωρ. Gregory (1983) 355–66 believes that this oracle is a genuine pronounce ment of Delphi, pleading with Emperor Julian to save the Oracle from lapse. Cf. Bowra (1959) 426–35; Fontenrose (1978) 353 (Q263). 54 Parke and Wormell (1956) 2.518; Fontenrose (1978) 349 (Q250): παῖς Ἑβραῖος κέλεταί με, θεὸς μακάρεσσιν ἀνάσσων, τόνδε δόμον προλιπεῖν καὶ Αἵδην αὖθις ἱκέσθαι. λοιπὸν ἄπιθι σιγῶν ἐκ βωμῶν ἡμετεράων. 55 Delatte (1927) 325: Ὀψέ ποτέ τις [[φησιν]] ἐπὶ τὴν πολυσχεδῆ ταύτην ἐλάσειε γῆν καὶ δίχα σφάλματος γενήσεται σάρξ: ἀκαμάτοις δὲ θεότητος ὅροις ἀνιάτων παθῶν λύσει φθοράν: καὶ τούτῳ φθόνος γενήσεται ἐξ ἀπίστου λαοῦ: καὶ πρὸς ὕψος κρεμασθήσεται ὡς θανάτου κατάδικος: ταῦτα δὲ ἑκὼν πείσεται φέρειν: θανὼν δὲ εἰς ζωὴν αἰώνιον ὀρεῖται. See also Fontenrose (1978) 354 (Q268). 56 Eusebius certainly consulted Porphyry’s On the Philosophy to be Derived from Oracles, but how scientifically Porphyry went about gathering his sources and authenticating them is the issue here. 57 August. Contra Faust. 13.1: Ita nihil, ut dixi, Ecclesiae Christianae Hebraeorum testimonia conferunt, quae magis constet ex Gentibus quam ex Judaeis. Sane si sunt aliqua, ut fama est, Sibyllae de Christo praesagia, aut Hermetis, quem dicunt Trismegistum, aut Orphei, aliorumque in Gentilitate vatum, haec nos aliquanto ad fidem juvare poterunt, qui ex Gentibus efficimur Christiani: Hebraeorum vero testimonia nobis, etiamsi sint vera, ante fidem inutilia sunt, post fidem supervacua; quia ante quidem eis credere non poteramus, nunc vero ex superfluo credimus (‘Thus, as I have said, the testimonies of the ‘Arrows fletched from our own wings’ 129 Hebrews offer nothing to the Christian Church, which consists more of Gentiles than of Jews. Indeed, if, as is the rumor, there are any prophecies of the Sibyl, or Hermes, whom they call Trismegistus, or Orpheus, or the other prophets of the Gentiles, these will be able to assist us Gentiles, who become Christians, toward faith considerably. But the testimonies of the Hebrews are useless before we believe, even if they may be true, and they are redundant afterwards, because we were not even able to believe them before, but now we believe them unnecessarily’). Hardly any of the Early Church Fathers would have agreed with Faustus that the Hebrew prophets were of no value to the Christian converted from paganism. Augustine himself spends the next several sections arguing against Faustus’ claim, explaining that the whole conception of the Christ depends on the Hebrew prophets. In this instance, the two seem to be talking past each other. 58 Lipsey (2001) 2. 59 Stating the purpose of his book, Lipsey says: ‘The project of this book is to stake a claim in rocky Pytho on behalf of modern minds for which some Delphic material – not all, by any means – can be uplifting and provocative, as it was for many ancient listeners and readers’ (2001: 7). 60 Clem. Alex. Protr. 11.1; Eus. PE 5.16; Greg. Naz. Or. 5.32. Eusebius draws extensively from Plutarch for evidence of the decline of the Oracle. 61 Parke and Wormell (1956) 1.288; Scott (2014) 236. Scott’s work does indeed provide a more nuanced appreciation of the decline of Delphi by introducing often-ignored evidence of public works projects and statuary and monumental dedications as indicative of some continuity of the status of Delphi. However, although this evidence can be demonstrative of the importance of Delphi as a place of history and social memory, they can only be suggestive of the continuity of its oracular operations. 62 See, for example, Cic. Div. 1.38, 2.117; Plut. Mor. 409E–38E. 63 Cic. Div. 1.38: nunc minore gloria est …; Cic. Div. 2.117: cur isto modo iam oracla Delphis non eduntur non modo nostra aetate, sed iam diu, ita ut iam nihil possit esse contemptius? 64 For a few possible reasons for the decline of the Oracle, see Dempsey (1918) 167–81; Flacelière (1965) 72; Parke and Wormell (1956) 1.277–81; Heineman (2012). 65 Fontenrose (1978) 5, 240–354. The seven responses are numbered H68, H70, Q258, Q261, Q262, Q263, Q268. Thus, Scott’s claim that ‘we have evidence that the oracle of Delphi continued to give responses right through into the fourth century AD’ is not very well supported, as the authenticity of all of these oracles is dubious or debatable. 66 I grant possible exceptions for the two oracles that were supposedly given to the Emperor Julian. Parke and Wormell (1956) 2.194–5, 232–3 (476, 600); Fontenrose (1978) 352–3 (Q262, Q263). 67 There is a common sense in the field that a clear picture of the operations of the Delphic Oracle is not possible with the current state of our evidence. This feeling is often couched in hedging phrases that come on the coattails of claims made on the basis of a subset of evidence, but occasionally scholars are more transparent. For example, Bourguet (1914: 250) famously wrote, ‘la dernière Pythie a emporté son secret,’ and Fontenrose (1952: 445) says in his review of Amandry: ‘There is too little evidence upon which to base sound and convincing conclusions. Most ancients who wrote about the Delphic oracle did not themselves know what was done. Those who did, e.g. Plutarch, tell us little ….’ 68 There is already some work on the processes by which the construct was formed. See Amandry (1950: 47–8), who argues that the conception of the Pythia’s raving and bacchant-like behavior stems ultimately from a misunderstanding of the μηνία discussed by Plato at Pl. Phdr. 244A–B. However, Amandry’s purpose is positivist, as he uses this argument to discredit the authority of later authors and to isolate the historical practices of the Delphic Oracle as the only meaningful reality. See also Will (1942) for a similar study on the construction of the prophetic pneuma. 10 Egyptian necromancy in Heliodorus Aethiopica (6.12–15) and the Witch of Endor narrative (1 Sam 28)

John Hilton

Introduction There are many episodes in the ancient Greek novels in which magic, divination and sorcery are practised.1 This chapter will concentrate on a single case. In the Aethiopica of Heliodorus (6.12–15) the heroine, Chariclea, and her mentor, the Egyptian priest of Isis, Calasiris, witnessed an old woman (variously referred to as ἡ πρεσβῦτις, ἡ γραῦς, τὸ γύναιον) from the Egyptian village of Bessa performing ‘an unholy ritual, but one commonly practised by Egyptian women’, by the light of a full moon (6.14 σκηνῆς τινος οὐκ εὐαγοῦς μὲν ταῖς δὲ Αἰγυπτίαις ἐπιχωριαζούσης θεωρὸς ἐγίνετο).2 The woman poured libations of honey, milk and wine into a pit (βόθρον). She then took a cake made of wheat and shaped into the likeness of a man, crowned it with bay and fennel, and threw it into the pit together with a branch of bay sprinkled with her own blood (πέμμα στεάτινον εἰς ἀνδρὸς μίμημα πεπλασμένον δάφνῃ καὶ μαράθῳ καταστέψασα εἰς τὸν βόθρον ἐνέβαλλεν). She then uttered outlandish spells over her dead son’s body (βαρβάροις τε καὶ ξενίζουσι τὴν ἀκοὴν ὀνόμασι κατευξαμένη) and succeeded in raising him up (ὀρθὸν ἑστάναι τῇ μαγγανείᾳ κατηνάγκαζεν). Next she asked the dead man whether her second son would survive the war against the Persians and return home, but the corpse relapsed into death before replying. The old woman persisted, however, and employed more powerful magic to revive the body. Chariclea became afraid at these strange actions and asked Calasiris whether they should ask the old woman the whereabouts of her fiancé, Theagenes, from whom they had become separated, but he refused to entertain this suggestion for he regarded the ritual as unclean and polluting.3 The old woman succeeded in reviving her son again (6.15) but he was angry at what she had done and told her that by using black arts she had disturbed him in the afterlife. He told her that she and her second son would soon meet violent deaths. He added that he was aware of the presence of Calasiris and Chariclea and revealed that Calasiris’s sons were currently engaged in a battle against one another at Memphis, urging Calasiris to make peace between them. He also predicted that Chariclea would find Theagenes and would live her life with him as royalty on the boundaries of the earth. He then relapsed into final death. Egyptian necromancy in Heliodorus 131 The old woman became enraged at Chariclea and Calasiris because they had witnessed her performing necromantic rituals (ὡς ἐπιβούλως τε καὶ πρὸς ἐναντίου τοῖς αὐτῆς μαγγανεύμασι κατασκόπους γεγενημένους, ‘[on the grounds] that they had spied on her necromancy from malicious and hostile motives’) and charged at them, intending to kill them. However, she failed to see a broken spear projecting from the ground on the battlefield and ran into it, thus meeting the violent death her son had predicted. The element of secrecy and the fear of detection in this narrative are also unprecedented in necromantic sources. This chapter investigates the literary, religious and historical context of this episode, proceeding on the basis of the widely held view that the Aethiopica was written halfway through the fourth century, at the approximate time of the Emperor Julian’s attempt to revive paganism as the official religion of the Roman Empire.4 It argues that the necromancy of the old woman of Bessa can only be fully understood in the context of the polemic among Christian writers, especially in the fourth century, over the practice of divination, and especially necromancy, as told in the Witch of Endor story.

Hypotexts A number of hypotexts have already been noted for the Heliodorus passage above.5 It is clear that it recalls the necromantic scene in the Odyssey (11.24–224) in which Odysseus summoned the spirits of the dead in the land of the Cimmerians where Helios does not shine or, as Circe described it, ‘the groves of Persephone’ and the ‘mouldering home of Hades’ (10.509–10), in order to discover a way home. The ritual performed by the witch of Bessa, especially the libations of milk, wine and honey that are poured into a pit, resemble the actions of Odysseus very closely (where three libations also are poured: milk mixed with honey, wine, and water). The necromantic episode in Heliodorus has a similar prophetic function, delivering guidance to both Calasiris and Chariclea and advancing the action of the plot. Both of these prophecies turn out to be true. Just as in Homer’s epic, the information provided by the dead gives direction and a sense of urgency to the leading characters.6 Another likely hypotext of the Aethiopica passage is provided by Lucan in the Pharsalia (6.413–830).7 Here, before the decisive Battle of Pharsalus, Sextus Pompeius’ desire to know the outcome of the battle drives him to consult a Thessalian necromancer, Erichtho, rather than the holier oracles of Delphi or Dodona. Erichtho successfully infuses blood and ‘lunar poison’ into a recently dead corpse and brings it back to life through spells and enchantments. The revived corpse predicts the deaths of ‘all the leaders’ but is ambiguous about the fate of Sextus himself. This episode emphasizes that Sextus Pompey’s action in consulting the witch was ‘unholy’ (nefas, 6.510) and the witch invokes the powers of the underworld rather than the heavenly gods to reveal the future. However, despite this Sextus Pompey deliberately searches for Erichtho, whose reputation and expertise he has heard of from the local population of Thessaly, and per- sists in asking her to foretell his future, since he is anxious to know whether he 132 John Hilton will become the ruler of the world (dominus rerum) or inherit only death (tanti funeris heres, Phars. 6.592–603). Lucan states that, by the very act of consulting Erichtho, Sextus disgraced his father (Pompeii ignava propago, 6.589) and gave legitimacy to such disgraceful acts. While the Odyssey passage and the long scene in Lucan’s Pharsalia may have influenced Heliodorus’ account in the Aethiopica, there is a clear difference between them. There is no suggestion in Homer and Lucan, or indeed any of the other necromantic texts, that necromancy is contrary to the law and in none of these sources does the revived corpse condemn the actions of the necromancer as directly as Heliodorus does in the Aethiopica. The practice of necromancy is criticized by the ‘prophet’ Calasiris, who disassociates himself from the rite. He contrasts necromancy with the genuine art of the true prophet, which was based on ‘legitimate sacrifice and pure prayer’ (ἐκ θυσιῶν ἐννόμων καὶ εὐχῶν καθαρῶν):

ὁ δὲ παρῃτεῖτο φάσκων καὶ τὴν θέαν οὐκ εὐαγῆ μὲν κατ’ ἀνάγκην δ’ οὖν ὅμως γενομένην ἀνέχεσθαι· εἶναι γὰρ οὐ προφητικὸν οὔτε ἐπιχειρεῖν οὔτε παρεῖναι ταῖς τοιαῖσδε πράξεσιν, ἀλλὰ τὸ μαντικὸν τούτοις μὲν ἐκ θυσιῶν ἐννόμων καὶ εὐχῶν καθαρῶν παραγίνεσθαι, τοῖς δὲ βεβήλοις καὶ περὶ γῆν τῷ ὄντι καὶ σώματα νεκρῶν εἰλουμένοις οὕτως ὡς τὴν Αἰγυπτίαν ὁρᾶν ἡ τοῦ καιροῦ περίπτωσις ἐνδέδωκε. (Aethiopica 6.14.7)

But he refused, saying that the mere sight of such things was unclean and that he could only tolerate it because he had no alternative; it was not proper for a priest either to take part in or to be present at such rites; the prophetic powers of priests proceeded from legitimate sacrifices and pure prayer, whereas those of the profane were obtained literally by crawling upon the ground and skulking among corpses, as the accidents of circumstances had permitted them to see this Egyptian woman doing.

The contrast between high and low forms of magic is found elsewhere in the Aethiopica (3.16.3), as has been noted by Meriel Jones, who relates it to the Platonic notion of heavenly and earthly love.8 The necromantic episode in the Aethiopica belongs to the low form of magic explicitly rejected by Calasiris in this earlier passage. The old woman is the embodiment of base magic, which is treated critically in the Aethiopica.9 Such scenes are nevertheless not avoided entirely in the novel. Even the old woman’s son, who is initially compassionate towards her and inclined to filial obedience, nevertheless criticizes her actions severely:

ἐγὼ μὲν” ἔφη “σοῦ τὰ πρῶτα ἐφειδόμην, ὦ μῆτερ, καὶ παρανομοῦσαν εἰς τὴν ἀνθρωπείαν φύσιν καὶ τοὺς ἐκ μοιρῶν θεσμοὺς ἐκβιαζομένην καὶ τὰ ἀκίνητα μαγγανείαις κινοῦσαν ἠνειχόμην, σῴζεται γὰρ ἡ περὶ τοὺς φύντας αἰδὼς ἐφ’ ὅσον οἷόν τε καὶ ἐν τοῖς ἀποιχομένοις. Ἀλλ’ ἐπειδὴ καὶ ταύτην ἀναιρεῖς τὸ κατὰ σαυτὴν καὶ ἐλαύνεις οὐκ ἀθεμίτοις μόνον τὴν ἀρχὴν ἐπιχειρήσασα Egyptian necromancy in Heliodorus 133 ἀλλ’ ἤδη καὶ εἰς ἄπειρον τὸ ἀθέμιτον ἐπεξάγουσα, οὐκ ὀρθοῦσθαι μόνον καὶ νεύειν ἀλλὰ καὶ φθέγγεσθαι σῶμα νεκρὸν ἐκβιαζομένη κηδείας μὲν τῆς ἐμῆς ἀμελοῦσα καὶ ταῖς λοιπαῖς ἐπιμίγνυσθαι ψυχαῖς ἐμποδίζουσα χρείας δὲ μόνης γενομένη τῆς σῆς, ἄκουε ταῦθ’ ἃ πάλαι σοι μηνύειν ἐφυλαττόμην.

I tolerated your transgression of the laws of man’s nature, your affront to the ordinances of destiny, your use of the black arts to move the immovable, for even in the afterlife we continue to respect our parents so far as we may. But the respect I had for you is now forfeit by your own actions: not content with the first sin of compelling a dead body to stand upright and nod its head, you are taking your sinfulness to the extreme of extorting speech from me as well. With no thought for anything but your own concerns, you neglect the rites that are my due in death and keep me from the company of the other souls. Learn now what I have hitherto kept from telling you!

This distinction between legitimate and illegitimate forms of divination makes more sense when seen within the context of the legislation by Christian emperors to suppress pagan forms of sacrifice with the intent of gaining knowledge of the future (discussed below).10 Calasiris’ refusal to participate in these rites and the strong moral condemnation of the old woman by her own son indicate a departure from previous treatments of the theme. These differences between the Heliodorus passage and earlier accounts of necromancy are significant when contextualized within the struggle for moral legitimacy between Christianity and paganism in the fourth century.

Attempts to suppress necromancy and divination in the fourth century Papyrological discoveries have shown that in the third century there was renewed interest in necromancy and in the Homeric variety in particular. A papyrus fragment from Book 18 of Julius Africanus’ Kestoi (3rd century POxy 412 = PGM XXIII) adds a hymn in hexameter verse to Homer’s account of Odysseus’ consultation of the dead. However, it was in the fourth century that the most significant examples of necromantic spells on magical papyri were produced. PGM IV.1928–54 invokes Helios and his holy angels to grant power over the ‘spirit of this man who died a violent death’.11 Further examples of necromantic spells can be found in PGM IV.2006–125, 2145–240 and XIII.277–83. Pachoumi argues, despite some evidence to the contrary, that these spells do not merely invoke the spirits of the dead but also imply the physical resurrection of the corpse, who then serves as the divine assistant (paredros), or medium, of the magician.12 Matthew Dickie points out that in the fourth century ce the Emperor Constantine introduced legislation forbidding haruspices from entering private houses on penalty of death (CTh 9.16.1).13 Ammianus Marcellinus informs us that Constantius II introduced legislation punishing even minor acts of magic (16.8.2).14 Constantius II was also concerned about the practice of magic in Egypt and sent his agent Paul the Chain to the area to investigate magical practices (19.12.3–6). A law passed by Constantius in 357 made it illegal to consult fortune tellers and diviners (CTh 134 John Hilton 9.16.4), especially those who summoned up the spirits of the dead – again on pain of death:

multi magicis artibus ausi elementa turbare vitas insontium non dubitant et manibus accitis audent ventilare, ut quisque suos conficiat malis artibus inimicos. hos, quoniam naturae pergrini sunt, feralis pestis absumat. (CTh 9.16.5)

Many have dared to disturb the elements with magic arts and do not hesitate to disrupt the lives of the innocent and dare to ventilare by summoning the shades of the dead in order to destroy their enemies by evil arts. May a deadly plague consume these men who are alien to nature!15

This legislation against magic continued in the reign of Valentinian, Julian’s eventual successor after the brief reign of Jovian.16 Imperial hostility towards those who consulted diviners in order to know how the future would unfold derived from the treasonous nature of such enquiries, when they related to the death of the emperor or the fortunes of his competitors. As far back as 11 ce, when people began to speculate about his death, Augustus made divination in secret illegal.17 However, this and later imperial decrees had very little effect on such activities, which escalated in intensity.18 In the fourth century, Firmicus Maternus, for example, advised astrologers to avoid ‘illegal’ questions relating to the emperors.19 Ammianus Marcellinus provides a lurid and extensive account of a séance that was held to determine who would succeed Valens on the throne of the Roman Empire (29.1.5–44). In this episode a certain Palladius gave information about how Fidustius, Hilarius and Patricius had made use of an ouija board to spell out the name of Theodorus and in this way identified him as the successor of the emperor. These men and others associated with them were interrogated under torture and executed after they had revealed details of the ritual.20 It is also clear that the Christian Church vigorously opposed magic of every sort. Canon Law excluded magicians from catechism unless they repented of their actions and underwent severe penance.21 Basil of Caesarea, Gregory of Nyssa and Athanasius of Alexandria all opposed pagan magic and divination.22 Graf refers to the fourth-century Christian writer Arnobius’ denunciation of the harm caused by diviners and prophets (Against the Gentiles 1.43.5).23 Even Christian bishops were not immune from punishment. In 386 ce, the Spanish bishop Priscillian of Avila was condemned to death in Trier for practising sorcery.24 It is self-evident that this legislation was repeatedly ignored, since episodes involving magical divination continued unabated during this period.25 The name of the village in which the old woman in the Aethiopica lived is also significant in this regard. Heliodorus clearly has in mind an Egyptian village, although not one remotely close to the other places mentioned in his narrative.26 Ammianus Marcellinus mentions an oracle of the Hellenized pantheistic god Besa (more usually spelled Bes) near the town of Abydum in a remote part of Egyptian necromancy in Heliodorus 135 the Thebaid in the south of Egypt, which was famous for foretelling the future and at which ancient local rituals continued to be practised, especially in the fourth century (19.12.3: Oppidum est Abydum in Thebaidis parte situm extrema. Hic Besae dei localiter adpellati oraculum quondam futura pandebat, priscis circumi- acentium regionum caerimoniis solitum coli).27 The location of this oracle was so remote that many consulted it by correspondence rather than travelling there in person (Amm. Marc. 19.12.4). It was certainly so far away from the Nile Delta, where most of the rest of the action of the Aethiopica takes place, that the reader is led to suppose that Heliodorus, who leaves the precise location of the village of Bessa extremely vague, wanted to include it for the associations that the name carried, even if this made the narrative less plausible.28 These associations have to do with the religious conflict between pagan and Christian in the fourth century.29 According to Ammianus, those who consulted this oracle about the will of the gods came mainly from Antioch and Alexandria (19.12.8), and communicated with it either in person or by correspondence (19.12.4). Some of the letters (chartulae vel membranae) were sent to Constantius out of spite (maligne, 19.12.5). Constantius, who was narrow-minded (angusti pectoris), then instituted an enquiry into the oracle, setting up a court in Scythopolis in Palestine, and sending the notorious Paul the Chain to conduct the interrogations, despite the fact that the charges were concocted by the investigator and the petitions were not of a serious nature (19.12.5, 9.12.13), as many of the resulting cases showed.30 The scope of the prosecutions was very wide and came to include accusations of sorcery associated with necromancy:

Nam siqui remedia quartanae vel doloris alterius collo gestaret, sive per monumentum transisse vesperum malivolorum argueretur indiciis, ut venefi- cus sepulchrorumque horrores et errantium ibidem animarum ludibria col- ligens vana pronuntiatus reus capitis interibat. (19.12.14)

For if anyone wore charms against the quartan or other fever around his neck, or was accused by malicious information of passing a tomb at nightfall, he was pronounced guilty of being sorcerer and of collecting the horrors and empty mockeries of wandering spirits there, and was condemned to death.

Ammianus adds that the petitioners were prosecuted as if they had been consulting important oracles, such as Claros, Dodona or Delphi, rather than an obscure one in the remote south of Egypt (19.12.15). This suggests that those who consulted the oracle did so because it was remote and little known, and would attract less attention from the Christian authorities.31 The proceedings as a whole resemble a campaign by the Christian emperor to repress pagan magical practices that were undertaken in order to ascertain knowledge about the future. This is certainly indicated by the later history of the site, which was fiercely contested by pagans and Christians32 before it eventually became a bishopric under the Christian emperor Valens (364–378). 136 John Hilton The oracle of Besa was also sometimes associated with Antinopolis33 – the site of the burial of the emperor Hadrian’s beloved Antinous. It was here that Antinous died during the festival of Osiris on 24 October 130 in mysterious and controversial circumstances.34 Dio Cassius (69.12.2–4) states that Hadrian was curious about divination by human sacrifice and made use of ‘prophetic magic of all kinds’ (τά τε γὰρ ἄλλα περιεργότατος Ἁδριανός … ἐγένετο, καὶ μαντείαις μαγγανείαις τε παντοδαπαῖς ἐχρῆτο). The historian also suggests that Antinous volunteered to die because it was necessary for someone to die willingly in order that Hadrian might achieve the (undefined) ends he had in view (ἑκουσίου γὰρ ψυχῆς πρὸς ἃ ἔπραττεν ἐδεῖτο). Cassius Dio’s narrative of this event (69.12.2– 4) and its association with the festival of Osiris, who, according to the well- known Egyptian myth, died and came back to life, provides intriguing context to the story of the necromantic rituals of the old woman of Bessa in Heliodorus’s Aethiopica, and might explain why the oracle retained such a following in the fourth century:35

[2] ὁ γὰρ Ἀντίνοος ἦν μὲν ἐκ Βιθυνίουπόλεως Βιθυνίδος, ἣν καὶ Κλαυδιούπολιν καλοῦμεν, παιδικὰ δὲ αὐτοῦ ἐγεγόνει, καὶ ἐν τῇ Αἰγύπτῳ ἐτελεύτησεν, εἴτ᾽ οὖν ἐς τὸν Νεῖλον ἐκπεσών, ὡς Ἁδριανὸς γράφει, εἴτε καὶ [3] ἱερουργηθείς, ὡς ἡ ἀλήθεια ἔχει. τά τε γὰρ ἄλλα περιεργότατος Ἁδριανός, ὥσπερ εἶπον, ἐγένετο, ,καὶ μαντείαις μαγγανείαις τε παντοδαπαῖς ἐχρῆτο. καὶ οὕτω γε τὸν Ἀντίνοον ἤτοι διὰ τὸν ἔρωτα αὐτοῦ ἢ ὅτι ἐθελοντὴς ἐθανατώθη (ἑκουσίου γὰρ ψυχῆς πρὸς ἃ ἔπραττεν ἐδεῖτο), ἐτίμησεν ὡς καὶ πόλιν ἐν τῷ χωρίῳ, ἐν ᾧ τοῦτ᾽ ἔπαθε, καὶ συνοικίσαι [4] καὶ ὀνομάσαι ἀπ᾽ αὐτοῦ. καὶ ἐκείνου ἀνδριάντας ἐν πάσῃ ὡς εἰπεῖν τῇ οἰκουμένῃ, μᾶλλον δὲ ἀγάλματα, ἀνέθηκε. καὶ τέλος ἀστέρα τινὰ αὐτός τε ὁρᾶν ὡς καὶ τοῦ Ἀντινόου ὄντα ἔλεγε καὶ τῶν συνόντων οἱ μυθολογούντων ἡδέως ἤκουεν ἔκ τε τῆς ψυχῆς τοῦ Ἀντινόου ὄντως τὸν ἀστέρα γεγενῆσθαι καὶ τότε πρῶτον ἀναπεφηνέναι. διὰ ταῦτά τε οὖν ἐσκώπτετο, καὶ ὅτι Παυλίνῃ τῇ ἀδελφῇ ἀποθανούσῃ παραχρῆμα μὲνοὐδεμίαν τιμὴν ἔνειμεν. Antinous was from Bithynium, a city of Bithynia, which we also call Claudiopolis; he had been a favourite of the emperor and had died in Egypt, either by falling into the Nile, as Hadrian writes, or, as the truth is, by being offered in sacrifice. For Hadrian, as I have stated, was always very curious and employed divinations and incantations of all kinds. Accordingly, he hon- oured Antinous, either because of his love for him or because the youth had voluntarily undertaken to die (it being necessary that a life should be surrendered freely for the accomplishment of the ends Hadrian had in view), by building a city on the spot where he had suffered this fate and naming it after him; and he also set up statues, or rather sacred images of him, practically all over the world. Finally, he declared that he had seen a star which he took to be that of Antinous, and gladly lent an ear to the fictitious tales woven by his associates to the effect that the star had really come into being from the spirit of Antinous and had then appeared for the first time. On this account, then, he became the object of some ridicule, and also because at the death of his sister Paulina he had not immediately paid her any honour. Egyptian necromancy in Heliodorus 137 The Witch of Endor narrative The above discussion has shown that necromantic divination was highly controversial in the late Roman Empire and especially in the fourth century. Christian emperors and writers alike condemned the practice and laws were passed to outlaw the ritual. These attempts to suppress pagan practices provides a context which explains the polemical tone of the debate surrounding the well-known Old Testament story of the Witch of Endor (1 Sam. 28), which was hotly debated in the fourth century.36 This episode occurs in the context of the war between the Philistines and Israel. Shortly before his death, the prophet and kingmaker Samuel, made Saul King of Israel. Saul expelled the necromancers from the land, as such practices were forbidden by Mosaic Law (Leviticus 19.31, 20.6), but when he was confronted by the Philistine army he became afraid and felt the need to consult God about what to do. However, he receives no answer from dreams, divination or prophecy. He therefore consults a female medium (ἐγγαστρίμυθος, in Hebrew ‘ob’) at Endor. After fasting and putting on a disguise, Saul meets the woman at night – a suitable time to summon the dead – and asks her to ‘prophesy’ (μάντευσαι) and to ‘bring up’ (ἀνάγαγε) whomever he names. She is afraid to do this at first in the light of Saul’s expulsion of necromancers, but after receiving assurances that she would not be punished, she asks who he wants her to raise up. He tells her to bring Samuel up from the dead. At this point the narrative jumps forward. No details of the necromantic ritual are given. Instead, the woman cries out when she sees Samuel and realizes that her client was Saul, who had earlier enforced the laws against necromancy. The king tells her not to be afraid but to tell him what she saw. She replied that she saw gods (θεούς) emerging from the earth. Saul recognizes Samuel from the medium’s description of an old man and his cloak, and he prostrates himself before him. Samuel then addresses Saul, asking him why he has disturbed him by bringing him back to life. Saul tells him of his anxiety about the war with the Philistines, adding that God was not listening to him and did not communicate with him through prophets or dreams, and asks him what to do. Samuel tells him that God has taken Israel away from Saul because he has not attacked the Amalekites and that David would henceforth rule Israel. He adds that Saul will die at the hands of the Philistines and join him in death the next day, together with his sons. After this encounter, Saul, who is weak from fasting, collapses in terror at the prophecy uttered by Samuel. The woman of Endor revives him with a meal before they part ways. The prophecy comes true when Saul and his sons are killed in the subsequent battle. This narrative is extremely curtailed and many details are unclear. No details are given of the ritual or the words used by the woman to invoke Samuel and there is a severe distortion of the narrative focus, which switches abruptly from the woman’s vision of gods emerging from the ground,37 as she reports to Saul, to Samuel’s direct conversation with Saul. Furthermore, Samuel holds Saul account- able for disturbing him in death, not the woman, and Saul’s future death is predicted as a result of his failure to attack the Amalekites – there is no indication that he is considered culpable for indulging in necromancy. 138 John Hilton The story provoked a considerable controversy in Christian circles at the time of the Roman Empire. In the second century, for example, Justin Martyr uses the story as proof of the existence of souls after death (Dialogue with Trypho 105.4). Tertullian (On the Soul 57–8) argues that those who die violent deaths go to the afterlife accompanied by demons which appear to the relatives of the dead person when exorcized. In doing so, these demons undermine belief in the afterlife and cause people to lose faith in the final judgement and the resurrection. By divine grace, however, these impostures can be detected. This was the way in which the Witch of Endor had managed to bring Samuel back to life – it was not true that a demon had the power to compel a prophet to return to the living, because it was not Samuel but a phantom that appeared and took possession of Saul. Similarly, the apparitions of the dead at night do not prove that the dead come back to life. It is only God who can bring the dead to life, on the day of judgement and the resurrection. In the third-century text The Martyrdom of Pionius, the author alleges that some Jews stated that Christ himself practised necromancy and divination by his death on the cross, or that Christ was raised from the dead by necromancy.38 The author refutes this view by arguing that Samuel was not brought back to life by the necromancer in 1 Kings 28 – to believe that she had this power is evil (14.5) and undermines the righteousness of the true believer. Instead ‘underworld demons’ took on the likeness of Samuel and manifested themselves to the belly-myther and Saul (14.11). Saul was an idolater who was deceived by these demons. Since it is not possible to revive a soul, Jesus could not have risen up from the earth. Instead he ascended to heaven as his disciples testified (14.14). Such was the controversy over this text in the third century that Bishop Alexander of Jerusalem asked Origen of Alexandria to interpret the story in a homily (Origen Homily on 1 Kings 28). For Origen, it was troubling that Samuel’s soul resided in Hell and that the woman of Endor supposedly had sufficient power over it as to be able to bring it back from the dead. Origen’s solution to these problems distinguishes between the author of the story, whom he identifies with the Holy Spirit, and the narrative persona of the scripture which was merely a ‘petty demon’ (daimonion), without power to resuscitate Samuel and to prophesy about the future of Israel. Nevertheless, although it was the persona of the Holy Spirit and not the Holy Spirit itself that invoked the presence of Samuel, the narrative was entirely true. Besides, Origen argues, even Christ himself descended into Hell after his death and before his resurrection, so it would not be implausible for Samuel to have done the same. Moreover, Samuel was able to deliver a true prophecy because he had seen the truth in the afterlife. The debate continued in the fourth century, when Origen’s interpretation of the Old Testament text (Homily on 1 Kings 28) was contradicted by Eustathius of Antioch, Bishop of Beroea around 320 ce. His treatise On the Belly-myther bears the subtitle Against Origen. Eustathius’ account of the passage was written because people were dissatisfied with Origen’s views and more especially because they were disturbed by the doctrinal implications of the story. Eustathius adhered to the literal text of the narrative and avoided Origen’s appeals to allegory. According to Eustathius, both the belly-myther and Saul himself had been driven Egyptian necromancy in Heliodorus 139 mad by a demon and it was for this reason they had engaged in the evil act of necromancy. However, what the belly-myther saw was a false and unsubstantial vision of a demon pretending to be Samuel, as was evident from the fact that Saul had not actually seen this apparition – only the medium had. As a parallel example, Eustathius cited the miracles performed by Moses in Exodus, such as turning a staff into a snake. Such actions he takes as real miracles, while the Egyptian magicians who imitated them were only presenting illusions brought about through magic. The belly-myther, according to Eustathius, had been in the same position as the slave girl possessed of a Python whom Paul encountered and exorcized at Philippi (Acts 16:16–19). Similarly, Eustathius argued, had the apparition actually been Samuel, he would have admonished Saul as Paul did the slave girl. With regard to the actual prophecy delivered by the demon in the form of Samuel, Eustathius contended that it was merely repeating prophecies that had already been pronounced (12). This trick, he added, was one commonly performed by priests and soothsayers who ascertain as much information about the person consulting them as they can before making their prophecies (12.9). Eustathius conceded that the prophecy of the deaths of Saul and his sons was new, but refuted it by arguing that they did not occur the next day as the phantom Samuel had said they would (13.1–2). The words of the false Samuel, that Saul and his son Jonathan would join him in the afterlife, were also blasphemous as they echoed the words on Christ on the cross that the remorseful thief would be with him in Paradise (Luke 23:43). Moreover, just as a chasm separated Lazarus from the rich man in the afterlife, so Saul and Samuel would never have been able to meet and so the phantom Samuel lied when prophesying that Saul and Jonathan would join him in death. Eustathius concluded that many prophets claim to have foretold the truth and argued that they were often allowed to perform their deceptions by God in order to test the faith of true believers (24.10). So in the case of the belly- myther, the righteous should be able to see that necromantic rituals were not a legitimate form of prophecy but abominations frequently condemned in scripture (25.4–5). Even Plato, Eustathius notes, condemned myth-making seen in Homer and Hesiod for the false ideas they inculcate in the young (Plato Rep. 376e–7a).

Conclusion Eustathius’ elaborate attempt to refute Origen’s interpretation of the Endor story provides substantial evidence for how seriously necromancy and magic were taken in the fourth century. This, taken together with the increased interest in necromancy at this time – which is evident in the magical papyri from Egypt, the consultations of the Bessa oracle at Abydos described by Ammianus Marcellinus and the legislation against such practices by the Christian emperors – explains the different nuance given to the consultation of the dead in Heliodorus’ Aethiopica. It is particularly striking that Ephrem compares Julian to Saul – thus connecting the emperor directly to the Witch of Endor story (Hymns against Julian 4.8).39 This is surely another clear indication of the heightened awareness of the issue of necromancy in the fourth century. 140 John Hilton Elsewhere in the novel, it is clear that Heliodorus must have known Christian doctrines well40 and it is more than likely that he was familiar with the contemporary debate on the story of the belly-myther of Endor. Internal evidence confirms this probability. Heliodorus is exceptional in having both a bystander and the revived corpse upbraid and condemn the medium who had brought them back to life in such strong terms. Calasiris specifically implies that old woman’s actions are contrary to law. The Old Testament passage also emphasizes the secrecy and illegality of the encounter between Saul and the medium. Both narratives present prophecies that turn out to be true and in both, consultation of the dead in turn results in the violent death of the parties involved in the ritual: in the Old Testament story, both Saul and his son Jonathan die in battle; in Heliodorus, the old woman herself dies on a battlefield, impaled on a spear. In both cases, there is a degree of empathy with the old woman who resorts to this form of magic: in Heliodorus, the old woman loses both of her sons to war; in the Old Testament, the belly-myther slaughters her last remaining animal to feed Saul. While there does appear to be some awareness of the Old Testament narrative in Heliodorus, it is not necessary to assume that his text is therefore sympathetic to Christianity or that Heliodorus was a Christian at the time he wrote the Aethiopica, although, given the pressure on pagans to convert to Christianity in the fourth century, one cannot necessarily rule out the possibility that he later became a Christian bishop as the testimonia assert.41 While the debate between Origen and Eustathius was an argument between Christians, it is unlikely that the problems it contains would not also have been raised by pagans against Christians. The anti-Christian treatises of Celsus and Julian are fragmentary and no mention of the Endor necromancy is made in the extant texts. However, it would be very surprising if this anomalous text was not used by these pagan authors in their attacks on Christian doctrine in the same way as Jewish writers intimated that the resurrection of Christ was an act of necromancy. Julian also frequently brackets Christians with corpses because of their reverence for the relics of martyrs (Mis. 361b; CG 335b). Morgan has argued that Heliodorus’s text promotes Hellenism over Christianity.42 Indeed, the very fact that Heliodorus chose to include a necromantic episode in his fictional narrative must have been a challenge to the laws enacted by Christian emperors such as Constantius. It would have been impossible for an account of divination through Egyptian magic using the body of a man who had recently met a violent death to avoid controversy in a novel as ambitious as the Aethiopica, especially since Constantine and later Constantius had recently enacted laws against such practices. The only context that makes sense for such a provocative narrative would be the period between this legislation and the measures introduced by Julian’s eventual successor Valentinian – broadly in the middle of the fourth century.43 The narrative would have had most resonance when the apostate Emperor Julian was attempting to revive traditional pagan beliefs.44 Julian himself was accused of conducting human sacrifices and polemical Christian narratives were composed after his reign implicating him in such rites.45 Egyptian necromancy in Heliodorus 141 In such a framework, a narrative that features a true prophecy resulting from the consultation of the dead by means of an Egyptian necromantic ritual would have had strong ideological undertones. Nevertheless, such practices were deemed lowly and polluting by educated Hellenists such as Julian and contrary to belief in the Olympian gods, particularly as Julian was at pains to restore beliefs in the traditional pantheon of Greek gods within an overarching superstructure of Neoplatonic ideas focused on the sky god Helios. So Calasiris’s condemnation of the necromantic ritual, the hostility of the son towards his mother and his description of her actions as sinful, are in line with the ideas of the educated intelligentsia of the Roman Empire in the fourth century. These philosophers tolerated such practices, which were endemic in superstitious lands such as Egypt, while at the same time making use of them to offer a challenge to the legislation of the Christian emperors and to affront fourth-century Christian apologists such as Eustathius and Gregory of Nyssa.

Notes 1 Ruiz-Montero (2005) 38–56. 2 Ogden (2001) 137. 3 This refusal to take an opportunity to obtain information about the future is unique in ancient accounts of necromancy, although it is true that the information is nonetheless later communicated to both Calasiris and Chariclea, when the dead man eventually speaks. See Slater (2007) 57–69. 4 It is not possible to enter fully into the question of the date of the Aethiopica here, since the issues are complex and the bibliography very extensive. However, a consensus has developed among Heliodoran scholars that the novel must be dated to the fourth century. On this, see Morgan (1996) 417–56; Bowersock (1994) 149–60. The present discussion reinforces the argument that a fourth-century date of composition provides the best fit for the interpretation of the novel, and seeks to find connections between the novel and the intellectual and religious controversies of this period. Hilton (2012: 57–68) provides further evidence linking the Aethiopica with the controversial religious reforms of Julian. The present discussion forms part of a series of papers linking the novel of Heliodorus with aspects of the reign of Julian and the controversy aroused by his religious reforms. 5 Slater (2007) 59; Jones (2005) 83 n. 23. 6 Other ancient necromantic texts are less directly relevant to the scene in the Aethiopica. Aeschylus’s Persians (607–99) has the Persian queen Atossa raise the ghost of King Darius from the dead in order to discover what would happen after his defeat at the battle of Salamis. In Seneca’s Oedipus (530–626), Creon describes how a priest revives Zethus, Amphion, Niobe, Agave, Pentheus and finally Laius, who reveals that Oedipus has killed him and married his mother. In Apuleius (Met. 2.21–30), an Egyptian necromancer, Zatchlas, revives the corpse of a young man, who informs a crowd that he had been poisoned by his adulterous wife. As proof of this, he tells them that in the middle of the night the young man hired to guard his corpse from mutilation by Thessalian witches had himself been mutilated as he had responded to the witch’s magic instead of watching the corpse, because he had the same name as the dead man, Thelyphron. The young man finds that his nose had indeed been gnawed off and replaced with a wax one. For all these episodes, see Ogden (2001). 142 John Hilton 7 There is no reason why Heliodorus would not have been able to read Lucan’s Latin narrative, since bilingualism was widespread in the late Roman Empire. See Adams (2003) index s.v. ‘Greeks’. Even before the fourth century, Achilles Tatius clearly had Virgil in mind in the storm scene at the beginning of Book 3 of his Leucippe and Clitophon. Ammianus Marcellinus was of course bilingual and records the remarkable career of the bilingual Strategius Montanus (15.13.1–2). See Drijvers (1996) 532–7. Julian himself undoubtedly knew Latin. See Smith (1995) 13. On Heliodorus and the Lucan passage specifically, see Slater (2007) 65 n. 20; Ogden (2001) 14; Jones (2005) 83 n. 23. For a discussion of the Erichtho episode, see Fauth (1975) 325–44. 8 Jones (2005) 79–98. 9 Both Slater (2007: 60) and Jones (2005: 83) note that Aeth. 3.16.3 anticipates the necromantic episode in Book 6. 10 For the Christianization of the Roman Empire and the conflict with magicians and diviners, see MacMullen (1984: 96, 162 n. 28) and the discussion below. 11 Bremmer (2002) 81; Kahane (1999) 815–36. 12 Pachoumi (2011) 729–40; Pachoumi (2015) 391–413; Ogden (2001). 13 Dickie (2001); Graf (1997) 195. On the Theodosian Code, see Harries and Wood (2010). For sanctions against magic more generally in Roman law, see Pharr (1932) 269–95. It is true that there was legislation against magic and divination throughout the imperial period, as Pharr shows, but, given that the generally accepted date for Heliodorus is the fourth century, it is the laws passed during this period and the greater prominence of magic at this time that deserve special attention. 14 On Ammianus and Roman religion in general, see Davies (2004) 226–85. On Ammianus and divination see Rike (1987) 71–5. 15 Here the term ventilare is derived from ventus (‘wind’) but suggests a link to venter (‘stomach’) and so to ventriloquism, which preoccupies accounts of the ‘belly-myther’ (ἐγγαστρίμυθος) of Endor in the Old Testament. 16 CTh 9.16.7–11. 17 Cassius Dio 56.25.5 discussed by MacMullen (1966: 129). 18 See the extensive discussion by MacMullen (1966: 128–62) and, more recently, Potter (1994). 19 Sogno (2005) 170–1. 20 Discussed by Rike (1987) 38–9; MacMullen (1966) 135–7. 21 Canons of Pseudo-Athanasius 41; Dickie 2001: 257–61. 22 Basil of Caesarea Letter 188.8 (concerning love magic); Gregory of Nyssa Against Fate 62–3 (divination as a form of demonic possession); Athanasius of Alexandria Life of Antony 79.1–2. Gregory mentions a woman who prophesies through her ‘stomach’ (i.e. an ἐγγαστρίμυθος). 23 Graf (1997) 195. 24 Dickie (2001) 270. 25 MacMullen (1984) 96–101. Dodds (1965: 38–9 n. 3) quotes Synezius On Dreams 8 to the effect that no emperor can prevent his subjects from dreaming. It was precisely the secrecy of acts of divination that made them difficult to suppress. 26 According to the grammarians, the name was spelled with either a single or double ‘s’. The name Bessa (‘glen’ in Greek) is attested as a Greek city in Locris (Steph. Byz. Ethnica 167, Strabo 9.4.5; Dakoronia (1993) 115–27), while Besa is also attested as the name of a deme in Attica (Strabo 9.4.5; Barrington Atlas s.v. ‘Besa’). 27 The Hellenistic demon Besa features as an oracle giver in the Greek magical papyri (e.g. PGM VIII.64–110, which contains a prayer to the setting sun). Cf. Betz (1992) 147–8; Malaise (1990) 680–729; Meeks (1992) 423–36. For the connection between Bes, divinity and the soul in African mythology, see Nicholas (1954) 551–94. 28 Heliodorus is vague about the exact geography of Egypt elsewhere too. Cf. Naber (1873) 145–69, 313–53. For example, he refers to Philae as ‘a city on the Nile’ (8.1), Egyptian necromancy in Heliodorus 143 though it is actually an island in the Nile. This lack of topographical precision has been used to refute the suggestion that Heliodorus was a native of Egypt, or Ethiopia, as argued by Glava (1937). As Rattenbury (1938: 145) points out, Heliodorus may have been ‘a writer to whom effect was of more importance than truth’. 29 Scholarship on the conflict between pagans and Christians in general is, of course, huge. See, for example, Edwards (2015); Elm (2012); Hopkins (1999); Lane Fox (1986); MacMullen (1984); Croke and Harries (1982); Dodds (1965); Momigliano (1963). 30 The cases against Andronicus and Demetrius Chytras were shown to be entirely without substance. See Ammianus (19.12.11–12). Nevertheless, offenders such as Simplicius and Parnasius were banished or went into exile (19.12.10) and others were executed (19.12.13). 31 The use of necromancy as opposed to the consultation of legitimate oracles is noted also in Lucan’s narrative (6.423–30). 32 For the climate of violence between Christians and upholders of traditional Egyptian religion in Egypt see Frankfurter (2000) 273–95. 33 See, for example, Stillwell et al. (1976). For the archaeological exploration of Antinopolis, see Gayet (1902a); (1902b); (1903). 34 Opper (2008) 174. 35 The cult of Bes was active also in the third century, but given the likely date of Heliodorus in the fourth century, it is the attention paid to this site during the later period that carries special significance for the purpose of this chapter. The cult of Osiris features significantly in Aethiopica (9.9). 36 Now often referred to as the Medium of Endor (Mitchell 2005) or the Belly-Myther of Endor. See Greer and Mitchell (2007). The latter term comes from the Septuagint translation of the passage, which identifies the woman as ἡ ἐγγαστρίμυθος. See also Mitchell (2005) 414–45; Smelik (1979) 160–79; Scheidweiler (1953) 319–29. 37 For the story that the Etruscan seer Tages rose out of the earth see Ammianus (21.1.10). 38 Greer and Mitchell (2007) 29. 39 I owe this valuable observation to one of the anonymous referees of this chapter. Christian propaganda against Julian was nothing new and would continue long after Julian’s death, as noted below (see Ephrem, Hymns against Julian 2.4 and 4.8). Julian did not allow the vociferous opposition to his religious reforms to deter him, as can be seen throughout his works, especially in the Misopogon. 40 Rohde (1914) 443 n. 1; Koraes (1804–6). 41 For the testimonia, especially Socrates Scholasticus (Test. I) and Nicephorus Callistus (Test. XVII), see Colonna (1938); Morgan (1996) 417–56; Morgan (2005) 309–18. 42 Morgan (2005) 309–18. 43 For Valentinian’s legislation against diviners, see CTh 9.16.7–11. 44 As stated above (n. 4), the purpose of this paper is not to enter directly into the question of the dating of the Aethiopica, except to confirm the preponderant likelihood of the fourth-century date. 45 Garstadt (2005) 83–135; Lieu (1986). For rumours about Julian’s involvement in such rites, see Ephrem, Hymns against Julian 1.16, 2.2 (Julian as sorcerer), 2.4, 4.8 (use of necromancy); Gregory of Nazianzus Or. 4.92,4–7 (nocturnal murders at Antioch covered up by Julian); John Chrysostom de Babyla 79.4; Socrates HE 3.2.4–5; Theodoret HE 3.21, 3.26–7; and the pro-Julianic account of Ammianus Marcellinus 21.1.7 (malicious people attributed to Julian the use of the ‘dark arts’). For the use of fiction in propaganda for and against Julian, see Hilton (forthcoming, 2017). It is no doubt for this reason that the actions of the witch of Bessa are condemned by, but not entirely excluded from, the Aethiopica. 11 Sosipatra: prophetess, philosopher and theurgist Reflections on divination and epistemology in late antiquity1

Crystal Addey

Sosipatra, who is attested only in Eunapius’ biographical work Lives of the Philosophers and Sophists (Vitae Sophistarum; c. 396–405 ce), was a prophetess and philosopher, as well as an indirect successor of Iamblichus’ philosophical school, in late antiquity.2 Throughout her biography, Eunapius consistently characterizes Sosipatra as a ‘divine woman’ – and as spiritually and intellectually gifted and superior to various men, including her father, husband and several male philosophers.3 Sosipatra’s gifts are evident in the story of her childhood and training by Chaldaean prophets, who apparently educated her and initiated her into the Chaldaean mysteries. This episode links her with the tradition of the Chaldaean Oracles. In relation to this, Sosipatra is depicted as a theurgist and prophetess, and her words are consistently characterized as oracular.4 Her ability to know of events from a distance, or ‘remote viewing’ of past, present and future events, is very close to modern notions of ‘psychic’, clairvoyant power and represents an unusual example of such power in the ancient world.5 However, some episodes of clairvoyance or psychic power, including responding to clients’ unspoken questions, were attributed in antiquity to the Pythia at Delphi, the oracle of Amphiaraus and the priest at the oracle of Claros – Sosipatra’s ability implicitly links her with some of the most respected oracles in antiquity.6 She is also reputed to have read and understood effortlessly many philosophical, poetic and rhetorical works, and to have established her own philosophical school in Pergamon.7 What is especially striking about Sosipatra is that she combines rigorous philosophical erudition and learning with her roles as a prophetess and theurgist. This chapter will use the tradition of Sosipatra’s life and activities as a case study for examining the complex relationship between divination, theurgy and epistemology in late antiquity, focusing especially on the connections between rationality and divinatory, prophetic states. In relation to this, the chapter will also examine the influence of the Pythagorean biographical tradition (which was particularly important for Iamblichus’ philosophical school and successors), Plato’s portrayals of Diotima and Socrates, Porphyry’s depiction of Plotinus and Iamblichus’ philosophical praxis upon Eunapius’ representation of Sosipatra as prophetess, philosopher and theurgist. I argue that Eunapius’ Sosipatra exemplifies the inherently and inextricably linked relationship between divinatory and philosophical expertise and cognitive states within late antique theurgy, with her philosophical insight Sosipatra: prophetess, philosopher and theurgist 145 enhancing her oracular visions and vice versa. Sosipatra is presented as a theurgist who has attained an advanced state of praxis; she is the philosopher and prophetess par excellence. In a recent article, Sarah Iles Johnston argues that Sosipatra is passive because (1) she does not actively perform ritual within Eunapius’ biography, but rather asks Maximus to perform a ritual on her behalf; (2) the Chaldaean prophets who train Sosipatra order the chest containing her ritual objects to be sealed before they leave (which Johnston interprets as implying that she will not use the ritual implements henceforth); and (3) because her divinatory, prophetic episodes are presented as spontaneous rather than as the result of her use of physical ritual tools and practices.8 In relation to this, I hope to demonstrate that Sosipatra is presented as an advanced theurgist who is an expert on the kairos, the most appropriate or best time for any given undertaking or activity, and as an active ritual teacher (as well as a proficient teacher of philosophy) who educates Maximus regarding the appropriate timing, ethics and contexts of theurgic ritual practice. In this respect, Sosipatra is an active ritual practitioner and teacher whose advanced, intellectual form of ritual entails a spontaneity in her oracular practices because it is envisaged as based on a constant and active ritual connection to the gods. Finally, I argue that Sosipatra is represented as an important and paradigmatic role model for late Platonist philosophers and theurgists, both female and male, against Ilinca Tanaseanu-Döbler’s claim that Sosipatra is not presented as a role model because she is not depicted as educated following a methodical course that could be imitated by others.9 As we will see, Eunapius depicts Sosipatra as an active role model for her student Maximus. Additionally, Eunapius’ portrayal of Sosipatra, and quite possibly the historical philosopher-prophetess herself, acted as a role model for Asclepigeneia, who trained the later Neoplatonist Proclus in theurgic ritual practices.10 Asclepigeneia is presented as a female philosopher who is descended from the heads of the Athenian School and thus a vital link in the ‘golden chain’ of the Platonic philosophical succession. She is also depicted as possessing considerable ritual expertise, as the central teacher of theurgy in the later Athenian School: ‘For she alone preserved the rituals, and the whole process of theurgy, handed on to her from the great Nestorius by her father’.11

Methodological issues First, it is important to consider the methodological issues surrounding the portrayal of Sosipatra in Eunapius’ work, which is largely anecdotal and hagiographi- cal.12 None of Sosipatra’s writings survive, though presumably she must have written some works as the head of a philosophical school. Nor do we have any other evidence of her life and activities. Therefore we see Sosipatra only through Eunapius’ representation, through the lens of a male biographical writer.13 This is typical for many female philosophers in the late Platonic tradition, including Arete, another philosophical associate of Iamblichus, and Asclepigeneia, who trained Proclus in theurgy.14 Richard Hawley has noted that ancient female philosophers are mostly treated as anomalies within the biographical tradition: they are portrayed as exceptions to the rule, with emphasis placed on their beauty, sexuality 146 Crystal Addey and male connections.15 Eunapius certainly does treat Sosipatra as an anomaly, introducing her biography with the apologetic statement that this extraordinary woman was so famous that she merits inclusion in his catalogue of wise men.16 In relation to gender issues, Sosipatra is portrayed by Eunapius as a ‘divine woman’ or ‘holy woman’. Yet, in the prologue of his narrative, Eunapius asserts that his aim is to write of the main achievements of ‘distinguished men’ (τῶν σπουδαίων ἀνδρῶν), using the specifically masculine term rather than the more neutral ἄνθρωπος, although it may be relevant that he is citing Xenophon here and that the plural masculine noun ἀνδρες could refer to women as well as men.17 The topos of the ‘holy man’, the theios anēr, a term used to describe the wise sage conceived as a locus of divine power in late antiquity, has been set out by Peter Brown and Garth Fowden in relation to Christian and pagan philosophical male figures respectively.18 Nicola Denzey-Lewis and Ilinca Tanaseanu-Döbler have highlighted the gender issues raised by this term; alternative terms such as ‘persons of power’ have been suggested to delineate those attributed with superhuman powers in late antiquity.19 In relation to biographical representation, the romantic and legendary tone of Eunapius’ account of Sosipatra has been noted.20 The story of the Chaldaean priests is clearly modelled on an ancient topos: the theoxenia, myths of humans being visited by the gods, such as the myth of Philemon and Baucis, who were visited in their Phrygian village by Jupiter and Mercury.21 Eunapius hedges his bets in describing the Chaldaean prophets as either heroes, daimones or divine figures. Sarah Iles Johnston argues that the Chaldaean priests are portrayed as ‘angelic souls’ in a theurgic sense, as non-human divine figures. She then extrapo- lates that Eunapius wishes to emphasize that Sosipatra becomes what she is without human tutelage.22 However, even if the Chaldaean priests are presented as non-human, from a Neoplatonic perspective, Sosipatra has reached an advanced stage of spiritual development through cultivation of her receptivity and her progress through the scale of virtues, through her ethical, intellectual and ritual conduct in previous lives, as will be explored below (see the section ‘Sosipatra the Chaldaean’). Furthermore, the mythic associations of Sosipatra’s story can also be clarified within a Neoplatonic context, since Neoplatonists considered myth to be inextricably linked with the journey of the human soul, the nature of the universe and reality.23 Neoplatonic philosophers tended to treat myth as mystical allegory and offered some of the first extensive allegorical accounts of Homeric epic.24 Sosipatra’s life is presented with mythical connotations because within the Neoplatonic tradition, the journey of the human soul is mythical, in the sense that it ‘plays out’ mythological themes and motifs. Thus, by actualizing the full potential of her soul (through ritual, intellectual and ethical means), Sosipatra can be presented as a ‘divine woman’, a figure with mythical associations like Pythagoras. The question of whether any historical information can be gleaned from late antique biography is a vexed and contested issue, particularly contentious in the light of recent poststructuralist critique.25 Although Sosipatra is not attested in other extant ancient sources and none of her writings survive, few would doubt her historical existence given that Eunapius presents her as associated with historical figures who are attested elsewhere, such as Maximus of Ephesus (discussed also Sosipatra: prophetess, philosopher and theurgist 147 by Ammianus Marcellinus and Libanius) and her husband Eustathius, who is the recipient of a letter from Iamblichus, of which only a small fragment survives.26 Furthermore, Sosipatra’s philosophical associate Aedesius and her son Antoninus are entwined with so many historical and philosophical events in this period that it is almost impossible to believe they are fictional; had Eunapius presented a fictional character so closely connected with the lives of prominent philosophers and figures such as Maximus, Eustathius, Aedesius and Antoninus, it would surely have strained acceptance of his narrative.27 Moreover, Iamblichus’ Letter to Arete demonstrates that women were involved in philosophical discussion and activities within Iamblichus’ circle, as was the case with Plotinus’ philosophical school, the Alexandrian School (which included the well-known philosopher Hypatia, almost a contemporary of Sosipatra) and the later Athenian School of Neoplatonism.28 However, beyond the historical fact of Sosipatra’s existence, the extent to which Eunapius’ portrayal reflects historical features of Sosipatra’s life – the ‘real’ Sosipatra – is uncertain and fraught with methodological issues. Many scholars have doubted that the portraits of women in ancient literary texts reflect specific historical women for at least two reasons. First, these accounts were written by male writers and may tell us more about male views, preoccupations and biases than about their female subjects.29 Second – and consequently – these depictions often portray women in terms of literary tropes, which are frequently taken to reflect the ideological nature (particularly regarding ‘proper’ female behaviour) of such depictions.30 However, when considering the portrayal of philosophic women in late antiquity, it seems important to bear in mind that philosophy was considered to be a way of life in antiquity and exemplary ‘real-life’ philosophers were often treated as role models in biographical texts. Furthermore, it seems sensible to distinguish carefully between biographies which claim to be written by eyewitnesses or near contemporaries (such as Eunapius’ work) from those written about biographical subjects in the distant past, such as Porphyry’s biography of Pythagoras and Iamblichus’ work On the Pythagorean Way of Life.31 In relation to this issue, Eunapius was almost a contemporary of Sosipatra and had oral sources and personal testimony from those who knew her well. Eunapius’ account of Iamblichus’ circle in particular seems to be based on oral sources and oral tradition, derived from Eunapius’ own place within the intellectual ‘family tree’ of Iamblichus’ philosophical circle and successors. In an examination of Eunapius’ work (particularly his programmatic statements about his sources in the prologue of the work), which draws on late antique educational practices and customs, Edward Watts has compared Eunapius’ work to oral history.32 Watts emphasizes the oral tradition upon which Eunapius drew in his vivid and lively portrayal of Iamblichus’ philosophical circle, and contrasts this with the much briefer (and somewhat bland) accounts of Plotinus and Libanius, which are drawn from written sources, primarily Porphyry’s Life of Plotinus and Libanius’ Oration One respectively.33 Eunapius’ lively portrait of Iamblichus’ associates is based on his own intellectual genealogy: Eunapius himself was an inheritor of Iamblichus’ extended philosophical circle, since he was a pupil of Chrysanthius. Chrysanthius had been a pupil of Aedesius, Iamblichus’ main successor – and was 148 Crystal Addey almost certainly Eunapius’ main oral source, particularly for Sosipatra.34 Eunapius relates that Aedesius established his own philosophical school in Pergamon after Iamblichus’ death.35 Since Chrysanthius had been taught by Aedesius, he was almost certainly also taught by Sosipatra herself, since Eunapius relates that students would attend her classes after those of Aedesius:

After the passing of Eustathius, Sosipatra returned to her own estate, and dwelt in Asia in the ancient city of Pergamon, and the famous Aedesius loved and cared for her and educated her sons. In her own home, Sosipatra held a chair of philosophy that rivalled his, and after attending the lectures of Aedesius, the students would go to hear hers; and though there was none that did not greatly appreciate and admire the accurate learning of Aedesius, they positively adored and revered the woman’s inspired teaching.36

Sosipatra’s inspiration, ἐνθουσιασμός, refers to the theurgic and inspired nature of her philosophical teaching, as well as her divine inspiration and oracular abilities: Eunapius implies that the more advanced students studied with her after preliminary study with Aedesius.37 Theurgy was a kind of ritual practised by many Neoplatonists: it aimed for the soul’s assimilation to and union with the divine, to which end it used divination among other ritual methods. Eunapius also speaks of meeting Maximus, who performs a ritual for Sosipatra.38 Thus, Eunapius had direct access to eyewitness accounts and anecdotes about Sosipatra.39 Furthermore, he locates himself as a link in the ‘golden chain’ of the Platonic philosophers, beginning his collective biography with Plotinus, Porphyry and Iamblichus, whose lives he narrates because of his philosophical lineage and that of his teacher Chrysanthius.40 Thus, Eunapius gives us an ‘insider’s’ account of Iamblichus’ circle – a kind of ‘family tree’ of Iamblichus’ philosophical associates and successors; this strengthens Eunapius’ insights into Iamblichus’ wider philosophical circle.41 Eunapius’ ideological agenda was to promote and reinforce Iamblichus’ philosophical lineage and authority, and by extension that of Iamblichus’ pupils.42 In relation to this, one of the main functions of the biography of Sosipatra is to reinforce the connection between Iamblichus’ school and the Chaldaean tradition – and thus the Chaldaean Oracles.43 However, despite this ideology, Eunapius may well reflect certain features of the historical philosopher Sosipatra. Yet many methodological problems surround this issue, particularly in terms of developing any clear-cut criteria for ascertaining precisely which features of late antique Vitae (if any at all) might be used as sources for historical information. This chapter focuses primarily upon Eunapius’ representation of Sosipatra (and the arguments herein are based upon this representation, they do not depend on her historicity); yet it is important to leave open the question of the historicity of Sosipatra.

Sosipatra the Chaldaean To emphasize the links between Sosipatra and Iamblichus’ philosophical circle, Eunapius includes her biography after those of Iamblichus and his followers Sosipatra: prophetess, philosopher and theurgist 149 Sopater, Aedesius and Eustathius, and before that of her son Antoninus, who also became a philosopher and holy man with oracular abilities.44 Historically, Sosipatra was an indirect successor of Iamblichus’ philosophical circle, through her marriage to Eustathius and her association with Aedesius.45 In relation to her configuration of philosophical and prophetic gifts, Sosipatra is presented as the female counterpart of Iamblichus, an exemplary model of philosophical wisdom and prophetic knowledge. Eunapius relates Sosipatra’s childhood training by Chaldaean priests. Just before these Chaldaean priests depart, they hand over to Sosipatra the ritual garments in which she had been initiated, as well as mystic symbols and books:

The others left the table, and taking the girl with them, they very tenderly and scrupulously handed over to her the whole array of garments in which she had been initiated, and added certain mystic symbols thereto; and they also put some books into Sosipatra’s chest, and gave orders that she should have it sealed.46

Sarah Iles Johnston interprets the Chaldaean priests’ instructions to seal the chest as evidence of Sosipatra’s ritual passivity.47 Yet the sealing of the chest seems to be for the safe-keeping of the sacred books (most especially if these were the Chaldaean Oracles or related Chaldaean sacred books) and ritual implements, since the Chaldaean prophets tell Sosipatra to take care of them until their return.48 The Chaldaean priests also give Sosipatra her ritual garments and symbols, the latter referring to ritual tools commonly employed within theurgic ritual, demonstrating that Sosipatra had participated in ritual practices with the Chaldaean priests.49 The fact that the Chaldaean priests entrusted these ritual garments, books and symbols to Sosipatra implies that they considered her fully ready to continue ritual practices after their departure. To confirm this, Eunapius adds that Sosipatra was fully initiated (τεθειασμένην) and full of divine inspiration (ἐνθουσιῶσαν) after their departure, terms which denote her ritual and divinatory expertise.50 The books which the Chaldaean priests handed over to Sosipatra may well have been the Chaldaean Oracles themselves, which were inextricably linked with theurgy in late antiquity.51 Eunapius implicitly depicts Sosipatra as a theurgist. Although he does not explicitly name her as such, this is largely because of his concern not to divulge initiatory knowledge to the uninitiated.52 We must also take into account the historical and religious circumstances of Eunapius’ time (Vitae Sophistarum was written at the end of the fourth century ce or the beginning of the fifth); he displays some reticence about contemporary pagan religious practices precisely because such practices were threatened by the growing Christianization of the Roman Empire and the corresponding laws of Christian emperors against pagan religious practices.53 In relation to Eunapius’ reticence about divulging initiatory knowledge to the uninitiated, it is vital to note that theurgy was conceptualized in relation to initiation and that the Chaldaean tradition seems to have been a mystery cult of some kind.54 Indeed, Eunapius describes Sosipatra’s ritual training by the Chaldaean priests in exactly these terms: 150 Crystal Addey Then [i.e. the Chaldaean priests] – whether they were heroes or daimones or of some race still more divine – took charge of the child and into what mysteries they initiated her no one knew, and with what religious rite they consecrated the girl was not revealed even to those who were eager to learn.55

Although limitations of space preclude discussion here of the mysterious provenance of the Chaldaean Oracles, it is worth noting that Eunapius depicts the Chaldaean tradition as living and active, exemplified by the Chaldaean prophets.56 In the De mysteriis, Iamblichus also relates a conversation he had with Chaldaean prophets regarding ritual and divinatory knowledge: ‘I shall tell you, indeed, the account I once heard about these matters from the mouths of Chaldaean prophets’.57 This conversation focuses on the nature of the gods, dai- monology, divination and divine assimilation, particularly human ethical and ritual receptivity to the gods, all of which were central to theurgic praxis. All of this adds to the impression that the Chaldaean tradition was a living religious tradition, configured in the form of a mystery cult, of which Sosipatra herself was an active and fully initiated member. Indeed, Eunapius almost certainly intro- duces Sosipatra into his work and provides a lengthy biography of her in order to reinforce and emphasize the close links between Iamblichus’ extended circle and the Chaldaean tradition. As noted above, Johnston claims that Sosipatra is passive because she does not actively perform ritual within the biography and because her divinatory, prophetic episodes happen spontaneously rather than as the result of using physical ritual tools and practices.58 However, as we have seen, Sosipatra was trained in ritual and initiated by the Chaldaean prophets as a child and so ritual practices form a consistent cornerstone of her expertise. Eunapius’ presentation of Sosipatra’s theurgic development ties in well with Iamblichus’ account of the mutually inclusive and cumulative stages and contexts of theurgic ritual. Iamblichus maintains that those who live governed by nature and generation perform ritual using physical and natural objects and tools, while those who live in accordance with intellect practise an intellectual and incorporeal worship, and those in between practise both modes of worship.59 It is vital to note that Iamblichus’ intellectual and incorporeal worship was still conceived as ritual, because it was held to transform the soul rather than merely teach it discursively and because it represented a state of close contact, connection and assimilation to the gods.60 It is also important to note that Iamblichus here describes stages of development and contexts for theurgic ritual practice, rather than discrete ‘types’ of theurgy (often described as ‘lower’ and ‘higher’ levels), since they were considered to be mutually inclusive and cumulative.61 The inclusive and cumulative nature of the stages is based on the concept of receptivity (ἐπιτηδειότης) and its role in enhancing the suitability of the theurgist for contacting the gods, and the idea of the transmigration of souls, rebirth or reincarnation.62 According to this view, those who have attained an advanced level in theurgy have enhanced their receptivity in this life and in previous lives, engaging in each stage of ritual cumulatively in an orderly manner. Theurgy is thus conceptualized as a lifelong and as a multi-lifelong endeavour. Moreover, the Sosipatra: prophetess, philosopher and theurgist 151 stages of ritual were conceived as inclusive, integrating the previous stage rather than discarding it: the more advanced stages of theurgy presuppose expertise in the less advanced stages and in all contexts of ritual practice (including the use of physical tools and practices).63 In relation to its metaphysics, cosmology and psychology, theurgy is characterized by a ‘total unity of spirit and action’ based on its completely unbroken coherence: ‘Rather, those who aspire to unite themselves absolutely with the gods should exercise themselves equally in all its branches, and strive to achieve perfection in all of them’ (ὁμοίως δὲ πᾶσιν ἐγγυμνάζεσθαι καὶ δι’ ὅλων αὐτῶν τελειοῦσθαι χρὴ τοὺς ἐθέλοντας εἰλικρινῶς τοῖς θεοῖς συνάπ- τεσθαι).64 Thus, Sosipatra is implicitly presented as an advanced theurgist who practises the most advanced stage of ritual, intellectual and incorporeal worship.65 Yet her oracular episodes demonstrate that intellectual ritual was presented as bringing practical benefits to humans in daily life.66 Her divinatory episodes are portrayed as spontaneous and instantaneous because her whole life has become ritualistic, in the sense that she is presented as constantly connected with the gods through her receptivity (ἐπιτηδειότης) and service to them, and thus able to hear their oracular advice consistently; in this respect, Sosipatra is presented as similar to Socrates, as will be explored further below. According to Iamblichus, the theurgists had to enhance their receptivity (ἐπιτηδειότης), their suitability to receive the gods: Sosipatra is implicitly presented as advanced in the cultivation of her receptivity, yet receptivity is characterized as an active openness and humility, rather than as a purely passive state. Iamblichus presents advanced theurgists as prophets but conceives of their divination as entailing an instrumental agency, rather than passivity.67 Receptivity is thus construed as a kind of instrumental power for the theurgist which is neither wholly autonomous nor passive, enabling him or her to cultivate and develop the capacity (affinity with and assimilation to the divine) to act as an appropriate instrument for the gods.68 Johnston particularly treats the episode where Sosipatra develops feelings for Philometor, suspects he is performing love spells upon her, and asks Maximus to perform a counteractive ritual for her, as an example of the female philosopher’s passivity.69 However, Sosipatra – as an expert, receptive and humble theurgist – may have considered it inappropriate to perform a counteractive ritual upon an individual for whom she had developed romantic feelings. Eunapius also suggests that Sosipatra asked Maximus to perform this ritual because he was Philometor’s relative.70 Most importantly, instructing Maximus to perform this ritual may place Sosipatra in an educational role as a transmitter of ritual knowledge. Sosipatra may have already been Maximus’ teacher in philosophy, since Maximus was a distinguished pupil of Aedesius and Eunapius tells us that the pupils of Aedesius also attended classes with Sosipatra.71 After Maximus’ ritual, Sosipatra subsequently relates all the details of the ritual to Maximus through her divinatory ‘remote viewing’ ability: ‘[she] described to Maximus his own prayer and the whole ceremony; she also told him the hour at which it took place, as though she had been present, and revealed to him the omens that appeared’.72 This knowledge of Maximus’ ritual and her instruction to Maximus to perform the ritual on her behalf suggest her active mastery of ritual rather than passivity, 152 Crystal Addey an issue we shall return to below. Although Sosipatra is clearly presented as exceptional, the concept of the transmigration of the soul – and the theurgic concept of receptivity – means that other late Platonist philosophers would almost certainly have viewed her abilities as proof of her development in previous lives. In relation to this, Eunapius shows Sosipatra as endowed with natural virtue and higher levels of virtue.73 Sosipatra is presented as an advanced theurgist whose philosophical and theurgic expertise interact and are mutually inclusive.

Sosipatra the theurgist and philosopher Sosipatra’s ability to know of events from a distance, or ‘remote viewing’ of past, present and future events, and her subsequent oracles, show us a theurgist in action. Her ‘remote viewing’ of events is similar to that attributed to Iamblichus himself: perceiving the presence of a funeral procession on the road upon which he and his students are walking as they return from a sacrificial ritual, he advises them to take an alternative route home, so as not to incur religious pollution from the funeral.74 This is also similar to the bi-location attributed to Pythagoras within the biographical tradition, although Pythagoras was reputed to have the ability to be in two different locations at once and to be able to interact and converse with people in both locations.75 Both Pythagoras and the Pythagorean way of life were extremely important for Iamblichus’ philosophical circle.76 Apollonius of Tyana, a neopythagorean sage whom Eunapius also mentions, was reputed to be able to ‘teleport’ or to move to another location instantaneously and several episodes of clairvoyance or psychic power are attributed to him.77 All of these examples are varieties of soul projection which have been compared to the shamanic tradition.78 Eunapius implicitly draws on the legendary soul journeys of these philosophical- religious figures in his portrayal of Sosipatra’s psychic abilities. Sosipatra’s remote viewing and subsequent oracles are connected closely to her philosophical expertise. For example, we hear of Sosipatra giving a lecture in her philosophical school in Pergamon on the nature and descent of the soul, including discussing proofs against various theories, when she suddenly falls silent and then falls into an oracular state, characterized as corybantic (κορυβαντιασμοῦ) and Bacchic (ἐκβακχεύσεως), and exclaims:

‘What is this? Behold my kinsman Philometor riding in a carriage! The carriage [ochêma] has been overturned in a rough place in the road and both his legs are in danger! However, his servants have dragged him out unharmed, except that he has received wounds on his elbows and hands, though even these are not dangerous. He is being carried home on a stretcher, groaning loudly’. These were her words, and they were the truth, for so it actually was. By this all were convinced that Sosipatra was omnipresent, and that, even as the philosophers assert concerning the gods, nothing happened without her being there to see.79

Sosipatra’s omnipresence and her ability to see from the perspective of the gods marks her divine assimilation through both her philosophical and theurgic Sosipatra: prophetess, philosopher and theurgist 153 expertise. There is a close connection between Sosipatra’s philosophical discourse on the descent of the soul and her oracular vision. Porphyry and Iamblichus considered that the descent of the soul into the physical body occurred within ‘the vehicle of the soul’ (ochêma pneuma), an intermediary entity thought to connect the incorporeal soul to the corporeal body.80 Within Neoplatonism, this descent into the physical body and mortality was considered to represent a kind of inver- sion or reversal of the soul’s proper and perfected state, a forgetting of its true home with the gods: this philosophical notion is reflected in the overturning of Philometor’s carriage (ochêma).81 So, while she is speaking philosophically about the descent of the soul, which almost certainly included discussion of the ochêma pneuma, Sosipatra falls into an oracular state in which she sees Philometor’s carriage overturn: not only does her philosophical discourse blend seamlessly with her oracular state, with philosophical thought affecting oracular vision, but her oracular vision reflects her discourse and vice versa. Sosipatra’s oracular episodes also depict her as an expert on the kairos, the best or most appropriate moment for any given action, undertaking or utterance. Eunapius portrays her prophesying to Eustathius the number of children she will have with him and their destinies. She then tells Eustathius about his death and his destiny after death and begins to relate her own destiny to him:

‘But you will go hence before me, and be allotted a fair and fitting place of abode, though I perhaps shall attain to one even higher. For your station will be in the orbit of the moon, and only five years longer will you devote your services to philosophy – for so your phantom tells me – but you shall traverse the region below the moon with a blessed and easily guided motion. Fain would I tell you my own fate also’. Then, after keeping silence for a short time, she cried aloud: ‘No, my god prevents me!’82

In using the phrase, ‘No, my god prevents me!’ (ἀλλ’ ὁ ἐμός θεός με κωλύει), Eunapius explicitly connects Sosipatra to Socrates and Plotinus. In his biography of the latter, Porphyry relates an episode in which he and Plotinus are invited to a ritual for the invocation of the personal or guardian daimon in the Iseum in Rome: when the priest invoked Plotinus’ personal daimon to appear, Porphyry relates that a god rather than a daimon appeared, implying Plotinus’ superior status as a divine man.83 Sosipatra’s phrase is intended to evoke the same associations and to reinforce her status as a divine woman, an advanced soul who has greatly developed her assimilation to the divine. This formulation also evokes the precedent of Socrates, who is portrayed in various Platonic dialogues as receiving various negative admonitions and warnings from his daimonion, his divine ‘voice’ or ‘sign’, which Plutarch, Apuleius and Proclus interpreted as Socrates’ personal daimon.84 In being able to hear and follow the advice of her ‘god’, Sosipatra is presented as a wise sage akin to Socrates. Moreover, Proclus interprets Socrates’ daimonion as giving the philosopher access to the kairos, the right or best moment for any given utterance, undertaking or project to occur.85 Proclus relates the kairos to cyclical time and to the operation of divine providence throughout the cosmos: he states 154 Crystal Addey that the kairos represents the right moment when the good and purpose of a given action (both of which were conceived as stemming ultimately from divine providence) is most likely to be fulfilled and for the action to have the best results.86 Eunapius also depicts Iamblichus as an expert of the kairos in relation to theurgic ritual; when the philosopher’s students ask him to demonstrate his ritual expertise, he replies: ‘Nay, that does not rest with me, but wait for the appointed hour’ (“ἀλλ’ οὐκ ἐπ’ ἐμοί γε τοῦτο” ἔλεγεν, “ἀλλ’ ὅταν καιρὸς ᾖ”).87 Later (presumably at the time or moment which Iamblichus judges to be the kairos), during a visit to the baths at Gadara, Iamblichus pulls two spirits, Eros and Anteros, out of the spring.88 Sosipatra, like Socrates and Iamblichus, is presented as an expert on the kairos, who knows both when to speak and when not to speak. In relation to the utility and benefits of divination for humans, Iamblichus claims:

For the gods grant the power of defence against the dangers which menace us from the natural order. And when it is necessary to exercise virtue and an uncertainty of future events contributes to this, then (the gods) conceal what will be for the improvement of the soul. But whenever this (uncertainty) does not matter for this purpose, and foreknowledge rather is advantageous to souls for saving and leading them upwards, then the gods implant in the midst of their essences the foreknowledge inherent in divination.89

On this view, the wise and skilful seer will know when to speak the oracular utterance, and when is the right time to remain silent, because he or she is envisaged as communicating directly in some way with the gods, who convey a sense of when the oracular utterance will enhance human arête (excellence or virtue) and when it will inhibit such ethical development. Maximus’ counteractive ritual is relevant here, since Eunapius’ later biography of Maximus suggests that the latter had considerable ritual power but extremely bad timing and possibly some lack of judgement in its deployment.90 This suggests the possibility that Sosipatra recognized Maximus’ charisma and theurgic potential – and thus asked him to perform the counter-active ritual in her role as a ritual educator and teacher, as an attempt to teach him subtly about the appropriate timing, ethics and contexts of theurgic ritual practice.91 Indeed, when asking Maximus to perform the ritual, Sosipatra exhorts him to demonstrate his piety.92 After Sosipatra relates the full details of the ritual through her oracular ‘remote viewing’, and Maximus is amazed by her oracular accuracy, Sosipatra advises him: ‘The gods love you if you raise your eyes towards them and do not lean towards earthly and perishable riches’, a typically Platonic protreptic exhortation relevant to theurgic ritual which, according to its proponents, focused on the divine, and involved endurance and the rejection of physical distractions and luxury, including desires for personal gain.93 These words demonstrate her subtle guidance of Maximus’ own ritual power and expertise, and her attempt to instil in him the humility, caution and focus on divine guidance needed by the successful theurgist within ritual practices. In her transmission of ritual, spiritual and philosophical expertise, the portrayal of Sosipatra could be associated with – and may be modelled on – Plato’s Diotima, Sosipatra: prophetess, philosopher and theurgist 155 who is also presented as a wise philosopher and prophetess (Symp. 201d3–7; 206b11–12) who teaches Socrates the philosophic path in the Symposium (201d– 12c). Diotima is presented as more elenctic than Sosipatra, in the sense that the former’s speech, as reported by Socrates, features extensive use of philosophic elenchus (Symp. 201e4–203a10; 204d1–208c2), with Diotima as questioner and Socrates taking on the role of interlocutor. However, Socrates’ report of Diotima’s discourse also features the myth of the birth of Eros (Symp. 203b1–204a9) and sections of monologue (Symp. 204b1–c7; 208c3–212a10). Eunapius does not explicitly show Sosipatra engaging directly in elenchus with her students, but this difference can be partially explained by the conventions of the literary genre of the works which feature these women: Plato frequently uses Socratic elenchus within his dialogues, while ancient biographical works, even biographies of philosophers, rarely, if ever, feature philosophic elenchus. Overall, little evidence remains of Sosipatra’s teaching methods, which are difficult to ascertain, although Eunapius’ account of her discourse on the nature and descent of the human soul, particularly her refutation of various theories, suggests that dialectic and dialogue were an important aspect of her teaching methods.94 Given that Eunapius presents her as an indirect successor of Iamblichus’ circle and in relation to Neoplatonism, Plotinus’ and Iamblichus’ use of dialogue and dialectic as an important element of their teaching methods is indicative of the likelihood that Sosipatra used dialogue, dialectic and possibly elenchus in her own teaching practices.95

Conclusion Sosipatra neatly exemplifies the close and mutually inclusive relationship between divination and philosophical forms of knowledge in theurgic iterations of Neoplatonism in late antiquity. Her oracular expertise and psychic, ‘remote viewing’ ability puts her on a par with the Pythia at Delphi and the priest of the Oracle of Claros, yet she was also the head of a philosophical school in Pergamon. Eunapius presents her as a paradigmatic role model for late antique Platonist philosophers and theurgists (both male and female), an exceptionally gifted ‘person of power’ just as Pythagoras, Socrates, Diotima and Plotinus were often characterized in ancient philosophical and biographical tradition. Within Eunapius’ work, Sosipatra explicitly acts as a role model for Maximus in teaching him about the appropriate ethics and contexts of theurgic ritual. In addition, Eunapius’ Sosipatra (and perhaps even the historical philosopher and prophetess herself) almost certainly acted as an important role model for the rather similar theurgist and philosopher, Asclepigeneia, who over a century later trained Proclus, the head of the Athenian School, in theurgy. As we have seen, Marinus reports that Asclepigeneia was the daughter of Plutarch of Athens who had been trained by her own father in theurgy, handing on to her the ritual tradition from Nestorius.96 Marinus tells us that she instructed and taught Proclus to use the conjunctions and ritual sup- plications of the Chaldaeans, a clear reference to theurgic ritual practices as well as the Chaldaean tradition.97 Besides being the hierophant at Eleusis (from c. ce 355–380), Nestorius had also been a seer (with a reputation for giving accurate 156 Crystal Addey oracles), which suggests that Asclepigeneia inherited instruction in divination as part of her theurgic training.98 As an expert teacher of theurgic ritual and philosophy, Sosipatra acts as a central role model for Asclepigeneia, Maximus and her other students. Rather than conceiving of Sosipatra’s philosophical and theurgic (which includes oracular) expertise as dichotomous or discrete, her abilities are presented as mutually inclusive, as intimately related in the sense that they interact, reinforce and enhance each other. Sosipatra’s spontaneous oracular visions are intended to demonstrate her divine assimilation and closeness to the gods: her ritual is spontaneous and instantaneous because she has moved to the highest level of theurgic virtue and practice; she is presented as an advanced theurgist who has enhanced her receptivity and thus maintains a constant connection to the gods. Rather than being ‘passive’ in relation to ritual, Sosipatra’s whole life becomes ritualistic and she becomes a powerful instrumental agent of the gods. In this sense, her life is presented as an invocation of the gods, while her philosophical teaching and spontaneous divinatory activities are presented as service to the gods.

Notes 1 Earlier drafts of this chapter were presented at the International Society for Neoplatonic Studies Conference 2016, Seattle University, and at the workshop ‘God and the gods in Late Antiquity’, University of St Andrews: I thank all delegates for their useful comments and feedback. I also wish to thank Jill Harries, John Finamore, Robert Berchman, Victoria Leonard, José Manuel Redondo and Regina Fichera for reading and commenting on this chapter. 2 Eunapius, Vitae Sophistarum (VS), ed. J. Giangrande, Eunapii vitae sophistarum, Rome 1956. All citations from Eunapius’ work are drawn from this edition, while translations are drawn from W.C. Wright, Eunapius: Lives of the Philosophers/Lives of the Sophists (1921 [2005]). For discussion of the dating of this work see Penella (1990) 9–23. 3 Eunapius, VS 6.6.1–6.9.17 (466–471). See also VS 6.6.5.1–2 (466); 6.7.5.1–3 (468); 6.8.1.1–2.1 (469); 6.8.3.1–3 (469); 6.9.1.1–6 (469); 6.9.7.4–5 (470); Penella (1990) 60–1; Clark (1993) 133; Lanzi (2004) 275; Johnston (2012) 101; Tanaseanu-Döbler (2013a) 157–9; (2013b) 128–30; Denzey Lewis (2014) 276–7. 4 VS 6.7.3.3–4.5 (468); 6.8.1.2–3 (469); 6.8.3.4–6.3 (469); 6.9.7.1–4 (470); 6.9.12.6–15.1 (470): Penella (1990) 50, 56–9; Lanzi (2004) 275; Addey (2014a) 2. 5 Dodds (1973) 173; Denzey Lewis (2014) 276–7. 6 Pythia at Delphi: Plutarch, De garrulitate 20 (512e4–7); Herodotus 1.47–8, 1.65, 5.92. Oracle of Amphiaraus: Herodotus 1.49. Priest at the oracle of Claros: Tacitus, Annals 2.54; Dodds (1973) 168. 7 VS 6.8.2.1–6 (469); 6.9.1.1– 2.6 (469). 8 Johnston (2012) 110–14. 9 In this line of interpretation, Tanaseanu-Döbler (2013b: 134) is influenced by Johnston’s view of Sosipatra’s passivity. 10 Marinus, Proc. 28.8–15. As Edwards (2000: 100 n. 295) notes, Asclepigeneia was the daughter of Plutarch, the head of the Athenian School of Neoplatonism (he was also the teacher and predecessor of Syrianus, and Proclus’ teacher when he first arrived in the Athenian School) and she became the mother of Archiadas (Proclus’ closest friend, according to Marinus), whose daughter Asclepigeneia the younger married Theagenes and become the mother of the scholarch Hegias. The name attests to the family cult of Asclepius. See also Marinus, Proc. 12 (Proclus studies with Plutarch after his arrival in Sosipatra: prophetess, philosopher and theurgist 157 the Athenian School); 17.21–31 (Proclus’ close friendship with Archiadas, the son of Asclepigeneia); 29.16–28 (when she is very ill, Proclus prays at the shrine of Asclepius for Asclepigeneia the younger, the grand-daughter of Asclepigeneia the Theurgist. According to Marinus, she subsequently recovers from her serious illness). 11 Marinus, Proc. 28.13–15. Trans. Edwards (2000): παρ’ αὐτῇ γὰρ καὶ μόνῃ ἐσώζετο ἀπὸ Νεστορίου τοῦ μεγάλου ὄργια καὶ ἡ σύμπασα θεουργικὴ ἀγωγὴ διὰ τοῦ πατρὸς αὐτῇ π- αραδοθεῖσα. 12 Lanzi (2004) 277–8. On the development of biography in antiquity, see Momigliano (1971); Swain (1997) 22–37; Cox (1983) 3–16. 13 On female representation within male biographical discourse, see Clark 1993, 2–4, 133–4; Swain (1997) 14–15, 35 n. 91. 14 Iamblichus, Letter 3: To Arete, On Self-Control; Marinus, Proclus 28.12–15. See also Addey (2017). 15 Hawley (1994) 70–1. 16 VS 6.6.6.1–3 (466). See also Tanaseanu-Döbler (2013b) 128. 17 VS 1.1.2.3–5 (453). 18 Brown (1971) 99–100; Fowden (1977) 359–82; (1982) 33–59; Cox (1983). For a detailed discussion of the topos of the ‘holy man’ or ‘divine man’ or woman in relation to Sosipatra, see Tanaseanu-Döbler (2013b) 125–8. 19 For a comprehensive summary of the debate see Denzey-Lewis (2014) 278–9; Tanaseanu-Döbler (2013b) 125–8. Bingham Kolenkow (2002: 133–44) prefers to use the term ‘persons of power’, considering it more gender-neutral and appropriate. 20 See also for example, Pack (1952) 198–204; Lanzi (2004) 277; Johnston (2012) 99–100; Tanaseanu-Döbler (2013b) 129. 21 Johnston (2012: 102–3) adduces important parallels with the Homeric Hymn to Demeter. See also Denzey-Lewis (2014) 287. 22 Johnston (2012) 103–4. 23 See Cox (1983: xiii–xvi) on the mythic nature of late antique biographies of ‘holy men’: ‘in ancient biographies of holy men, the heroes come alive in the biography’s allusions, images, patterns, and themes. These are the interpretative “gestures” of the biographer himself, and they make possible that interplay between the mundane and the ideal, the earthly and the heavenly, by which the soul-revealing gestures of the hero are captured’. 24 Eunapius refers to the use of allegory and philosophers’ use of obscurity to hide secret, mystical ideas: ‘some philosophers hide their esoteric teachings in obscurity, just as poets conceal theirs in myths’: τῶν δὲ φιλοσόφων τὰ ἀπόρρητα καλυπτόντων ἀσαφείᾳ, καθάπερ τῶν ποιητῶν τοῖς μύθοις (VS 4.1.9.3–4 (456)). 25 For a representative example, see Clark (1998) 1–31. Although beyond the scope and parameters of the current study, Clark’s challenges and critique of the extrapolation of historical information from late antique Vitae deserve further examination. 26 Ammianus Marcellinus 22.7.3; 23.5.10–14; 23.3.23; 29.1.42; Iamblichus, Letter 7: To Eustathius, On Music = Stobaeus, Anth. 2.31.117. See also the commentary of Dillon and Polleichtner (2009) 72–3. 27 As noted by Johnston (2012) 100 n. 8. 28 Iamblichus, Letter 3: To Arete, On Self-Control = Stobaeus, Anth. 3.5.9, 45, 47–50. On the presence of women in the Neoplatonic schools, see Addey (forthcoming). 29 Dixon (2001) ix–xi, 5–7, 13–15, 16–25; Richlin (2014) 7–8, 108, 289–317. 30 Clark (1998: 21–30) has analysed the tendency of late antique Vitae to present their her- oines as teachers of wisdom. See also the conclusions of Richlin’s study of the Roman use of women as political icons (2014: 108–9). Dixon writes that ‘each text is designed to project ideology (e.g. of proper womanly behaviour) rather than circumstantial information about any given women, even when it purports to record a specific, historicised woman’ (2001: ix). 158 Crystal Addey 31 For example, Watts (2005: 334) notes that Eunapius’ work ‘has long been recognized as an important source for the intellectual history of the fourth century’. 32 Watts (2005) 334–61. 33 Watts (2005: 343–50, 359–61) notes that Eunapius considered oral traditions handed down specifically from teacher to student within an intellectual community as stable traditions (360). 34 VS 5.1.11.1–3 (458); VS 5.3.10.4–5 (461); 23.1.5–7 (500). Cf. Iamblichus Letters for a vivid glimpse of Iamblichus’ philosophic community. See also Athanassiadi (2013) 13–27; Penella (1990) 1, 5, 30–2; Watts (2005) 338, 345–50. Watts writes: ‘Chrysanthius also stands as the only known connection between his occasional teacher Sosipatra and Eunapius’ (2005: 349). 35 VS 6.4.7.2–4 (465). 36 VS 6.9.1.1–2.5 (469): Σωσιπάτρα, μετὰ τὴν ἀποχώρησιν Εὐσταθίου, πρὸς τὰ αὑτῆς ἐπ ανελθοῦσα κτήματα, περὶ τὴν Ἀσίαν καὶ τὸ παλαιὸν Πέργαμον διέτριβεν· καὶ ὁ μέγας Αἰδέσιος θεραπεύων αὐτὴν ἠγάπα, καὶ τοὺς παῖδας ἐξεπαίδευε. καὶ ἀντεκάθητό γε αὐτῷ φιλοσοφοῦσα κατὰ τὴν ἑαυτῆς οἰκίαν ἡ Σωσιπάτρα, καί, μετὰ τὴν Αἰδεσίου συνουσίαν, παρ’ ἐκείνην φοιτῶντες, οὐκ ἔστιν ὅστις τὴν μὲν ἐν λόγοις ἀκρίβειαν Αἰδεσίου <οὐ> περιηγάπα καὶ συνεθαύμαζεν, τὸν δὲ τῆς γυναικὸς ἐνθουσιασμὸν προσεκύνει καὶ ἐσεβάζετο. See also Watts (2005) 339, 342, 349–50; Johnston (2012) 107. 37 See also VS 6.8.12–13 (469). Fowden (1982: 39) describes Sosipatra’s philosophical circle as ‘a relatively closed coterie of Chaldaean adepts’. 38 VS 7.1.1.1–4.2 (473); Penella (1990) 8. 39 Eunapius’ extensive use of oral sources and testimony: VS 1.1.6.1–8 (453); 2.2.5.3 (455); 5.1.11.1–3 (458); 23.1–2 (500); Penella (1990) 23, 30–1; Watts (2005) 335–6, 346–8. 40 Biography of Plotinus: VS 3.1 (455); biography of Porphyry: VS 4.1–4.3 (455–7); biography of Iamblichus: VS 5.1–5.3 (457–61). Eunapius (VS 4.1.11.1–2 (457)) uses the image of the ‘golden chain’ to describe Porphyry (this image was drawn from Iliad 8.19 and symbolized for Neoplatonists the succession of the philosophers of their school). For his location of himself within this chain of transmission, see VS 5.11.1–12.1 (458); 6.1.6.3–4 (461); 23.12 (500). See also Watts 2005, 336, 338–9; Cox Miller 2000, 212–13; 237. 41 The term ‘family tree’ is taken from Cox Miller’s description of Eunapius’ portrait of Iamblichus’ extended circle (2000: 239. See also Fowden (1982) 37; Buck (1992) 141; Watts (2005) 345; Penella (1990) 31–3; Lanzi (2004) 278 n. 12. 42 Penella (1990) 32–3; Johnston (2012) 100. 43 Penella (1990) 39, 59–60; Watts (2005) 334–5, 338–9. 44 Antoninus: VS 6.9.15.1 – 17.8 (470-71). 45 VS 5.1.5.2 (no. 458); 6.1.1.1–2; 6.4.6. See also Fowden (1977) 376–7; Penella (1990) 49; Cox Miller (2000) 239. 46 VS 6.7.8–9 (468): οἱ δὲ ἀποχωρήσαντες τοῦ δείπνου καὶ τὴν παῖδα παραλαβόντες, τήν στολὴν τῆς ἐσθῆτος ἐν ᾗ τετέλεστο μάλα φιλοφρόνως αὐτῇ καὶ συνεσπουδασμένως παρέδοσαν, καὶ ἄλλα τινὰ προσθέντες ὄργανα καὶ τὴν κοιτίδα τῆς Σωσιπάτρας, κατασημήνασθαι κελεύσαντες καὶ προσεμβαλόντες τινὰ βιβλίδια. 47 Johnston (2012) 110. 48 VS 6.7.11.1–4 (468). 49 Penella (1990) 60 n. 47. 50 VS 6.8.1.2–3 (469). See also Fowden (1982) 37; Addey (2014a). 51 See also Pack (1952) 203; Penella (1990) 58–62; Cox Miller (2000) 237–9; Johnston (2012) 114; Urbano (2013) 258; Denzey-Lewis (2014) 280–3. 52 Eunapius displays concern about divulging initiatory knowledge to the uninitiated on several occasions. In his biography of Porphyry: ‘As for [his expertise] in natural philosophy and theurgy, let that be left to sacred rites/initiations and mysteries’: τὸ δὲ Sosipatra: prophetess, philosopher and theurgist 159

φυσικὸν καὶ θεουργὸν τελεταῖς ἀφείσθω καὶ μυστηρίοις·(VS 4.2.3.2–3 (457)). See also VS 6.1.5.4–6 (461). 53 Eunapius was born around 347 ce and wrote VS around 396–405 ce. See also VS 6.1.5.2–4 (461); 6.4–5 (461); 6.10.7.1–3 (471). Swain (1997: 35) writes: ‘Late pagan biography, however, looks more like an attempt to document an increasingly threatened way of life (and one which belongs to a different social and economic level than we see in many of our Christian texts)’. See also Fowden (1982) 33–59; Lanzi (2004) 277. 54 VS 6.7.5.4–6.7.6.1 (468); Lanzi (2004) 287. See also 7.3.1.3–2.1 (475) for Eunapius’ own initiation into the Eleusinian Mysteries. Eunapius has the Chaldaean prophets characterize their ritual knowledge as ἀπόρρητα (VS 6.6.10.4–5 (467)), a term frequently used to denote theurgic expertise and a traditional designation for mystery cults in antiquity. Denzey-Lewis (2014: 275–6) examines Eunapius’ avoidance of labelling Sosipatra in relation to the ritual context; he avoids using terms connected with either religion or magic. Denzey-Lewis argues that Eunapius’ narrative was constructed to defend Sosipatra from the charge of magic. 55 VS 6.7.1–2 (467): Οἱ δὲ παραλαβόντες τὸ παιδίον εἴτε ἥρωες, εἴτε δαίμονες, εἴτε τι θειότερον ἦσαν γένος τίσιν μὲν συνετέλουν αὐτὴν μυστηρίοις ἐγίνωσκεν οὐδὲ εἷς, καὶ πρὸς τί τὴν παῖδα ἐξεθείαζον ἀφανὲς ἦν καὶ τοῖς πάνυ βουλομένοις εἰδέναι. Trans. Wright (1921) with my own emendations. 56 On the Chaldaean Oracles, see Lewy (1956); Finamore and Johnston (2010) 161–73. 57 Iamblichus, DM 3.31 (176.1–2). 58 Johnston (2012) 110. 59 Iamblichus, DM 5.18 (224.6–225.9). 60 See Shaw (1985) 22–6; Addey (2014a) 39–40, 196–9 contra Rosán (1949) 213–14; Sheppard (1982) 217–18. Smith (1974: 90) also has reservations about distinguishing between ‘types’ of theurgy in terms of ritual. 61 ‘Lower’ and ‘higher’ types of theurgy. See also for example, Rosán (1949) 213–14; Lewy (1956) Excursus iv; Smith (1974) 90–1, 111–21; Sheppard (1982) 217–18. 62 Rebirth: Iamblichus, DM, 4.4 (186.9–187.1). Receptivity: DM 3.12 (129.9–11); Marinus, Proclus 5.1–2; George (2005) 290–3; Addey (2013) 14–20. 63 Iamblichus, DM 5.14 (217.8–13); 5.16 (221.1–222.3); 5.17 (222.4–11); 5.20 (227.11– 228.10); 5.22 (231.7–12). See also: Addey (2014a) 39–40. 64 Iamblichus, DM 5.26 (240.9–14). 65 For a later example of Proclus’ rapid progress through the scale of virtues and his advanced practice of theurgic ritual, see Marinus, Proclus 18; 22.12; 28. Marinus uses the same term, ἐνθουσιασμός (Proclus 22.20), to denote Proclus’ inspiration, which Eunapius uses to characterize Sosipatra. 66 For Proclus’ use of theurgic ritual to cause rains to release Attica from drought, to defend against earthquakes and to heal Asclepigeneia with Asclepius’ aid, see Marinus, Proclus 28–9. 67 On instrumental agency in Iamblichus’ account of the operation of divine inspiration and divination, see DM 3.3 (107.7–11); 3.5 (111.7–11); 3.12 (129.9–11). See also Addey (2013) 7–24 esp. 13–15, 20–2. The term ‘instrumental agency’ is taken from Keller’s 2002 analysis of spirit possession. 68 See Addey (2013) 20–2; Keller (2002) 9–10. Keller writes: ‘the concept of instrumental agency serves to highlight the way that receptivity has often been evaluated as an extremely powerful capacity among possession traditions. Rather than coding receptivity negatively as a type of passivity, instrumental agency accounts for this revaluation; one’s receptivity marks a developable sacred space’ (2002: 82). 69 VS 6.9.3.1–11.1 (469–70). This episode is based on Plotinus’ counteracting of Olympius’ magical attacks reported by Porphyry. See Vita Plotini 10; Johnston (2012) 110–12. 70 VS 6.9.3.5–6.9.4.1 (469). 71 VS 6.9.2.1–6 (469); 6.9.3.5–6 (469). 160 Crystal Addey

72 VS 6.9.7.1–4 (470): καὶ τήν γε εὐχὴν ἀπήγγειλε τῷ Μαξίμῳ καὶ τὴν ἅπασαν πρᾶξιν, καὶ τήν γε ὥραν προσέθηκεν, ὥσπερ συμπαροῦσα, καθ’ ἣν ταῦτα ἔπραττεν, καὶ τὰ φανέντα ἀνεκάλυψε σημεῖα. 73 See also Finamore (2012) 115. Compare Marinus’ presentation of Proclus’ progress through and attainment in the scale of virtues. See also Proclus 18; 21–2. 74 VS 5.1.13.1–15.1 (458–9). Porphyry (Vita Plotini 11) reports an episode in which Plotinus was able to tell which of the slaves had stolen a necklace belonging to Chione. He also states that Plotinus was in the habit of foretelling how adopted orphans would turn out, and that he noticed and offered advice when Porphyry was depressed and thinking of committing suicide. 75 Aristotle, On the Pythagoreans F191 (Apollonius, Historiae Mirabiles 6); Porphyry, Vitae Pythagorae 28–9. 76 Iamblichus characterizes Pythagoras as a divine soul sent to humans under Apollo’s leadership (De Vita Pythagorica 2.7–8). 77 Philostratus, Life of Apollonius 4.10; 5.24; 8.26–7; Eunapius, VS 2.1.3.5–4.1. 78 See also Dodds (1951) 140–7; (1965) 171. 79 VS 6.9.12.6–15.1 (470): “τί τοῦτο;” ἀνεβόησεν εἰς μέσους· “ὁ συγγενὴς Φιλομήτωρ φερόμενος ἐπ’ ὀχήματος, τό τε ὄχημα κατά τινα δυσχωρίαν περιτέτραπται, κἀκεῖνος κινδυνεύει περὶ τὼ σκέλη· ἀλλ’ ἐξῃρήκασί γε αὐτὸν οἱ θεράποντες ὑγιαίνοντα, πλὴν ὅσα περὶ τοῖς ἀγκῶσι καὶ χερσὶ τραύματα εἴληφε, καὶ ταῦτά γε ἀκίνδυνα· ἐπὶ φορείου δὲ φέρεται ποτνιώμενος”. ταῦτα ἔλεγεν καὶ εἶχεν οὕτως, καὶ πάντες ᾔδεσαν ὅτι πανταχοῦ εἴη Σωσιπάτρα, καὶ πᾶσι πάρεστι τοῖς γινομένοις, ὥσπερ οἱ φιλόσοφοι περὶ τῶν θεῶν λέγουσιν. 80 See also Proclus, In Tim V: 234.18–24 (311A) Diehl, for Porphyry’s view of the soul vehicle; Iamblichus, In Tim Fr. 49; Fr. 84 Dillon. See also Finamore (1985: 59–123) for a comprehensive discussion of the role of the soul vehicle in the descent of the soul, according to Porphyry and Iamblichus. 81 See also Plotinus, Enn. 1.2[19].1.1–6 citing Plato, Theaetetus 176a–b. 82 VS 6.8.4–5 (469): καὶ σὺ δὲ προαπολείψεις ἐμέ, καλὴν μεταλαχὼν λῆξιν καὶ πρέπουσαν, ἐγὼ δὲ ἴσως κρείσσονα. σοὶ μὲν γὰρ περὶ σελήνην ἡ χορεία, καὶ οὐκέτι λατρεύσεις καὶ φιλοσοφήσεις τὸ πέμπτον, οὕτω γάρ μοί φησιν τὸ σὸν εἴδωλον, ἀλλὰ καὶ τὸν ὑπὸ σελήνην παρελεύσῃ τόπον σὺν ἀγαθῇ καὶ εὐηνίῳ φορᾷ· ἐγὼ δὲ καὶ ἐβουλόμην μὲν εἰπεῖν τὰ κατ’ ἐμαυτήν”, εἶτα ἐπισιωπήσασα τῷ λόγῳ βραχύν τινα χρόνον, “ἀλλ’ ὁ ἐμός”, ἀνεφθέγξατο, “θεός με κωλύει”. 83 Porphyry, Vita Plotini 10.1534. Porphyry claims here that this ritual experience in the Temple of Isis led Plotinus to write the treatise On Our Allotted Guardian Daimon (Enn. 3.4). 84 Plato, Euthydemus, 272e1–5; Republic VI, 496c4; Phaedrus 242b8–c7; Apology 40a4–b6, 40c2–4, 41d5–6; Theaetetus 151a1–6; Ps.-Plato, Alcibiades I, 103a4–7, 105e6–106a1; Theages 129e1–130a5; Apuleius, De Deo Socratis, 16–17; Plutarch, On the Sign of Socrates, 588e3–5; Proclus, In Alc., 71.1–3, 72.18–20, 78.1–5. See also Johnston (2012) 105; Addey (2014b) 51–72. 85 Proclus, In Alc 121.1–11 (esp. 7–9); 79.15–80.3; 80.19–82.16; 93.21–4: 158.3–159.10. Within Greek religious and cultural traditions, the kairos was often linked to divination specifically since many kinds of divination – especially oracles and omens – aimed to discover the best or most appropriate time for a range of activities and undertakings. 86 Proclus, In Alc., 121.1–18. 87 VS 5.2.1.5–2.1 (459). See also Iamblichus on the centrality of observing the kairos in theurgic ritual: DM 8.4 (267.6–10). 88 VS 5.2.1.5–5.2.7.4 (459). 89 Iamblichus, DM, 10.4 (289.12–290.3): Οἱ γὰρ θεοὶ καὶ δύναμιν τοῦ φυλάξασθαι τὰ ἐπιόντα ἀπὸ τῆς φύσεως δεινὰ παραδιδόασι· καὶ ὅταν μὲν ἀσκεῖν δέῃ τὴν ἀρετὴν καὶ σ υμβάλληται πρὸς τοῦτο ἡ τοῦ μέλλοντος ἀδηλία, ἀποκρύπτουσι τὰ ἐσόμενα ἕνεκα τοῦ Sosipatra: prophetess, philosopher and theurgist 161

τὴν ψυχὴν βελτίονα ἀπεργάζεσθαι· ὅταν δὲ πρὸς τοῦτο μηδὲν διαφέρῃ, λυσιτελῇ δὲ ταῖς ψυχαῖς τὸ προγιγνώσκειν, ἕνεκα τοῦ σώζειν αὐτὰς καὶ ἀνάγειν, τὴν ἐν ταῖς μαντείαις πρόγνωσιν ἐν μέσαις αὐτῶν ταῖς οὐσίαις ἐντιθέασιν. 90 Maximus performs theurgic animation of a statue of Hekate in front of Eusebius of Myndus, presented as a sceptic regarding theurgy (performing theurgic ritual in front of sceptics was probably ill-advised for the cautious and humble theurgist) (VS 7.2.6.1– 10.6 (475)). When Julian invites Chrysanthius and Maximus to join him, they consulted the gods but the omens were hostile and forbidding. Therefore Chrysanthius decided not to go but Maximus refused to accept the omens and consulted the gods repeatedly until apparently favourable signs appeared – he is reported to speak approvingly of compelling the gods to provide favourable omens (VS 7.3.9.5–14.1 (476–7); 23.2.1–5.1 (500–1)). Here, Maximus puts his personal desire before the advice of the gods (VS 23.2.4.1–5.1) and uses manipulation to attempt to compel the gods. See also Iamblichus (DM 1.12; 3.17–18) on the impossibility of compelling the gods, a practice which he associates with magic rather than theurgy. Cf. VS 7.4.2.1–3.1 (477), where Maximus grows arrogant, insolent and luxurious in Julian’s court; VS 7.6.2–8 (480), where Maximus’ arrogant and careless display of his theurgic powers leads others to become envious and accuse him of involvement in a famous divinatory episode where the name of the next emperor was revealed. As a consequence, Maximus was hunted down and killed brutally by the emperor’s assassin, Festus. Ammianus Marcellinus (29.1.42) also reports the latter incident, as well as Maximus’ contemptuous dismissal of the warnings of Etruscan diviners regarding certain seemingly unfavourable omens during Julian’s advance into Persia (23.5.10–14). 91 VS 6.9.3.1–11.1 (469–70). Cf. Iamblichus, DM 8.4 (267.6–10). 92 VS 6.9.5.3–6.9.6.1 (469–70). 93 VS 6.9.7.6 (470): “θεοί σε φιλοῦσιν, ἐὰν σὺ πρὸς ἐκείνους βλέπῃς καὶ μὴ ῥέπῃς ἐπὶ τὰ γήϊνα καὶ ἐπίκηρα χρήματα”. 94 VS 6.9.11.3–12.5 (470). 95 Iamblichus: VS 5.3.4.1–3 (460). Plotinus: Porphyry, Vita Plotini 3.36–8; 13.11–18; 18.7–10. See also Addey (2014a) 134–5. 96 Marinus, Proc. 28.13–15. See also 28.10–13; 29; 14.19–27; 12.28–37); n. 10 above. 97 Marinus, Proc. 28.8–13. 98 Plutarch of Athens was the son or grandson of Nestorius. See Marinus, Proc. 12.1–3; 28.13–15; Eunapius, VS 7.3.2–4 (475–6), 493; Zosimus 4.18. See also Burkert (1987) 50–51, 85 n. 114, 113–14; Clinton (1974) 43; Edwards (2000) 73 n. 112; Addey (2014a) 280 n. 181. 12 One oracle too many? Corippus (and Procopius) on female prophecy in North African divination and profit in the Roman world1

Martine de Marre

Introduction The Iohannis or De Bellis Libycis was composed by the North African Flavius Cresconius Corippus2 around 550 ce and narrates in epic style the struggles of Justinian’s Byzantine governor, John Troglita, against the Mauri until his final victory over the African tribes in 548. As an epic, the poem borrows freely from a number of predecessors in the genre, from Vergil to Claudian, as has been demonstrated by various modern scholars.3 On the other hand, Corippus’ work is also firmly rooted in the historical events of his own time, and his epic is also utilized by historians (together with the works of Procopius) to point to the nature of the turbulence in Africa as Justinian’s forces strove to regain the African provinces.4 What are we to make, then, of the two passages (Ioh. 3.85–151 and 6.152–87) which describe the Africans solicit- ing prophecies from oracles? Are they simply a literary topos, imitating well- known oracular examples from the epics of Vergil, Lucan, Statius, Silius Italicus or Claudian? Or can we use them, as some scholars have done, as evidence that the oracle of Ammon was still active in the sixth century?5 What can we identify as literary artifice and which elements represent a verifiable aspect of local oracular practice? The present article intends to sift through the evidence from other ancient writers, such as Procopius, to attempt to answer these questions.

Historical and historiographical context The early part of the century in which Corippus’ epic is set saw a confederation of North African tribes known as the Laguatan rout the Vandals in 523 ce under the leadership of Cabaon (Procop. Vand. 3.8.15–29).6 The Laguatan had been moving steadily westwards, and most of the interior of Cyrenaica, particularly the desert oases (Ioh. 2.7), was now in their hands (Procop. Aed. 6.4.6–10).7 However, their control of the area was short-lived. Determined to regain the African provinces, the Emperor Justinian sent his general Belisarius to defeat the remaining Vandals in the province, with the intention of embarking on a policy of peaceful alliances with individual Laguatan tribes (Procop. Vand. 3.25.7). But the policy failed when seventy-nine Laguatan chiefs were executed in 543 (Procop. Vand. 4.21.2–11), One oracle too many? 163 and a large-scale revolt broke out, co-opting other African tribes in Tripolitania under the command of Antalas. It was at this point that John Troglita was sent to regain the territory for the Byzantine Empire. Our main historiographical source for these events is Procopius (particularly De Bello Vandalico and De Aedificiis), who accompanied Belisarius on his expedition against the Vandals in 533 and remained in Africa, attached to Solomon’s staff, until the mutiny of the African army in 536.8 Although Procopius’ De Bello Vandalico is considered to be a reasonably reliable source,9 and the historian well informed about events in Africa even after his departure from the continent, his account of Troglita’s campaigns is not as detailed as for the preceding period.10 And there is the fact that Procopius was essentially an outsider, a native of Caesarea. While this does not necessarily make him less reliable, he was probably not aware of the detail that a native African author could provide. Both he and, particularly, Corippus focus on two major battles in Troglita’s campaign, in the winter of 547 and the summer of that same year, or possibly the following year. On the whole, scholars favour Procopius’ account when it comes to information on battles and military tactics, since it seems that Corippus tends to adjust the historical record to suit the panegyrical purposes of his poem, a poem written, moreover, to court imperial favour.11 But where most ancient writers depict Africa as distant and strange, despite the continent’s increasing familiarity over time, a trap into which even Procopius falls,12 Corippus is clearly on familiar ground. Thus Corippus is also used as a historical source by many scholars, since his epic contains a wealth of detail not found elsewhere – for example, the proper Libyan names, such as Erancun (Ioh. 8.605), or the intricate hierarchy of the different tribes within the Laguatan alliance (Ioh. 2. 7–148; 6.191–201).13 Later Arabic sources like Ibn al-Athir, Ibn ‘Abd al-Hakam or Ibn Khaldûn refer much more often to the Laguatan than the Roman sources and are useful on cultural detail, although they are at least 400 years later than Corippus.14 Making use of them means that, historiographically, the present study spans about 1,800 years of human history, from references in Herodotus to information in Ibn Khaldûn. This may not be as problematic as it seems, since the focus is narrow and many customs and traditions among the desert peoples seem to have remained unchanged even to the present day.15

Religion among the North Africans Our written sources – mostly Greek, Latin and some Arabic – are not exhaustive on the issue of local African religious practices, nor always entirely reliable,16 but some credible general information can be deduced from these accounts which will provide a background against which to assess the pagan practices described by Corippus, specifically oracular practices. Libyan beliefs retained a strong animistic character, centring on the spirits within stones, mountains, desert winds, water sources, rainbows and mirages. The importance of the spirits or ayns of the oases/wells is particularly plausible because of the life-giving power of the water which they provided in the interior 164 Martine de Marre desert areas, and they remained important into Late Antiquity.17 The local African deities became associated with those of Punic, Greek and later Roman religion. Among these, Ammon was very prominent, and over the centuries of coloniza- tion had been syncretized variously as Egyptian Amun, Zeus Ammon and Jupiter Ammon.18 Throughout these assimilations, however, the god retained an association with desert paths and oases (as the protector of travellers), and with the cult of the dead and divination.19 The god Ammon had his main sanctuary and his oracle at Ammonium (modern Siwah).20 There were also a number of other deities, known only through Corippus (such as Sinifere, a local war god, or Mastiman, probably also related to war), who retained their African identity.21 Christianity, which Procopius would have us believe the local pagans received with enthusiasm, does not seem to have been deeply rooted among the tribes in the interior.22 The oracle of Ammon had developed a certain standing among the famous oracles of antiquity from quite early on,23 but the best known consultation of the oracle is probably the visit of Alexander the Great in 331 bce. After his conquest of Egypt, Alexander travelled west to consult the oracle about the future and was famously greeted as ‘O son of Zeus’.24 A great many ancient sources refer to this incident and, although some are sceptical,25 we may reasonably assume that the connection between Ammon and Alexander had gained legendary fame in ancient times.26 Not a great deal of reliable information about the oracle of Ammon is known.27 Diodorus (17.50.6) provides some information on the oracular tradition in which the image of the god was carried by eighty priests on a golden boat, with female followers singing hymns.28 Oracular responses were given to the priest by interpretation of the movements of the bearers,29 although Diodorus also later speaks of a ‘voice’.30 The oracle shrine consisted of a vestibule, a main chamber, an oracle cell and a side chamber.31 The chief prophet in the encounter with Alexander is described as ‘an elderly man’ (τὴνἡλικίαν, 17.51.1). As Parke and others have pointed out, this follows the Egyptian oracular tradition quite closely.32 The idea of a male chief priest is followed by all the other Alexander sources, following the tradition set by Callisthenes.33 Other brief notices of the oracle of Ammon, such as in the Second Alcibiades dialogue (2.148 d–149b), the various tales of the Spartan Lysander’s consultation of the oracle or a visit by Plutarch, also indicate that the prophet or prophets were male and interpreted certain signs to speak on behalf of the god.34

Motifs and motives in the Iohannis That Corippus adopts and adapts from many of his predecessors in the epic genre does not require much further discussion. There are too many similarities for these not to be intentional, and it is evident that Corippus self-consciously represents himself as Justinian’s Vergil:35

The bard of Smyrna described brave Achilles, and Vergil, that learned poet, described Aeneas. The task John accomplished prompted me to describe his One oracle too many? 165 battles and to tell men yet to come about all of his deeds. John is superior to Aeneas in valour, whereas the poem I write is unworthy of Vergil [Aeneam superat melior uirtute Iohannes, sed non Vergilio carmina digna cano]. (Ioh. Praef. 11–16)

Corippus makes another passing reference to the epic tradition in the first book, and the poets who have dealt with war: ‘as the ancient poets [ueteres … uates] tell in pagan song’ (1.452). With the broad ueteres … uates in the plural, he brackets together all the poets who have dealt with epic battles, Martis … opus (1.451). His models seem to be mainly Statius, Lucan and Claudian, but since the epic tradition displays many instances of direct reference and intertextuality which consciously link the poets within that genre, there are echoes of the other epic poems as well.36 As Shea and others have already pointed out, Corippus was working in a genre which ‘relied heavily on mythological topoi and devices’.37 A parallel trend in literature was the developing Christian influence and Corippus makes free – if erratic – use of renovatio, combining Biblical imagery with that drawn from the classical epic tradition.38 Although not a hard-and-fast rule, Corippus tended to use classical mythological figures that are not theological, as a stylistic decoration that rooted him in the tradition of his genre, but he also used the Roman gods as literary ‘personifications of divine power’.39 Pagans and Christians meet with this treatment indiscriminately, but the poet, in the tradition of Aeneas versus Turnus, paints John Troglita and the African leaders as hero and anti-hero without a great deal of subtlety,40 as will be demonstrated by the detailed discussion on the oracle passages themselves below.

The oracular consultations in Corippus There are three instances of oracular consultation contained within the eight-book epic of Corippus. Scholars have generally brushed over these episodes. Shea, for example, groups all three occurrences together, assuming a specific location for all: ‘Foremost among the native divinities … is Ammon, whose shrine was at Siwa near Cyrene … Several pilgrimages are made to his shrine by the Africans in Corippus’ poem’.41 A more detailed examination, however, does not support such a broad assumption. In the first instance, Corippus recounts that Guenfan, the African leader and father of the better-known Antalas, goes to consult the oracle:

His father himself went to the counterfeit temple of Ammon [Ammonis templa simulata]. Asking for dreadful prophecies about his criminal son [prolis iniquae fata tremenda rogans], he sacrificed in an unholy manner horrible offerings to Jupiter. (Ioh. 3.82–5)

This brief notice represents all the information we are given about the first oracular supplication, but already a number of leitmotifs are revealed. Since the passage is 166 Martine de Marre preceded by a much longer description of a ‘golden age’ of Africa, the tempore prisco (3.67), the description of the actions of Guenfan and his son are made to appear the more terrible by contrast. That Corippus is continuing to compare the pagan Africans with the Christian Byzantines is clear: the sentence at 3.82–5 is heavily laden with strong negative descriptives which stigmatize Guenfan, his son and pagan religion in equal measure. Words like templa simulata, with its emphasis on the counterfeit, are followed up by tremenda fata and prolis iniquae. The reader is left in no doubt about the poet’s attitude to the Mauri. Guenfan is cast in sharp contrast to John Troglita, the former credulous and superstitious, and the latter clear-headed and under the guidance of the Christian god, characterized by words and phrases such as placidus (1.128) erigit auxiliumque dei pietate magistra (2.283) and talibus orantis fletus et uerba recepit suscipiens dominus (1.310–11). Jupiter Ammon is evoked in Corippus’ first passage, and we can at least reasonably assume that here the literary tradition reflects that this deity was the most prominent in North Africa for the entire period of pagan activity.42 Just as the Cyrenaicans had seen Ammon as a collective symbol for the Libyan aspect of their identity,43 so local Libyan tribes claimed him as their own44 and it is therefore reasonable that in Corippus he is the pagan deity most often mentioned.45 At this point, Corippus is not specific about the location where Guenfan consults the oracle of Ammon, since this is immaterial to his literary purpose. But since Ammonium was the famous seat of the oracle of Ammon, we must at least consider that Corippus may have had this venue in mind at Ioh. 3.82–5, and that Guenfan’s visit to the oracle is, if not historical, more ‘historical fiction’ than pure invention. The lines immediately following that first succinct description of a visit to the oracle of Jupiter Ammon speak of another visit. Guenfan goes straight from the oracle of Ammon to that of Apollo:

Then, making for the gloomy altars of Apollo, he sought the tripods and laurel of Phoebus [tristes et Apollinis aras inde petens Phoebi tripodas laurusque requirita]. Blood of the most gruesome kind imaginable washed upon those horrid altars [funditur horrendis sanguis maestissimus aris], as the priestess [uittata sacerdos], her head bound with fillets, slaughtered beasts of every kind and stirred the fates. (Ioh. 3.84–8)

The cheek-by-jowl positioning of the visit to the temple of Ammon with the typically Graeco-Roman ‘tripod and laurel’ of Apollo’s oracle is the poet’s attempt to draw a typically pagan oracle scene. Since consulting more than one oracle was quite common in ancient literature,46 Corippus is utilizing this tradition to emphasize the polytheia of the Mauri by a rapid, almost hasty, piling up of the evidence, with a sideways bow to a literary tradition of oracular lore in bringing in the phoebi tripodas laurusque. It is at this point that one inevitably starts to question Corippus’ credibility on oracles – is it not a case of ‘one oracle too many?’ The superlatives such as maestissimus only add to this effect of exaggeration. The poet’s literary and propagandistic aims are now quite overt, One oracle too many? 167 as he contrasts this erratic and unsatisfactory polytheism with the worship of the single god by the Byzantines, whose prayers are invariably answered. Thus the depiction of Guenfan and the oracle follows a fairly predictable pattern as Corippus tries to show the Mauri bringing themselves to ruin by their worship of false gods.47 Introducing a reference to Delphi also recalls one of Corippus’ models, Claudian, who intones at the birth of Honorius (8.141–8) that ‘even the pagan oracles of Ammon and at Delphi broke their silence’ (tibi corniger Ammon et dudum taciti rupere silentia Delphi) (implying that they had been silent before this). Corippus’ use of the same two oracles make these two oracular sites almost metonymic for pagan oracular activity in the Christian tradition. While the prophetess in this passage seems to be that of Apollo, most scholars read that Corippus is describing the oracle of Ammon.48 The reason for this assumption is probably the fact that the mention of the oracle of Apollo is a purely literary interjection, an almost parenthetical mention. Moreover, the famous Pythia had always been a woman, while, according to our earlier sources, the prophecies of Ammon were conveyed by male priests in heavily ritualized ceremonial, a style which would hardly have suited Corippus’ purpose in describing the Mauri.49 From a literary point of view, therefore, the adoption of the female prophetess allows Corippus to make full use of the dramatic description of the priestess’ possession by the god, which paradoxically at once recalls an ancient Greek tradition while at the same time enhancing the barbarism and ‘otherness’ of the Mauri in the reader’s eyes.50 The rest of the description of the priestess, the uittata sacerdos, is garish in the extreme as the poet aims to draw a picture of a bloody and violent sacrifice, where the ecstatic possession by a divinity and the priestess’ powers of divination and prophecy represent all the evils of paganism. The priestess thus conducts the pagan rites one after the other – first the killing of the sacrificial animal, which she examines. She then stabs herself repeatedly with the sacrificial knife (3.88–93), and falls bleeding into a frenzy which, Corippus tells us, rubor igneus inficit ora numinis icta nota (3.97–8) – tacitly acknowledging, therefore, first the existence of the pagan divinity, and second that the divinity had a power over the priestess in the imagination of the author.51 This lengthy and lurid description is an echo of similar passages in Lucan, where Appius Claudius consults the Delphic oracle (Delphica Phoebi, 5.70), of which thirty-five lines are specifically devoted to her possession and prophecies (Phars. 5.146–50; 165–77; 190–3; 208–24).52 Some of the words and phrases Lucan used occur again in the Iohannis – for example, vittata sacerdos (Phars. 1.597; Ioh. 3.87) and Phoebi (Phars. 5.156; Ioh. 3.85) tripodas (Phars. 5.162; Ioh. 3.85) laurus (Phars. 5.155; Ioh. 3.85) – but Corippus’ imitation lies more in emulating Lucan’s rhetorical style with its emotive language, typical of silver Latin. This creates a sharp contrast with a description of, for example, Vergil’s Sibyl – although she too is described as in a frenzied state, she never assumes the grotesque aspects attributed to Lucan’s prophetess,53 and it was this grotesque aspect that featured as a model for Corippus in this passage.54 168 Martine de Marre Corippus goes on to describe the pagan priestess, devoting several lines to her leaps, gyrations and gasps, until eventually the prophecy is given, very lucidly, in lines 107–40, predicting not only the fall of the Vandals, but also the ruin of the Africans. The prophecy thus forms a very traditional function within the epic, as a ‘device to generate suspense … through sporadic clues about the resolution of the action and the fate of the characters’.55 And, at lines 125 to 126, ‘most remarkably’, says Shea, ‘it also contains a prediction and endorsement of Christianity’s coming to all Africa’56 in the mouth of the native priestess. It is she who says: ‘for weary Africa will call upon her maker, the God whom she worships and whom it is right that all men recognize’ (Africa namque suum factorem fessa rogabit, quem colit ipsa deum, quem fas cognoscere dignos). Shea is no doubt correct in thinking that Corippus intended this to be ‘a true prophecy, referring to the Christian God’.57 That the prophecy about the Christian conversion of Africa comes from the pagan oracle is not, however, as remarkable as Shea suggests. First, throughout the epic it is quite clear that Corippus does not see the pagan deities as without power – it is rather that their power is inadequate to help the Mauri and inferior to that of the Christian God. Second, a prophecy given by an oracle that favoured the opposing side in a war, for example, was seen as even more trustworthy by readers of such literary accounts, if not by the oracle’s audience, and was quite a popular theme in ancient literature. From North Africa alone we already have two examples: the first is the prophecy given by Lucan’s Thessalian witch, Erictho (Phars. 6.611–14), in which she incongruously gives utterance to Stoic doctrine.58 The second case is the seventh-century African prophetess El Kahina, who prophesied that the Arabs would be victorious in Africa.59 At this point, the preliminary conclusion must be that this particular passage does not contribute to a discussion on oracular practice, or the location of such. The literary fame of the Delphic Pythia has simply been spliced onto the oracle of Ammon with little regard for recorded precedents of the manner in which the priests of Ammon provided prophecies. The dramatic account presents numerous similarities with Lucan’s rendition in terms of content and style, and the prophecy clearly serves a traditional literary function – to enhance the element of suspense, of ‘imaginative anticipation of the catastrophe’60 – while at the same time serving Corippus’ aim of painting the pagans in as negative a light as possible. This does not mean that there was no such oracular activity, but simply that, at least up to this point, Corippus’ epic cannot be used to make a case for it. In the final oracular episode (Ioh. 6.142–87), Carcasan, who has been elected the new leader of the second Syrtic insurrection, makes his first action the consultation of the oracle ‘in the lands of the Marmaridans, where horned Ammon dwells’ (Marmaridum fines, habitat quo corniger Ammon). This time, the location – Marmaridum fines – is more specific. The Marmaridans are referred to by other classical authors, and occupied the area between Egypt and Cyrenaica, adjacent to the Psilli, today stretching from western Egypt to Derna.61 Ammonium is therefore indicated as the location. But Ammonium is not our only option. Augila (present-day Awjilah) was also a site sacred to Ammon. Mattingly refers to it as one of ‘the great oracular centres of the Ammon cult’ One oracle too many? 169 outside Ammonium,62 and states that ‘Augila was most likely meant’ for the oracle consultation at Ioh. 3.82–5, although he does not elaborate.63 Augila lay within the Cyrenaican desert on the north–south trade route connecting Cyrenaica with the desert, and also on an east–west route linking Garama, Ammonium and Egypt. Herodotus (4.170–82) already mentions it as one of the oases to be found to the west of the oasis of the Ammonians, about ten days’ journey away.64 In Herodotus’ account, the nomadic Nasamones (and the name itself is clearly linked to the worship of Ammon)65 roamed between Augila and the coastlines of the greater Syrtis. The links between Augila and the worship of Ammon are thus already well-established at this early date, virtually a millennium before Corippus. Herodotus (4.173) also mentions that the Nasamones went to sleep on the graves of their ancestors after prayer, and received their answers in dreams. We are told something similar by Pomponius Mela, writing on North Africa in the first century ce, specifically with regard to Augila and prophecy:

The Augilae think only the spirits of their ancestors [manes] are gods, they swear by them and consult them as oracles [eos ut oracula consulunt]. They pray to the ancestors for what they want [precatique quae volunt], and, after they have thrown themselves on burial mounds [ubi tumulis incubuere], the spirit ancestors bring dreams as oracular responses [ pro responsis ferunt somnia]. (Pomponius Mela 1.46)

These early writings reflect a general feature of prophetic practice that seems to have been prominent, though not unique, in both pagan and Christian Africa: the cult of the dead or ancestor worship. Brett and Fentress call the cult of the dead in North Africa ‘one of the distinguishing features of the Berbers in antiquity’,66 combining fertility worship with belief in prophetic powers. Archaeological findings of benches and rooms for sleeping in North African tombs and shrines indicate that the practice of incubation was common on the continent throughout antiquity.67 From the third century ce, evidence for the cult of the dead or ancestor worship appears among the Christian evidence, since local bishops simply adapted the pre-existing custom to the worship of Christian martyrs, as it proved impossible to root out.68 Sporadic later evidence also seems to support the importance of the oasis. Augila was of sufficient size in the sixth century to merit being called a πόλις by Stephanus Byzantinus (Ethnika 4.2), for example. Archaeological evidence, beyond the remains of the temple of Jupiter Ammon, which were still visible in the nineteenth century, is scarce, although there are indications that Augila seems to have had three settlements.69 Finally, there is a passage in Procopius’ De Aedificiis which also mentions Augila as it became the target of Justinian’s policy to enforce orthodoxy across the empire. Philae in Upper Egypt had suffered a similar fate,70 as Justinian’s forces marched on the city and deconsecrated the temple of Isis, expelling its clergy, destroying the images of the gods and turning the building into a church 170 Martine de Marre (Procop. Bell. 1.19.36–7).71 Augila’s fate was slightly different, as Procopius begins to relate in the following passage:

And there are two cities which are known by the same name [Πόλεις δέ πού εἰσι δύο ἐπ’ ὀνόματος ἑνὸς ᾠκημέναι], each of them being called Augila [Αὐγίλα γὰρ ἑκατέρα ἐκλήθη]. These are distant from Boreium about four days’ journey72 for an unencumbered traveller, and to the south of it. (Aed. 6.2.14–15)

It is possible, as some scholars have suggested, that Procopius was conflating Ammonium and Augila.73 We have no evidence that the historian actually travelled this far into the interior, since both oases lay in the desert.74 The duality could also be explained by the fact that there seems to have been more than one settlement, collectively referred to as Augila.75 Lastly, it could be that Procopius is referring to the Cities of Ammon (Ἄμμωνος πόλεις) mentioned by Diodorus (17.49.6), and was using Diodorus as a source. However, since the precise location of these ‘cities of Ammon’ remains obscure, as does the puzzling reference to the distance from Boreium, this must remain speculation. Procopius continues:

And they are both ancient cities [ἀρχαῖαι δὲ οὖσαι καὶ] whose inhabitants [τῶν οἰκητόρων] have preserved the practices of antiquity [ἀρχαιότροπα], for they were all suffering from the disease of polytheism [πολυθεΐας ἐνόσουν] even up to my day [καὶ εἰς ἐμὲ]. There, from ancient times [ἐκ παλαιοῦ] there have been shrines [ἕδη] dedicated [ἀνέκειτο] to Ammon and to Alexander the Macedonian [Ἄμμωνι καὶ Ἀλεξάνδρῳ τῷ Μακεδόνι]. The natives actually used to make sacrifices to them even up to the reign of Justinian [οἷς δὴ καὶ ἐσφαγιάζοντο μέχρι ἐς τὴν Ἰουστινιανοῦ βασιλείαν οἱ ἐπιχώριοι]. In this place, there was a great throng of those called temple slaves [ἱεροδούλων]. (Aed. 6.2.15–16)

The words καὶ εἰς ἐμὲ seems to imply that pagan worship had been actively practised until recently, although Procopius does later add μέχρι ἐς τὴν Ἰουστινιανοῦ βασιλείαν. Although Cameron states that ‘its pagan glories had already faded into the past’,76 had the pagan practices been a thing of the past, there would have been no need for the type of mass conversion77 which Procopius describes next:

But now the emperor has made provision, not alone for the safety of the persons of his subjects, but he has also made it his concern to save their souls, thus he has cared in every way for the people living there. Indeed, he by no means neglected to take thought for their material interests in an exceptional way, and also he has taught them the doctrine of the true faith [εὐσεβείας ἐδίδαξε δόξαν], making the whole population Christians and bringing about a transformation of their polluted ancestral customs [μεταπορευόμενος λελυ μασμένα σφίσι τὰ πάτρια ἤθη].78 Moreover, he built for them a church of the One oracle too many? 171 Mother of God to be a guardian of the safety of the cities and of the true faith. So much, then, for this. (Aed. 6.2.18–21)

Procopius makes no mention of oracular practice here, referring more generally to ‘polluted ancestral customs’, in addition to the information in the previous excerpt – that the native inhabitants made sacrifices to Alexander and Ammon (οἷς δὴ καὶ ἐσφαγιάζοντο … οἱ ἐπιχώριοι). This curious combination at once brings to mind the association of Alexander as the reputed ‘son of Zeus Ammon’. Scholars have made widely diverging interpretations based on this line. Constantelos assumes that the temple was dedicated to Zeus Ammon and ‘built by Alexander the Great’,79 whereas Cameron’s interpretation of this passage is ‘that [Augila] was the site of the shrine of Ammon visited by Alexander the Great’.80 If the statement were to be taken literally, that both Ammon and his son, Alexander, had shrines at Augila, and that sacrifices were made to them there, this connection would imply that Procopius or his sources had in fact confused the reputation of Ammonium, which Alexander had visited, with Augila. The Procopius passage on Augila therefore supports pagan ancestral practices in the sixth century, specifically related to Ammon (and Alexander) at this location. If Procopius confused Augila and Ammonium, however, the actual physical location must remain in doubt, and the πολυθεΐας ἐνόσουν in relation to Ammon does not actually specify a practising oracle at the site. Returning to Corippus, in the final oracle passage, Carcasan, like his predecessor, also asks Jupiter for a reply (responsa).81 This last oracular episode clearly describes a female priestess of Ammon as the one conveying the god’s response. A bull is also sacrificed in this scene, but this final consultation is much less bloody and more restrained in terms of the priestess’ behaviour. This oracle is therefore a closer parallel to Vergil’s Sibyl (Aen. 6.42–97), rather than Lucan’s oracle, who was used as the previous model. There are parallel references with the Aeneid to her changing colour, her heart swelling/heaving bosom, her hair no longer braided – but not the ripping apart of animal entrails and self-mutilation of the earlier scene. The prophecy on this occasion is ambiguous, however: the Africans will hold the fields of Africa forever, there will be peace and Carcasan himself shall enter Carthage. Corippus then conveys the true meaning of the message (6.177–87): that the Africans will be buried in their land forever, Rome will impose the peace and Carcasan will be led into Carthage in a Byzantine triumph.82 Ambiguous prophecies were seen by Christians as typical of the deceptiveness and duplicity of the pagan deities and had a long tradition in ancient literature going back to Herodotus’ tale of the prophecy given to Croesus.83 Again here it would seem that any historical aspect of this passage is overshad- owed by Corippus’ literary agenda. The description of the priestess is similar to Vergil’s Sibyl, and her misleading prophecy is also a topos, used here to show the deceitfulness of the pagan tradition. The only aspect of the passage that has any bearing on historical reality is the mention of the general location, which is known from other sources, and could indicate either Ammonium or Augila. 172 Martine de Marre While the description of a woman as priestess of Ammon does not follow the earlier sources in describing the oracular tradition of Ammon, it is true that what we know of prophetic practice in Africa from Procopius favours female agency. Procopius recounts the following incident which historically precedes the arrival of John Troglita in Africa, and the events related by Corippus:

When it came to be expected that the emperor’s expedition would arrive in Libya, the Moors, fearing lest they should receive some harm from it, consulted the oracles of their women [ταῖς ἐκ τῶν γυναικῶν μαντείαις ἐχρῶντο]. For it is not lawful [οὐ θέμις] in this nation for a man to utter oracles [μαντε ύεσθαι], but the women among them become possessed [κάτοχοι] as a result of certain sacred rites [ἐκ δή τινος ἱερουργίας] and foretell the future [προλέγ ουσι] no less than any of the ancient oracles [τῶν πάλαι χρηστηρίων οὐδενὸς ἧσσον]. So, on that occasion, when they made enquiry, as has been said, the women gave the response: ‘There shall be a host from the waters, the overthrow of the Vandals, destruction and defeat of the Moors, when the general of the Romans shall come unbearded … And then the remainder of the Moors recalled the saying of their women, to the effect that their nation would be destroyed by a beardless man.84 (Vand. 4.8.12–14 … 4.8.28)

Cameron maintains that Procopius has embellished his historical narrative ‘with a story of Berber prophecy which we would be right to regard as a picturesque detail, since it is hardly likely that Lepcis was entirely deserted’.85 Other scholars, however, have also described a general decay and even abandonment of the African cities in this time, so this description may not necessarily be fictional, nor the female oracles a ‘picturesque detail’. There is also considerable evidence for the role of female prophecy in Africa from later periods, as Whittaker and Smith have taken pains to illustrate.86 They see this against the background of the matrilineal and matrilocal African heritage,87 where Graeco-Roman and Arabic writers, from their patriarchal perspective, thought the freedoms given to women in Africa odd and variously interpreted these as licentiousness or promiscuity, as mentioned earlier.88 The plural μαντείαις used by Procopius is also interesting. This could of course refer to multiple oracles given on different occasions, but it could also link up to the African practice of inviting oracles through ancestral spirits, which is already attested at Augila by Herodotus and Mela. The γυναικῶν μαντείαις seem to be linked to a context less official or formal than that of the oracle of Ammon, and it is of course plausible that, with pagan temples becoming derelict due to imperial legislation, polytheis- tic practices may have moved to the domestic sphere.89 Female prophecy is mentioned in one other case, from around the middle of the seventh century, slightly later than the time in which Corippus and Procopius wrote. Early Arabic sources identify her as ‘Kahina’, which is thought to mean ‘seer’ or ‘prophetess’.90 This queen, locally known as Dehia, gained legendary status in North Africa and among the Tuareg.91 Although she was allegedly of One oracle too many? 173 Berber origin, she is said to have identified with the Christian Byzantines and from the Aures Mountains gave resistance to the Arab invaders.92 Ibn Khaldûn (7.11) mentions that she possessed the gift of prophecy and could correctly foretell the future.93 From all the legendary material that has come down to us, the only secure information would seem to be that she was a Berber queen and a prophetess, points on which all the ancient sources agree.94 Material evidence and a fair amount of literary testimony therefore seems to suggest that oracles via ancestor worship, whether Christian or pagan, were a feature of North African religious practice in antiquity. The evidence of Herodotus and Mela on ancestor worship and the connection with oracular prophecy is not reflected by Corippus’ oracle passages at all, whereas both passages from Procopius could conceivably be seen to represent continuity with the practices mentioned by the earlier authors. Procopius, whether he had Ammonium or Augila in mind, at least seems to have been under the impression that such pagan religious practices still prevailed there in Late Antiquity.

The contribution of Corippus and Procopius to the debate around the existence of late oracles All pagan sanctuaries and oracular activity had been famously closed down by the edict issued by Theodosius in 391,95 and some scholars support the idea that pagan oracular activity had been rooted out by the sixth century.96 Others maintain that it was still practised and that paganism continued in many pockets within the empire.97 All are agreed, however, that it was Justinian’s policy to create a unified empire by enforcing orthodoxy. This may have had its own repercussions, as we see in the following passage from Procopius, which would seem to indicate that there was some natural resistance to the enforcement of orthodox Christianity:

And all those of their number who were persons of any prudence and rea- sonableness showed no reluctance about adhering loyally to this faith [Christianity], but the majority, feeling resentment that, not by their own free choice, but under compulsion of the law, they had changed from the beliefs of their fathers [δόγμα τὸ πάτριον μετεβάλοντο], instantly inclined to the Manichaeans and to the Polytheists [Πολυθέους], as they are called. (Anek. 11.26)

Despite legislation, it seems that paganism was still not eradicated by the end of the century, and there is some evidence that much of pagan worship went underground.98 The interior of Cyrenaica and Tripolitania may provide an example of this. It would be quite plausible that rebellious indigenous tribes, removed from the Romanized littoral and in westward migration along the oases, may have been the source of a revival of Ammon worship in a re-establishment of their own African identity, as Mattingly surmises.99 Thus it is entirely possible that, when the more famous oracular site at Ammonium was closed down, Ammon worship, if not the oracle, moved to less public locations, to places like Augila,100 174 Martine de Marre where imperial control had been abandoned – until, as Procopius has informed us, Justinian targeted the area as part of a ‘cleaning up’ policy. Other centres, such as Ghirza to the south of Lepcis Magna, indicate that some temples were still actively used for pagan worship into the seventh century.101 Unfortunately, Augila has not been excavated to the same extent. We may also draw further support from modern Africa, as those areas in North Africa which have retained Berber traditions and their Afro-Asiatic languages are usually those which have had the least contact with their colonizers over the centuries.102 The oases at Augila and Ammonium still house Berber populations today. That the centre of Ammonian worship moved away from well-known Ammonium to Augila may also explain Procopius’ confusion regarding the two cities named Augila and the reference to Alexander and Ammon, since the description of these two oasis centres must have become rather similar in Late Antiquity. Pagan traditions and beliefs were not, it seems, dead, even if centres practising these traditions were being systematically rooted out. Procopius provides some credible support for the functioning of female oracles, and this is supported by Belayche’s study which found that the social function for the reassurance of oracular pronouncements did not necessarily wane with Christianity,103 even though they may have been frowned upon by rulers and administrators desiring unity and orthodoxy throughout the empire. In Late Antiquity, voices of scepticism had more to do with the ethics of oracles, rather than their existence or even their power to predict the future.104 For example, Socrates of Constantinople (4.8; 4.19; 5.17), writing about a century earlier than Corippus and Procopius, thought that pagan oracles could be accurate, even though they came from the devil.105 Another account about the Alexandrian philosopher Asclepiodotus, dating to 484 ce, relates how he received ‘an oracle (or rather, he was deceived by the demon appearing as Isis) …’ (Zach. Rhetor, V. Severus 3). When recording the famous last oracular pronunciation from Delphi in the twelfth century, the Byzantine historian George Cedrenus wrote that the Emperor Julian had sent Oribasius to rebuild Apollo’s temple at Delphi: ‘Arriving there and taking the task in hand, he received an oracle from the demon’ (ἔργου ἀψάμενος λαμθάνει χρησμὸν παρὰ τοῦ δαίμονος, Cedrenus 532). Many such accounts exist. Corippus’ literary rendition of oracular prophecies at least corresponds with the belief held by Christians at the time that oracles had a genuine power to predict the future, a power that did not come from the Christian God. Thus it was not so much that pagan oracles could not predict the future but that they were seen as demons – they had powers, but these were ungodly.106 Thus Corippus employed the suspicion surrounding the divine inspiration of persons and individual prophecy prevalent in his own time, in which oracles had become a part of the stories about contests between Christians and their demonic adversaries.107 Corippus makes no mention of the African tradition of oracles and prophecy from ancestral spirits, which he could plausibly have utilized in his epic. The reason for this may lie in the transfer of this tradition of ancient worship to the Christian context mentioned above, and there is a slight indication of this in the Iohannis. Instead of ancestors appearing to foretell the future to the pagan Mauri, One oracle too many? 175 we see Corippus’ hero, John, receiving two ‘visitations’. The first is in the form of a vision in which John is threatened by a tristis imago, a spirit described as cognata tenebris and Maura videbatur facies nigroque colore (Ioh. 1.243–5) and therefore clearly linked to the Africans and inimical to the Byzantines. It forbids John and his fleet to cross to the African shore, but the spirit is chased away by John and is replaced by another shade, this time aspectu placidus senior descen- dit Olympo (1.259). John addresses it as his father, pater optime (1.265), and Corippus may be modelling the interaction on Aeneas’ visitation by the shade of Anchises (Aen. 5.724–37). His father encourages him to go on, and dismisses the threats of the African angelus deiectus. These are described as visions, but John remains in control and is not possessed, like the African priestesses.

Conclusion It is quite possible that Corippus may have exploited what he knew of local pagan prophetic practice in the scenes he describes, and that Augila is the likelier venue in view of the evidence of activity there painted by Procopius. However, the echoes of Lucan and the clear literary and tendentious purpose of the passages – the superlatives and the superimposition of one tradition onto another – give the episodes a sense of inauthenticity, and make this case impossible to prove. Since the oracles of Ammon in the preceding tradition were reputedly given by male priests in the Egyptian tradition, this diminishes even further any likelihood that Corippus’ prophetesses of Ammon are a reflection of Ammonian oracular practices. The overriding literary techniques and reliance on these literary antecedents within these passages also make it unlikely that Corippus made use of local practices of female prophecy which are prevalent on the continent to the present day. Corippus’ priestesses must remain a literary creation set in a particular literary tradition of Late Antiquity. In this respect, then, we can agree with Cameron’s overall view that Corippus’ epic is almost entirely fictional and tendentious at best.108 Although Mattingly, among others, takes the epic literally – ‘the oracles [in Corippus] were initially used to excite tribal support and unite the confederated forces; second, the propaganda value of such prophecies was used to maintain confidence on the eve of battle’ – this presents an over-reliance on a source whose representation of oracles, in particular, cannot be seen as anything other than fictional.109 If ever there were a case to be made for the working together of archaeologists, philologists and historians to reconstruct a view into the past, this would surely be it. The evidence presented by Procopius, on the other hand, although slight, together with later – even modern – evidence, can point to a tradition of female prophecy that is much more believable, in the context of a continuation of pagan practices. While scholars such as Cameron may have pointed to the literary overtones in Procopius, and his text does need careful interpretation, there seems insufficient reason to doubt his evidence in the case of female prophecy. At the same time, we cannot specifically link such prophecy to Ammon, since it is equally 176 Martine de Marre possible, and even likely, that this use of oracles could be related to the prophecies given via ancestral spirit worship. In Procopius’ account we can find some support for the practice of paganism in Late Antiquity, and more support for the broader socio-historical views of Mattingly against Cameron, who is also sceptical on the possibility of the continuation of pagan practices in this period. It would seem entirely plausible, following Procopius’ observations, that the likeliest scenario is that pagan practices did not die out, but moved into the less well-known interior under the advance of an increasingly dogmatic Christianity. It is also conceivable, as Frankfurter and Mattingly have suggested, that the worship of Ammon in the Syrtic area and around the oases functioned as a unifying element among the Libyan tribes in their struggle against the empire – but this argument must be made without the corroborating evidence of Corippus.110 In conclusion, therefore, the evidence we have seems to suggest that there was female prophecy in North Africa, as Procopius suggests, but the likelihood that there were female oracles speaking for Ammon – probably not.

Notes 1 I would like to thank the editor of the volume and the two anonymous reviewers for their helpful comments and suggestions. 2 Biographical information on Corippus is sparse. He is called a grammaticus Afer in the Matritensis 10029 (ninth to eleventh century). On this see Cameron (1980) 534–5; also the PLRE 395–527: 329. 3 See inter alia Burck (1979); Ehlers (1980); Cameron (1984); Moreschini (2007); Riedlberger (2010). 4 For example Cameron (1982) 38–9; (1985) 183; Merrills and Miles (2014) 129, 228, 253–4; Mattingly (1994) 38–9. Shea also presents a lucid assessment of which aspects may be more trustworthy than others (1998) 20. 5 Such as Frankfurter (1998) 157; Gärtner (2008) 24–5, 50–1. Mattingly (1994: 39) takes the oracle episodes as factual, as does Moreschini (2002) 348–9. 6 On the name Laguatan: ‘It is generally accepted that the transliteration found in the work of the African writer Corippus is likely to be the closest to the original Berber (lagatan/Laguatan). Alternative forms in Procopius (Leuathae) and early Arab writers (Louāta or Lawāta) hint at the soft pronunciation of the g’. See Mattingly (2008) 4315. There is some argument about the eastern origins of the Laguatan (assumed by Mattingly to be in Tripolitania, while Modéran sees them as a new incarnation of the Nasamones of Augila). Fentress (1981: 208) sees their origins as southern Saharan, rather than eastern, while Smith (2003: 488) thinks they may be the Austuriani referred to in earlier Roman sources. But they were a confederation, eventually there would have been elements from various regions. 7 Synesius of Cyrene’s last letters, dating to 412–413, already indicate that the inland areas of Cyrenaica were no longer under Roman control (Ep. 81.10, 62). Much later, in the fourteenth century, the Arab historiographer Ibn Khaldûn notes the widespread presence of the ‘Louāta’ around the oases of western Egypt, Cyrenaica, Tripolitania, Syrtica, as well as the areas of Sfax, Kairouan, the Algerian Aures, the oases of Zab and at Bougie (Ibn Khaldûn I: 176–82, 232–6, 280–1). By this time, the Laguatan had ceased to resist and had joined the Arabs in spreading Islam. See Smith (2003) 488. One oracle too many? 177 8 Procopius and Corippus are our main sources, but references to the events in Africa for this period can also be found in a number of other sources, such as Jordanes, Victor Tonnennensis or Paulus Diaconus. 9 Cameron (1985) 134–51. Procopius is vindicated when tested against other material evidence. See Lillington-Martin (2013) 599, 627. For a general overview of recent scholarly opinions on the historian see Greatrex (2014). As with any source, Procopius’ accounts do need careful scrutiny individually, and some works fare better than others. For example, the underlying theme of Procopius’ De Aedificiis is propaganda for Justinian, depicting the emperor as the saver of lives and souls through the building of fortifications and churches. In this general aim, he is not far removed from Corippus, whose work is pro-Christian and overtly eulogistic. See Schindler (2009) 227–309. In terms of his accounts of the wars, Procopius is seen to be more reliable, as supported by Modéran (1986). 10 Evans (1972: 67) supports Procopius’ detail for the period, whereas Treadgold (2007: 199) asserts that Troglita’s campaigns were ‘appended as an afterthought’, supported by Cameron (1985) 179. 11 For comparison of various military and political accounts provided by Procopius and Corippus see Cameron (1980) 534–9; Zarini (2003); Shea (1998) 6–20. 12 Procopius, as Cameron (1985: 180, 184) and Smith (2003: 465–6) have demonstrated, does not pay a great deal of attention to local detail and his focus is on Byzantine action. 13 Scholars generally credit Corippus with the greater accuracy in most instances of this kind, for example Camps (2002); Mattingly (1983) 100; Mattingly (2008); Modéran (1986; 2003); Shea (1998) 20–8. Modéran (2003: 1.1 e-text): ‘Corippe connaît bien les Maures, et il le montre abondamment. Il ne s’agit pas pour lui d’une réalité étrangère et abstraite, mais de voisins qu’il a fréquemment côtoyés’. On Berber names, see Riedlberger’s use of Corippus (2010: 351–3). Kaldellis (2004: 271 nn. 138 and 141) estimates that Corippus’ epic is probably earlier than the De Belli of Procopius. But, according to Cameron (1985: 123), it is likely that the North African church to which Procopius refers in his later work (Aed. 6.2.21) was already standing when Corippus wrote his epic, whose composition she estimates to be about 549. 14 Smith (2003: 461–2) notes that little transfer of historical and geographical information seems to have taken place from the Graeco-Roman sources to the Arabic writers. 15 Bates (1970: 37–8) describes the life of the desert people as very conservative and not easily changed, but see also later discussion and references on prophecy and spirit possession in this chapter. 16 Smith (2003: 459–65) comments on how classical writers often used the accounts of their predecessors, rather than personal autopsy. 17 Joleaud (1933) 245. The beliefs around these water sources in Africa were written about by classical writers from Herodotus to Lactantius, for example Hdt. 4.181; Diod. 10.17.50.3; Pliny NH 2.106; Lucr. 6.848; Mela 1.39. 18 It is usually held that Ammon was the Hellenized form of the name of the Egyptian Amun, a traditional god of the oases, and that worship of the god spread to ancient Libya and Nubia. See Parke (1967) 194. Bénabou (1975: 335) puts this as early as between the sixteenth and twelfth century bce. Bates (1914: 190–1), followed by Mattingly (1994: 38), however, makes a strong argument for the differentiation of origin between Egyptian Amun and Libyan Ammon, most particularly because of the latter’s strong association with prophecy (while the former was essentially a fertility god). It is thought that Ammon had come to be Hellenized during the sixth century bce – coinage of Cyrene depicting Zeus Ammon appeared in the late sixth century bce and the deity was henceforth represented as a collective symbol of Libyan-Greek identity. See Marshall (1998) 56. The temple to Zeus in Cyrene was erected even earlier, under the reign of Battos IV. The earliest literary reference to Zeus Ammon is Pindar’s Hymn to Ammon, referred to by Pausanias (9.16.1), who claimed to have seen it. References are 178 Martine de Marre also made in Hdt. 2.29–31; Diod. 3.6. On colonizing influences and how the Libyans adapted these to their own models, see Mattingly (1994) 38–9. Marshall (1998: 56) emphasizes that the Cyrenaicans (specifically those descended mainly from the Greek settlers) distinguished their worship of Ammon from that of the Libyans by conflating him with Zeus. The Egyptian and Libyan gods in any case seem to have become intertwined in the Graeco-Roman understanding, for example, in the representation of Ammon with the horns of the ram, similar to the Theban Amun (the ram was sacred to both Egyptians and Libyans). See Joleaud (1933); Tsetskhladze (2008) 214. Ammon was also adopted by the Carthaginians and known locally as Baal Hammon, a deity which, influenced by Roman syncretism, was known as Saturn, but from Tripolitania to Egypt the worship of Ammon dominated. See Bates (1914) 198–200. See Bénabou (1975: 334–40) on the amalgamation of Zeus Ammon and later Jupiter Ammon, with cross-pollination of Saturn and the Punic Baal Hammon. 19 Joleaud (1933) 253; Mattingly (1994) 38–9, 167–8. 20 For details of the oasis city discovered through archaeological projects, see Bagnall et al. (2016) 132. 21 Camps (1990) 131–53. Although, as Cadotte (2007: 6, 385) has shown, syncretism was often a surface assimilation and further examination reveals gods who for the most part ‘preserved their original nature and remained African in spirit’. 22 Brandl (1975) 467–95; Brett (1979) 554–5; Smith (2003) 460–1; Oduyoye (2009) 63. 23 Herodotus (1.46) refers to several consultations of the oracle from as early as Croesus, demonstrating that the fame of the oracle of Ammon was well known by the mid-sixth century bce. See also Aristoph. Ornith. 716. Herodotus (2.54–7) also links the origin of the oracle of Ammon with the oracle at Dodona. See Ghazal (1986: 171–7) for a detailed list of sources on the Greek consultations of the oracle of Zeus Ammon up to Alexander. 24 Curt. 4.7.25; Plut. Alex. 27.5; Diod. 17.51.1. On the authority of Callisthenes, Strabo (17.43.1) states that this was confirmed by oracles of Apollo at Branchidae and of Athena at Erythryae. 25 Scepticism about the oracle’s authenticity is expressed by Plut. De Def, Orac. 2, 5; Luc. Phars. 7.192. 26 Egyptian hieroglyphs to this effect already preceded the visit to Ammonium, recording Alexander as ‘beloved of Ammon and selected of Ra’ and ‘son of Ra’. See Parke (1967) 223. 27 Parke (1967: 134) is sceptical about alleged early consultations of the oracle, seeing the lists of oracles as a rhetorical technique. 28 An almost identical account is provided by Curt. 4.7.23. See also Parke (1967) 200. 29 ‘Carried by eighty priests, who go without their own volition wherever the god directs their path’. 30 ‘[A]s the bearers now lifted the god and were moved according to certain prescribed sounds of the voice’. 31 Kuhlmann (1988) 127. 32 Parke (1967) 199–200; Nodzynska (2006) 121; Gillam (2015) 62–3. It is also possible that Diodorus’ standard processional oracle suggests that the Egyptian oracle cult may have been superimposed on a local religious practice. 33 Arr. An. 3.3–4; Diod. 17.49–51; Plut. Alex. 26.6–27.5; Curt. 4.7.5–31; Justin 11.11.1–13; Strabo 17.1.43. Discussion in Parke (1967) 225–6. 34 We are told by some of the ancient writers that Lysander, in trying to get the sanction of the oracle of Zeus Ammon, attempted to bribe the priests there: Plut. Lys. 25; Nepos, Lys. 3.2–4; Diod. 14.13.5–8. Curt. 4.7; Paus. 3.18.3; Strabo 17.1–43. Plutarch’s visit: De Defec. Orac. 3.410a. Herodotus (2.54–7) is an exception to this, since his anecdote on the origin of the oracle would imply that, at least originally, the oracle was a woman. Parke (1967: 94–241) considers most of the references to the oracle of Ammon to be of spurious historical value and mostly literary and derivative. One oracle too many? 179 35 Cameron (1984) 167, 175; Shea (1998) 26; Schindler (2009) 237–309; Gärtner (2008) 30–8. 36 Ehlers (1980) 109–35. 37 Shea (1973: 118), for example, points to the harping on the concepts of fides and imperium in the Iohannis, and the distinction between subiecti and superbi, adopted from the Aeneid. The literary quality of the Iohannis has been debated. It is rated as not being a pedestrian regurgitation of epic techniques – Alan Cameron (2001: 24), with qualified praise, calls the epic ‘a very creditable production for a sixth-century small-town African schoolmaster’. Likewise, Bruck (1979: 397) commends Corippus’ style for its ‘Flüssigkeit, Klarheit und leichte Lesbarkeit’. Cameron (1984: 68) and Shea (1998: 43–4) agree. Zarini (2000: 53) is more critical: ‘D’un seul mot chez Homère (mênin, andra), on passe à des compléments de plus en plus nombreux, chez Lucain ou Corippe par exemple’. 38 Detailed discussion in Gärtner (2008) 9–25, 41–51; Von Albrecht (1999) 338. 39 Shea (1973) 124. Also Lepelley (2010) 477, 486; Hofmann (1989) 361–7. 40 That Corippus polarizes the pagan Mauri and the Christians – black and white, evil and good, confusion and order, irrational and rational – has been amply demonstrated: Shea (1973) 122–3; Shea (1998) 27; Cameron (1984) 173. To some extent, this corresponds to Procopius’ theme in De Aedificiis, where the Byzantines are seen as the bringers of enlightenment. See Cameron (1985) 113–33. 41 Shea (1998) 38. 42 Ammon also has literary precedents in Roman epic, which had already taken up the association between Ammon and the Africans (specifically the Garamantes: Verg. Aen. 4.198; Sil. Pun. 2.58; Luc. Phars. 9.511–12). 43 Marshall (1998) 56. 44 Bates (1914) 197–8. Ancient writers were not always clear about links to Ammon. For example, Pliny (NH 5.5.3) incorrectly relates that the Greeks called the people of Ammonium the Mesammones because they were found in the middle of the desert, a false etymology because they were given this name on account of their worship of Ammon whose sanctuary was to be found at the oasis at Ammonium. 45 Ammon is mentioned twelve times: 2.110; 3.81; 6.116, 147, 179, 190, 556; 7.515, 519, 534; 8.252, 304; Gurzil nine times, Mastiman three times and Sinifere twice. 46 Such as Croesus (Hdt. 1.46), for example, or Xenophon’s advice to send to Dodona and Delphi for oracular pronouncements on what was best for the city of Athens (Vect. 6.2). 47 Lepelley (2010) 487; Gärtner (2008) 18. 48 Cameron (1984) 174; Shea (1998) 38–9. 49 Nodzynska (2006) 113–21. 50 On the comments of classical and Arabic writers regarding the unusual freedoms given by the African tribes to their women, see Smith (2003) 482, 496–7. 51 Shea (1998) 42. 52 Whittaker (1965: 22–3) outlines the arguments against the historicity of the more lurid religious practices described by a number of other classical authors like Lucan. 53 Dick (1965) 464–5; Whittaker (1965) 22. 54 In Lucan’s narrative (Phars. 9.544–86), Cato (his Stoic hero from Book 8 on), also passes the temple of Jupiter Ammon while he and his troops are in the Libyan desert, but he refuses to consult the oracle, despite Labienus’ encouragement that he try to find out how long the war will last. See Cameron (1984) 168, 173; Ehlers (1980) 119–20. 55 Dick (1963) 37. 56 Shea (1973) 127; Shea (1998) 40. 57 Shea (1973) 128. 58 Brett (1979: 509–12) suggests that she may have been a man, although this seems unlikely in view of Lucan’s full description. 59 For references, see Hendrickx (2013) 56 n. 53. 180 Martine de Marre 60 Moore (1921) 101, 108–9. 61 For example Strabo 2.5; Lucan Phars. 9.893. 62 Mattingly (1994) 168. Mattingly (1994: 33) suggests that the cult of Ammon had spread from Egypt in the east along the chain of oases to the west. Brett and Fentress (1996: 77) surmise that Augila may even have been the original base of the Laguatan, and that they made the religious centre of Ghirza their capital. 63 Mattingly (1994) 33, 39. 64 The existence of such a caravan route along the oases, described by Herodotus, is deemed credible by a number of scholars. For discussion and references see Corcella (2007) 704–5. 65 On their location see also Lucan Phars. 9.438–44; Silius Pun. 1.408–10; 3.320 and discussion in Corcella (2007) 698. 66 Brett and Fentress (1996) 35. See also Tishken, who writes: ‘Despite regional variations, all African indigenous religions manifested certain characteristics such as polytheism, ancestral reverence, witchcraft, and sacrifice’ (2007: 1471). 67 Brett and Fentress (1996) 35. There are interesting parallels in neighbouring Egypt, for example at the temple of Mandulis at Talmis, which maintained an incubation oracle until the fifth century. On this and the Isis cult, see Frankfurter (1998) 108–9, 160–2. 68 Aug. Ep. 29.9. Kotila (1992: 220) discusses attempts by local bishops to adapt the custom so that the practice would be limited to Christian martyrs. Camps (1986: 163) is of the opinion that the decorated tablets of El Mreïti and the painted or engraved stelae of Djorf Torba (western Algeria), which likely date from the sixth century, may well be evidence of the expression of these wishes made before propitiatory dreams, and indicate Christian influence. Bates (1914: 178) cites some other cases, indicating that this was probably a widespread practice among all the Berber tribes. 69 Mattingly (1994: 7, 33) refers to the Augilae as the ‘inhabitants of the Augila group of oases’. 70 Harran to the south of Egypt, where the Blemyes and the Nobadae had had a vested interest in the temples and had made a treaty with Diocletian, was left unmolested. Leone (2013: 121–87) discusses issues around the moving and reuse of statues. Procopius (Aed. 6.2.21–3) also mentions Boreium, a city to the north of Augila and to the west of the Pentapolis, where Justinian is said to have brought about the conversion of the local Jewish population and built a church in place of the Jewish sanctuary. In another example, where a pagan centre is replaced by a Christian church, the cult centre of Isis of Mebnouthis (which granted healing via incubation and dreams) was destroyed and replaced by a Christian church with relics which apparently had ‘healing powers’ (484 bce) – essentially a continuation of the tradition of healing in another form. Frankfurter (1998: 165) points to the ‘vital importance of the Isis oracle in serving regional needs’, presenting an interesting parallel with the present evidence on Augila. 71 The sanctuary was still standing, dedicated to the martyr Stephen, with an inscription at the entrance that is believed to have had an apotropaic function against the demons rumoured to inhabit pagan sanctuaries. See Lee (2016) 146–7. 72 Bahariya oasis, about 48 km to the south-west. See also Strabo (17.3.19) for the same information. 73 Parke (1967: 233) is of the opinion that ‘Procopius has mistakenly reduplicated the name of the actual oasis of Augila and means to refer to the oasis of Ammonium’. Frankfurter (1998: 157) simply takes the passage to be referring to the oracle at Ammonium, and ignores the reference to Augila. Kuhlmann (1998: 174) suggests that ‘Procopius got his geography mixed up, placing a temple of Ammon and of Alexander the Great at Augila Oasis instead of Ammonium and Bahariya, respectively’. Bahariya is about 420 km from Siwah as the crow flies. 74 Smith (2003: 470–1) shows that ancient writers were ‘at the mercy of their sources’ and that few Graeco-Roman writers or their sources ventured into the Sahara, and nor did most of the Arab writers before the opening of the trans-Sahara trade route. One oracle too many? 181 75 See note 69 above. 76 Cameron (1985: 123 contra Athanassiadi (1993: 26): ‘in the oasis of Augila on the other hand, where religious life still thrived, Justinian reformed the locals’ ‘polluted ancestral customs’ by ‘intense missionary work’. 77 There is some debate as to whether this beatific picture of mass conversions to Christianity presented by Procopius is a realistic one, as mentioned earlier in this chapter. For example, Cameron (1982: 40) writes: ‘To begin, there seem to be no Moors on record to have carried Christian or Hebrew names’, an indication that the Bible did not play a great role in their education or folklore. 78 Literally translated, the meaning is more ‘bringing to ruin/destroying their ancestral pagan customs’. 79 Constantelos (1964) 376. 80 Cameron (1985) 89. 81 Shea (1973: 126). Shea (1998: 38) interprets this to mean that Jupiter’s message was delivered through Ammon and that this passage ‘clarifies the relationship between Jupiter and Ammon’, since ‘Ammon is then similar to Apollo; he makes Jupiter’s will known to man’. This explanation seems to over-rationalize the poet’s use of syncretic synonyms, however. 82 This allows the poet to play with a bit of irony, for example that John crucifies the leaders of the rebellion, when the oracle had promised that the land would be theirs (Ioh. 7.498–542). See Gärtner (2008) 22. 83 Hdt. 1.51–3. Silius Pun. 3.700–12; 9.249–66. See also Plutarch (Mor. 407 A–B) on general suspicion around deliberate oracular ambiguity. 84 The ‘unbearded general’ was the eunuch Solomon. 85 Cameron (1985) 182. Mattingly (1994: 304) supports the abandonment of Lepcis as plausible. See also Brett and Fentress (1996) 79–80. 86 The Igbo of Nigeria are known to use oracles, usually women, as criminal courts, serving several communities. There are a number of other examples in Whittaker (1965) 30–47; Smith (2003) 497–8. 87 Smith (2003: 497) writes: ‘Almost every account of the Amazigh or Tuaregs from the last two centuries marvels at the high position women enjoyed’. 88 Smith (2003) 482–3, 496–7. Smith also comments that tribes deeper in the interior provide more evidence of matrilineal descent. 89 Leone (2013) 46–55. 90 The name Kahina is possibly derived from the ancient Jewish KHN, but this is not certain. Ibn Khaldûn names her as a Jewess, but she may have been Christian. 91 The earliest reference is from Waqidi in the ninth century, followed by Ibn al-Athir in the thirteenth. For later sources, see Hendrickx (2013) 48–9. 92 Hendrickx (2013) 57. 93 Brett (1979: 509–12) also suggests that she may have been a man, a debatable point in material that seems to be largely legendary. See also Hannoum (2001) 15. 94 Hendrickx (2013) 56. 95 Cod. Theod. 16.10.9; Cod. Iust. 1.11.2. In 400 ce, another edict of Honorius and Arcadius reinforced this, closing down all pagan temples. See Cod. Theod. 16.10.13. 96 Kuhlmann (1998) 174; Cameron (1985) 89 n. 40. 97 Constantelos (1964: 372) divides this survival into two groups: those with Graeco- Roman or Hellenic beliefs, and the pagan beliefs of other peoples within the empire. 98 See also John of Ephesus (Hist. Eccl. 3), who relates that many pagans continued, some even holding office. The argument is supported by Fontenrose (1978: 5 n. 7), who argues that although pagan temples had been closed, pagan worship continued well into the sixth century. 99 Mattingly (1994) 39. 100 Bates (1914) 197. 182 Martine de Marre 101 See the archaeological reports of Brogan and Smith (1984: 80–92), citing epigraphical evidence (both Latin and Libyan), altars, Islamic pottery, coins, wood and charcoal samples, skeletal and botanical remains and textiles. 102 The Tuareg and Kabyle are probably the best known; other large groups are the Chaouis and Tashlhiyt. On modern Berber history and traditions, see Brett and Fentress (1996); Gélard (2006) 81–102; Wysner (2013) 1–12; Tishken (2007) 1468–82. 103 Belayche (2007) 171–91. 104 Eunapius of Sardis (Vitae 7.6.3–9, 480), for example, expresses scepticism, pointing out that plotting on the basis of an oracle was foolhardy, even if the oracle was correct, and believed firmly in portents, especially through dreams. See Zos. 4.18. 105 Treadgold (2007) 135. 106 Gregory (1983: 365): ‘Whatever the original source and inspiration of the oracle, to the Christian authors the text was a demonstration of the falseness of the pagan oracles and the futility of recourse to them’. 107 This was part of a long tradition which held oracles to be evidence in religious conflict. See Gregory (1983) 356. See also Shea (1998) 34; Jerard (1995) 16. 108 Cameron (1984); Cameron (1985). 109 Mattingly (1994) 39. 110 Frankfurter (1998: 167) writes: ‘the traditional association of oasis cultures with the god Amun’. See also Mattingly (1994) 33. 13 Deconstructing divination Superstition, anticlericalism and Cicero’s De Divinatione in Enlightenment England, c. 1700–1730

Katherine East

Introduction Anticlericalism was one of the defining topoi of heterodox writing within the English discourse of the Enlightenment.1 The clergy possessed an immense authority over the spiritual well-being of their flock; they were there to ensure the laity’s access to the spiritual world, to interpret the Bible on the people’s behalf and to exercise their special position as interpreters of the divine will. It was to the clergy that people went for reassurance about future worries, and for indications of divine intent. This was an unacceptable level of influence over the lives and minds of men in the view of those who declared war on this ‘priestcraft’, a term used broadly to encompass the crimes of the clergy, their harvesting of power from the dependency of their flock and the theological foundations of that power. The anticlerical challenge was mounted on the argument for a natural religion in place of a revelatory one: if religion was bound by the laws of nature, as these men believed it to be, then so was God, a conclusion which eliminated the possibility of miracles, portents, providence and, most importantly, revelation.2 In this understanding of God and nature, the clergy became superfluous; what need was there of a body to interpret the divine on behalf of the laity when all truly divine acts were fully accessible to man’s natural reason? If it proved inaccessible to that reason, if it required divination or interpretation, then it was simply not part of the true religion. It was for this reason that the rhetoric of superstition, a feature of religious discourse throughout its history, became one of the means by which anticlerical writers constructed their attacks. The association of the clergy and its power with superstition would be the most effective strategy for solidifying its segregation from true religion. As pamphlets and treatises waging this war on priestcraft flew from the presses, Cicero’s De Divinatione assumed a position of prominence in these texts.3 The work contains two books, in the first of which ‘Quintus Cicero’ presents the Stoic arguments in favour of divination, and in the second ‘Marcus Cicero’ (hereafter Cicero-as-Marcus) counters with an extensive critique of Quintus’ arguments by employing the rational strategies of an Academic Sceptic, a structure which provides a wealth of material for different theological stances. The explanation of the Stoic conception of a providential god in the first book 184 Katherine East provided material for orthodox Christians, particularly those Latitudinarians hoping to reconcile Newtonian physics with the traditional tenets of their faith, and for whom Stoic theism had yet to lose its lustre.4 The rebuttal of Stoic providence and the associated rationalization of religion in the second book inevitably attracted the attention of the champions of natural religion in the Enlightenment.5 It is the attraction of the second book of De Divinatione for the anticlerical cause among heterodox writers which will be considered here, and the evolution of this text into an effective weapon which could be deployed in the war on priestcraft. The material was clearly present; the second book, in which the arguments are presented by a character bearing Cicero’s own name (a controversial point which will be expanded upon later), amounts to an extended refutation of the belief in divination, and consequently the belief in the possibility of interpreting the divine will, and even the idea that the divine seeks to communicate that will with mankind. This target was sufficiently close to that of the anticlerical writers that the arguments and strategies in evidence were adopted and deployed in their own battle against claims to a special relationship between the clergy and God. While the examples of how to disprove divination provided by the second book of De Divinatione were useful, there was an additional facet of its argumentation which appealed particularly to anticlerical writers: the accusation that divination was a superstition, and the consequential argument that it must therefore be held separate from true religion, as religio and superstitio are separate and mutually exclusive entities. This Ciceronian superstitio offered a means by which sacerdotal authority could be first identified as superstitious, and then rejected on the basis that a superstition could not be part of the true religion.6 The engagement with De Divinatione which grew from this identification will be illustrated here with reference to three of the most prominent anticlerical writers of the English Enlightenment: John Toland, Anthony Collins and Matthew Tindal.7 After establishing the significant role played by the rhetoric of superstition in the heterodox discourse of the early Enlightenment, the efforts of these three men to associate De Divinatione with the fight against superstition, particularly by characterizing Cicero as an enemy of superstition and De Divinatione as his most explicit challenge to superstition, will be investigated. It was the understanding of Ciceronian superstitio in this dialogue which informed how these anticlerical writers constructed their argument that the clergy itself drew its power from superstition, and moreover exploited that superstition for its own gain, severing it from true religion.

Heterodox approaches to superstition In 1741, David Hume (1711–1776) – that foremost figure of the Scottish Enlightenment – included among his Essays, Moral, Political, and Literary a very brief essay on the subject ‘Of Superstition and Enthusiasm’.8 While the primary purpose of this essay was to demonstrate the fundamental opposition of these two afflictions, with his treatment of superstition Hume also provided an extremely Deconstructing divination 185 useful summary of the interpretation of superstition prevalent among the heterodox in the early Enlightenment. Accordingly, it will be used here to illustrate some of the principles which guided the integration of superstition into Enlightenment discourse. Hume begins with the uncontroversial and fundamental assumption that superstition should be explained in terms drawing on its identification as separate from, indeed opposed to, religion itself: ‘That the corruption of the best things produces the worst, is grown into a maxim, and is commonly proved, among other instances, by the pernicious effects of superstition and enthusiasm, the cor- ruptions of true religion’.9 This is a definition which had governed discussions of superstition since antiquity, and which is indeed strongly associated with Cicero himself, due to his pledge at the end of De Divinatione ‘to extend the influence of true religion, which is closely associated with the knowledge of nature, so it is a duty to weed out every root of superstition’.10 While the opposition of superstition and religion was an established principle in intellectual discourses concerning religion, what proved more fluid, with infinite repercussions for that discourse, was the understanding of what constituted this ominous entity ‘superstition’.11 Encompassing understandings from inappropriate attempts to influence the future, to a means of condemning particular practices, most notably witchcraft, to an accusation thrown at Catholicism in its entirety during the Reformation, it was a flexible tool for denouncing the ‘other’ in religion. Returning to David Hume, his description of the origins of superstition articulates well the meaning superstition had come to assume among the heterodox by the mid-eighteenth century:

The mind of man is subject to certain unaccountable terrors and appre- hensions, proceeding either from the unhappy situation of private or public affairs, from ill health, from a gloomy and melancholy disposition, or from the concurrence of all these circumstances. In such a state of mind, infinite unknown evils are dreaded from unknown agents; and where real objects of terror are wanting, the soul, active to its own prejudice, and fostering its predominant inclination, finds imaginary ones, to whose power and malevolence it sets no limits. As these enemies are entirely invisible and unknown, the methods taken to appease them are equally unaccountable, and consist in ceremonies, observances, mortifications, sacrifices, presents, or in any practice, however absurd or frivolous, which wither folly or knavery recommends to a blind and terrified credulity. Weakness, fear, melancholy, together with ignorance, are, therefore, the true sources of Superstition.12

Evident here is the understanding that superstition takes root and thrives where reason is absent. The opposition between superstition and reason became the new framework around which the rhetoric of superstition functioned in the debate, particularly in the works of the heterodox.13 In 1683, Charles Blount (1654–1693), whose works performed a crucial service for English Deism by transmitting the ideas of both Herbert of Cherbury and 186 Katherine East Baruch Spinoza, had defined superstition in terms very similar to those used by Hume. The belief in miracles, he explained, was rooted in superstition:

For the Minds of men being naturally prone to be agitated betwixt Fear and Hope of the future (the two grand Passions that govern humane life) thence it comes to pass, that they very often fancy a certain extraordinary divine power in all Contingents which are unusual, and the natural Causes of which they do not comprehend, as if those Contingents certainly proceeded, not from the order of Nature, but from an immediate operation of God transcending or changing that order; and that they presignified some good or evil Fortune to themselves.14

It is apparent that this was an echo of the broader issues governing the debate. The relationship between God and nature had become the focal point of disputes as support for a natural religion, in which the divine power was entirely constrained by the laws of nature, gained traction. In this natural religion, reason became a guiding force. The association of the antithesis of religion (namely superstition) with the antithesis of reason and nature (namely irrationality) was thus a logical conclusion.15 In the heterodox discourse, the function of superstition was increasingly defined by its complete and direct opposition to reason. This facilitated the integration of superstition into another prominent part of the English Enlightenment discourse: anticlericalism. This is once more illustrated by Hume who, having established what superstition and enthusiasm were, began enumerating their consequences for society. Regarding superstition, foremost among these consequences was the enhancement of the power of the clergy:

My first reflection is, That superstition is favourable to priestly power, and enthusiasm not less or rather more contrary to it than sound reason and philosophy. As superstition is founded on fear, sorrow, and a depression of the spirits, it represents the man himself in such despicable colours, that he appears unworthy, in his own eyes, of approaching the Divine presence, and naturally has recourse to any other person, whose sanctity of life, or perhaps impudence and cunning, have made him supposed more favoured by the Divinity. To him the superstitious intrust their devotions: To his care they recommend their prayers, petitions, and sacrifices: And by his means they hope to render their addresses acceptable to their incensed Deity.16

Numerous heterodox writers sought to cement this association, from Herbert of Cherbury to Charles Blount to John Toland to Matthew Tindal. In 1709, the Whig writer John Trenchard (1662–1723) used The Natural History of Superstition to perpetuate the anticlerical agenda which dominated so much of his work:

Though true Religion improves the Faculties, exhilarates the Spirits, makes the Mind calm and Serene, renders us useful to Society, and most active in the Affairs of the World, yet I dont know how it has happened, that in Deconstructing divination 187 all Ages and Countries, Fanatical, Melancholly, Enthusiastick, Monkish, Recluse, Sequestred Persons have passed upon the World for Religious, such who lived in Cloisters and Caves or became Pilgrims and Hermits, who seeming not to mind the Affairs of this World, were believed to know more of the next.17

This, then, was how the rhetoric of superstition developed among the heterodox into a tool to be employed in the deconstruction of clerical authority, by arguing on the basis that authority fed off irrational hopes and fears, and must consequently be identified as a superstition.

Cicero: the enemy of superstition The promotion of superstition within heterodox discourse was accompanied by the integration of Cicero, here assuming the role of the ‘enemy of superstition’, achieving an almost paradigmatic status in this incarnation due to his forceful rejection of superstition at the conclusion of the second book of De Divinatione, referred to above and here quoted in full:

Speaking frankly, superstition, which is spread among the nations, has taken advantage of human weakness to cast its spell over the mind of almost every man. This same view was stated in my treatise On the Nature of the Gods; and to prove the correctness of that view has been the chief aim of the present discussion. For I thought that I would be rendering a great service both to myself and to my countrymen if I could tear this superstition up by the roots. But I want it distinctly understood that the destruction of superstition does not mean the destruction of religion. For I consider it the part of wisdom to preserve the institutions of our forefathers by retaining their sacred rites and ceremonies. Furthermore, the celestial order and the beauty of the universe compel me to confess that there is some excellent and eternal Being, who deserves the respect and homage of men. Wherefore, just as it is a duty to extend the influence of true religion, which is closely associated with the knowledge of nature, so it is a duty to weed out every root of superstition.18

This passage represents the conclusion of a book in which the character Cicero-as- Marcus had rejected the arguments for divination point by point, primarily Stoic in their understanding, offered by his brother Quintus in the first book. Cicero’s decision to present this attack on religious divination under his own name, and to conclude it with such a statement of intent against superstition still under his own name, inevitably drew the enthusiastic attentions of heterodox writers. As noted above, Cicero’s positioning of religio and superstitio in direct opposition to each other solidified a way of comprehending superstitio which was influential until well into the eighteenth century.19 The consequence in De Divinatione is the use of superstitio as a means of characterizing divinatio so that it can be ostra- cized from appropriate religious practice. The practices identified with divination 188 Katherine East are dismissed as superstitions, defined as such on the basis that they draw their power from irrational fear: ‘what wonder, then, if in auspices and in every kind of divination weak minds should adopt the superstitious practices which you have mentioned and should be unable to discern the truth?’20 Again and again, the ability to reject a particular divinatory endeavour as a superstition, and to consequently make the case for the exclusion of said practice from religion, is utilized:

What a conflict this is! In view, then, of the differences between different nations in the responses, in the manner in which observations are made and in the kinds of birds and signs employed, need I assert that divination is compounded of a little error, a little superstition, and a good deal of fraud? And to these superstitions you have actually joined omens! … Then you go on and speak of the order of silence, favete linguis and the ‘prerogative’, or omen of the elections. This is indeed turning the artillery of one’s eloquence against oneself! For while on watch for these ‘oracles’ of yours could you be so free and calm of mind that you would have reason and not superstition to guide your course?21

In this sustained use of superstitio as a means of condemning aspects of religious practice as inappropriate to the true religion due to their irrationality, existed a strategy for employing the rhetoric of superstition to segregate certain elements of religion deemed unacceptable. This image of Cicero as the enemy of superstition was eloquently presented by John Toland (1670–1722), whose anticlerical and heterodox works provoked controversy in the first decades of the eighteenth century. In 1712, Toland wrote Cicero Illustratus, in which he presented his plans for a new edition of Cicero’s complete works to Prince Eugene of Savoy, his intended sponsor and his cor- respondent in matters heterodox.22 Indicating his plans to include an index in the edition recording all of the Ciceronian passages relevant to the Christian faith, Toland declared that ‘Tully can be called the hammer of Superstition before all other mortals’.23 Earlier in the work, Toland had reflected on the doubts which arose from Cicero’s philosophical dialogues, doubts which emanated from the obscurity regarding where Cicero’s voice should be located in these works, when often the author distanced himself through the use of different characters and historical settings. The subject of De Divinatione naturally arose in this context, as the work in which this dilemma was most pertinent: Cicero wrote the second book apparently under his own name, suggesting that it might reflect his own personal views on the matter, yet it seemed to contradict the stance taken in favour of Stoic theism in the preceding dialogue, De Natura Deorum, and the stance he often assumed in his speeches in favour of traditional Roman religion and its divinatory practices.24 Toland confronted this dilemma, arguing that the statements at De Divinatione 2.148–9, must be read as representative of Cicero’s true views, not simply the articulation of a possible standpoint by a character in a dialogue:

I would like [the reader] to notice that Cicero plainly removes his mask in De Divinatione (which, as he often says himself, is simply a continuation of Deconstructing divination 189 De Natura Deorum), and confirms these things completely in his own name. But, fearing that Readers would not finally understand his mind, he declares the meaning of these Books at the end of the second book of De Divinatione.25

Cicero’s words on superstitio and religio at the end of De Divinatione are then quoted in full. Toland’s determination to affirm these words as truly Ciceronian reflects the prominence they had and would continue to possess in his broader corpus. Lines from these passages had been employed by Toland in 1709, appearing on the frontispiece of his Adeisidaemon, a work seeking to defend Livy from the accusation of being a ‘superstitious man’: ‘Ut Religio propaganda etiam, quae est juncta cum cognitione Naturae; sic superstitionis stirpes omnes ejiciendae’.26 Cicero’s repudiation of superstition is again quoted in full by Toland in the partner work to Adeisidaemon, Origines Judiciae, a refutation of Pierre Daniel Huet’s Demonstratio Evangelica (1679) and its presentation of the traditional view of Moses as a prophet, in which Toland crafted an alternative account of Moses as an historical and political figure. De Divinatione is quoted towards the beginning of this work, followed by the declaration that Toland wants the same sentiment understood about himself: that while he impugns superstition, he will fight for religion.27 In 1720, the passage was once more quoted in full by Toland in his work Pantheisticon, which reimagined the Christian liturgy to suit the purposes of a pantheistic society. The Modiperator, leader of the congregation’s meetings, after reciting Cicero’s definition of reason from the third book of De Republica, calls on his audience to always follow that law, before reciting for them the passage from De Divinatione.28 An exchange between the Modiperator and the congregation follows in which the difficulties faced by the superstitious man are enumerated, the obvious conclusion being that Pantheism will liberate men from the tyranny of such superstitions, much in the manner Cicero deems desirable in De Divinatione. Anthony Collins (1676–1729), a Freethinker and Toland’s friend and ally in the anticlerical cause, also saw in De Divinatione an opportunity to depict Cicero as the adversary of superstition.29 In 1713, Collins published A Discourse of Free- Thinking, the work for which he is best remembered, and which ensured that his primary legacy was as one of the foremost proponents of that philosophy. This is a text in which Collins reiterated those beliefs which had already gained him notoriety – namely a radical Deism and an overt hostility to the established Church – and advocated a philosophy in which any belief could be challenged, and discarded if it failed to meet the requirements of rational law. To this end, in the Second Section of the Discourse, Collins presented a series of arguments intended to confirm his thesis that it is the duty of all men to think freely on questions concerning God and the Scriptures. The third argument offered is that ‘there is no remedy for the great Evil of Superstition, but thinking freely on these Points’.30 There Collins made his case that ‘Superstition is an Evil, which either by the means of Education, or the natural Weakness of Men, oppresses almost all Mankind. And how terrible an Evil it is, is well describ’d by the antient Philosophers and Poets’.31 Whom should he quote in order to confirm this point, 190 Katherine East but Cicero? Moreover, later in the Discourse he describes De Divinatione as a work in which Cicero ‘baffles all the Stoical Arguments for Superstition, openly under his own name’, and ‘destroy’d the whole Reveal’d Religion of the Greeks and Romans, and show’d the Imposture of all their Miracles, and Weakness of the Reasons on which it was pretended to be founded’.32 Over a decade later, Cicero was still being utilized as something of a Gospel on Superstition (or the Gospel against Superstition) in anticlerical discourse. In 1730, Matthew Tindal (1657–1733) – another Freethinker and heterodox writer – looked to De Divinatione in his Christianity as Old as the Creation.33 This, his final work, once more took up the argument for the supremacy of natural over revealed religion, directing it particularly against the power of the clergy. As can be expected, when disputing the possibility of revelation, the question of superstition arose. Written as a dialogue, the first interlocutor asks, ‘but if every Thing, as you contend, ought to be look’d on as superstitious which is not of a moral Nature, Superstition has spread itself over the Face of the Earth, and prevail’d more or less in all Times and Places’.34 The second interlocutor replies:

this is no more than what has been own’d long ago by a very good Judge, who says, Superstition, which is widespread among the nations, has taken advantage of human weakness to cast its spell over the mind of almost every man. And the Universality of Superstition is in Effect own’d by every Sect, in affirming that Superstition is crept into all other Sects; and that ’tis the chief Business of their respective Teachers to promote it.35

Later in this text, Tindal calls once again upon De Divinatione as evidence of Cicero’s campaign against superstition, undertaken despite his own status as a priest: ‘Of this, Cicero is a remarkable Instance; who, in his Book de Divinatione, exposes the Superstition of his own Country-men, and ridicules those Miracles, with which the Annals of the Church-Priests were fill’d’.36 The characterization of Cicero as the enemy of superstition was of great importance to these anticlerical writers seeking to engage in a discourse in which the accusation of superstition remained a powerful rhetorical and ideological weapon. What particular value did Ciceronian superstitio hold for these men?

Interpreting Ciceronian superstitio A revealing insight into the understanding of Ciceronian superstitio which under- pinned its anticlerical popularity is provided by an exchange between Anthony Collins, vocal critic of the clergy, and Richard Bentley who, as well as being one of the most notable classical scholars of the English tradition, was a clergyman and a great champion of the Anglican rational cause. In 1713, Bentley had felt compelled to pen a response to Discourse of Free-Thinking in order to defend the clergy against Collins’ barely veiled accusations and condemnations.37 He addressed Collins’ arguments point by point, including his use of Cicero to support his assertion – noted above – that Freethought was the only means by which Deconstructing divination 191 superstition’s power over the minds of men could be overturned. The Ciceronian passage used by Collins reads as follows in his translation of the text:

If you give way to Superstition, it will ever haunt and plague you. If you go to a Prophet, or regard Omens: if you sacrifice or observe the Flight of Birds; if you consult an Astrologer or Haruspex; if it thunders or lightens, or any place is confirm’d with Lightning, or such like Prodigy happens (as it is necessary some such often should) all the Tranquillity of the Mind is destroy’d. And sleep it self, which seems to be an Asylum and Refuge from all Trouble and Uneasiness, does by the aid of Superstition increase your Troubles and Fears.38

Responding to Collins’ Freethinking strategy to guard against superstition, Bentley wrote that:

One of his Capital Arguments is from the Evil of Superstition, which terrible Evil and great Vice can never be avoided, but by turning Free-thinker, that is (in plainer English) abandoning all Religion. Strange! That Superstition and Religion, which have been distinguish’d and divided this two thousand Years, should yet stick so fast together, that our Author cannot separate them: so that to ease himself of the One, he must abdicate Both.39

This dismissal precedes an extended critique of Collins’ translation of the Ciceronian passage, a critique which challenges and disparages Collins’ linguistic abilities and understanding of his Ciceronian source clause by clause. Concluding his rebuttal, Bentley declares that:

His dismal Description of [superstition] is in the words of Cicero; which chiefly relate to little Bigotries in Civil Life, not to fabulous Conceptions of the Supreme Being. And his Inference from thence is exactly, as if I should now say to You: Sir, you must renounce your Baptism and Faith, or else you can never be rid of those terrible Superstitions about the Death-watch, Thirteen at one Table, Spilling of Salt, and Childermas-day.40

Herein lies the distinction. According to Bentley’s reading, the superstitions which Cicero rejected in De Divinatione were the small incursions, the out-dated or inappropriate practices, or practices performed incorrectly or for untoward ends. For Collins, however, ‘there is no just Remedy to this universal Evil [of Superstition] but Free-Thinking. By that alone can we understand the true Causes of things, and by consequence the Unreasonableness of all superstitious Fears’.41 Collins explicitly adopts the understanding of superstition which places it in direct opposition to reason, determining all that is irrational in religion to be superstitious, and provoking Bentley’s outraged response that by employing this definition Collins is in fact dismissing religion as a whole. Here the matter of revelatory and natural religion ultimately divides the two men: only an entirely rational religion is acceptable to 192 Katherine East Collins, while Bentley’s religion still requires space for a providential god. So, while Bentley perceives in Cicero’s treatment of superstition in De Divinatione an account of the usual inappropriate religious practices to be encompassed as superstitions, Collins perceives a confirmation of the equation of superstition with all that is contrary to rational law. Collins’ interpretation depends heavily on the second book of De Divinatione and its conclusion. Not only does Cicero-as-Marcus associate the origins and power of superstition with ‘human weakness’, with credulous men who lack reason, but throughout the second book he deploys ratio as the means of exposing religious practices as superstitions.42 When addressing Quintus’ examples of the success of divination through dreams, Cicero-as-Marcus asks, ‘which is more consonant with philosophy: to explain these apparitions by the superstitious theories of fortune-telling hags, or by an explanation based on natural causes?’43 For Cicero-as-Marcus, who takes the part of the Academic Sceptic in this dialogue, the deployment of ratio against the examples and precedents provided as evidence by Quintus is the natural means by which to challenge the legitimacy of the arguments presented to him. This repudiation of divination as superstitious on account of its inconsistency with rational argument was clearly going to appeal to the heterodox readers of the early Enlightenment, for whom it would serve as a model for engagement when faced with their own notion of superstitious practice: the authority claimed by the clergy.

Using Ciceronian superstitio: profiteering from priestcraft If we return to John Toland’s Pantheisticon, the connection forged between Ciceronian superstitio and the clergy is articulated when Div. 2.148–9 is recited. First, the Modiperator and his congregation have a series of exchanges in which is reiterated the principle that superstition encompasses all that is irrational, as the congregation declare, ‘We want to be made ready and ruled by this Law: [the law of reason, whose definition is provided by Cicero in the third book of De Republica, which has just been recited] Not at all by the mendacious, and superstitious fabrications of men’. The Modiperator then states that ‘False Laws are neither clear, nor universal, nor always the same, nor ever efficacious’, to which the congregation replies: ‘Therefore they are useful to few, or to none at all, With the sole exception of the Interpreters’.44 The lesson being that those beliefs which exist outside the realm of reason and which consequently rely on the interpretation of others for their meaning to become clear must be categorized as superstitions, profitable solely to those whose role it was to ‘interpret’ their meaning. The reader is left in no doubt as to who these interpreters exploiting superstition are, when, after quoting the passage from De Divinatione, the congregation recites: ‘The Superstitious Man is tranquil Neither awake nor asleep; He neither lives happily, Nor dies fearlessly: Alive and dead, he is made the prey of Priests’.45 The clergy’s claim to power, that they are in possession of a special authority as interpreters of the divine on behalf of the laity, is therefore identified as a superstition according to the terms identified in Cicero’s De Divinatione. Deconstructing divination 193 The same use of Ciceronian superstitio is in evidence in Anthony Collins’ celebration of Freethought in 1713. In this work Cicero becomes the subject of focused discussion when he is identified as one of the forebears of Freethought. While considering the theological dialogues De Natura Deorum and De Divinatione, Collins makes the following accusation:

now the modern Priests, whenever they meet with any Passage favourable to Superstition, which Cicero puts in the mouth of the Stoick, or any false Argument which he makes the Epicurean use, and which they have thought fit to sanctify … they urge it as Cicero’s own, and would have the Reader believe Cicero look’d on it as conclusive.46

Again, the clergy are denounced as figures who foster and feed off superstition for their own benefit. This is also the stance of Matthew Tindal, as demonstrated by his discussion of superstition cited above, and his assertion there that in all religious sects: ‘’tis the chief Business of their respective Teachers to promote [superstition]’.47 Each of these anticlerical writers presents a vision of the clergy as a body which exploits the irrational fears of men – the fear of death, most particularly, and fear of divine retribution – for their own profit, constructing their power from that fear. The Ciceronian definition of superstition as that which contradicts reason thereby provides the means of turning the rhetoric of superstition against the clergy, for their very authority is based in the irrational. The clergy are in fact further condemned, for not only did they exploit superstition to ensure the dependence of the laity, but also to advance their influence in the political sphere, by facilitating the accumulation of power by others. Ciceronian examples are again invoked to illustrate this additional facet of the clergy’s profiteering from superstition. One particular instance, in which Cicero-as-Marcus scorns the divinatory power of the Sibylline Books due to their exploitation by their interpreters, the quindecimviri, garnered attention:

We Romans venerate the verses of the Sibyl who is said to have uttered them while in a frenzy. Recently there was a rumour, which was believed at the time, but turned out to be false, that one of the interpreters of those verses [Lucius Cotta] was going to declare in the Senate that, for our safety, the man whom we had as king in fact [Julius Caesar] should be made king in name.48

John Toland referred to this dismissal of the Sibylline Books and their interpreters in his Two Essays in a Letter from Oxford (1695).49 Addressing what he terms the ‘Rise, Progress, and Destruction of Fables and Romances’, Toland used the condemnation expressed by Cicero-as-Marcus and directed it against the efforts of the Church to arm itself with equivalent oracles through which it might influence contemporary affairs. The Sibylline Books and their potential for corruption are also a point of discussion in the writings of Ralph Cudworth who, as a defender of the orthodox and a foremost figure among the Cambridge Platonists, was situated in a very 194 Katherine East different theological arena from Toland, yet who in his True Intellectual System of the Universe (1678) also cited Cicero’s condemnation of those oracles when considering their exploitation by early Christians.50 In an extensive discussion of the Sibylline Books, during which several excerpts from the De Divinatione are quoted, Cudworth argues that, although the verses were undoubtedly abused and corrupted, the foretelling of the rise of Christianity can be discerned therein. The abuse identified by Cicero is paralleled to that by those early Christian priests:

Now as Cicero seems to complain, that in his time these Sibylline Oracles were too much exposed to view, so is it very probable, that notwithstanding they were to be kept under the Guard of the Quindecimviri, yet many of them might be copied out, and get abroad, and thereby an occasion be offered, to the ignorantly zealous Christians, who were for Officious Lyes and Pious Frauds, to add a great deal more of their own forging to them.51

Not only was Ciceronian superstitio identified with priestly authority (although not by Cudworth, whose purpose was the endorsement of the established Church), but De Divinatione provided a rich resource for examples of how priests profited from that authority through the exploitation of superstition to facilitate the increased power of secular rulers. For heterodox writers such as Toland, this fostering of superstition among the laity amounted to a tyranny of the mind, made even more dangerous by its association with tyranny in the civil world.

Conclusion The profit to be won from prophecy can take innumerable different forms, as this volume demonstrates emphatically. Among the anticlerical writers of the English Enlightenment, it was fury at the perceived profit derived by the clergy from their special relationship with the divine – in the form of the dependence of their priests, and the ability of those priests to influence secular matters – which drove their war on priestcraft. When fashioning the strategies to be employed in that war, Cicero’s De Divinatione, with its unrelenting critique of claims to divine communication and intervention, proved an invaluable resource. A tract which explicitly condemned superstition, and in which superstition was characterized as the intrusion of the irrational into the true, natural, rational religion, the De Divinatione provided an understanding of superstition which could be turned against the clergy. Superstition became the pseudo-religious practices constructed by men – contrary to reason – in order to enforce their own power; their claims to power which existed outside the realms of man’s reason – essentially as interpreters of an aspect of divinity incomprehensible to the laity – could be identified and condemned as superstitions. In the hands of these anticlerical writers, the attack on superstition in De Divinatione became the outright rejection of the irrational in the true, natural religion and, consequently, the ideal weapon to turn against the power of the clergy. This is only one of the functions performed by De Divinatione in a broad and complex discourse, but it serves to illustrate the significance this Deconstructing divination 195 ancient text on ancient religious practices could attain in the context of an entirely modern discourse.

Notes 1 This anticlericalism – or the ‘war on priestcraft’ – continues to be a point of debate with respect to its radicalisation of the English Enlightenment. The assumption that this anticlerical fervour in English discourse signified a wish to see the Established Church overthrown can be found in the works of Israel (2001: 566–627) and Beiner (2011: 156–75), and in narratives championing the traditional perception of the period as the overthrow of the Age of Faith by the Age of Reason. An alternative reading – championed by Goldie (1993: 209–31); Popkin and Goldie (2006: 79–109); Champion (2003: 249–51); Champion (1992: 1–24, 173–9) – has worked to show that in attacking the privileges of the clergy with this rhetoric, these heterodox writers sought only to reform the Church, not to overpower it. 2 On the political context of these theological debates see Wigelsworth (2009) 109–41. 3 While no English translation of De Divinatione was made available until the nineteenth century, it was available in various forms to the men of learning engaging in this discourse. In 1721, John Davies of Queens’ College, Cambridge University, published an edition of De Divinatione, together with Cicero’s De Fato, as part of his project to contribute editions of Cicero’s philosophical works to the series begun by Johann Georg Graevius in 1684. De Divinatione was also available through the complete editions of Cicero’s works, the most recent of which had been published in 1692 by Jacobus Gronovius in Amsterdam, using the recension of the text produced by Janus Gruterus in Hamburg in 1618. Further complete editions would be produced throughout the eighteenth century by Isaac Verburg in Amsterdam in 1724, Johannes Albertus Ernesti in Leipzig between 1737 and 1739, and by Olivetus in Paris between 1740 and 1742. Several French translations of De Divinatione were published in the early eighteenth century, including those by Roland Desmarets (or Maresius) in 1710 and by L’Abbé Le Masson in 1721, both printed in Paris. The library of Anthony Collins provides an indication of the versions of De Divinatione actually consulted by men such as him. It includes the two French translations, the complete editions by Verburg, Gronovius and Gruterus, and the edition by John Davies. See Tarantino (2007). 4 The fate of Stoic theology in this period, in particular its shift from a favoured resource for theism to its later rejection as atheism, is recounted by Brooke (2012) 127–48. 5 The contribution of De Divinatione to theories of natural religion and the rise of Deism in the English Enlightenment has received some attention. See Gawlick (1963) 657–82; Zieliński (1929) 260–86. 6 East (2014) 970–5. 7 There is clearly a wide body of potential evidence regarding the use of Cicero’s De Divinatione beyond these three particular individuals, but in this chapter some chrono- logical and geographical boundaries are necessary. Tindal, Toland and Collins are the best examples of the particular engagement with Ciceronian superstitio in De Divinatione which I am illustrating here. 8 Hume (1788 [1741–1742]) 69–74. While there is insufficient space to discuss it here, it is worth noting that Hume’s works and thoughts also bore the marks of Cicero’s influence. See Berman (1980) 150–4; Fosl (1994) 103–20; Olshewsky (1991) 269–87; Harris (2015) 186–95. 9 Hume (1788 [1741–1743]) 69; emphasis original. On the separation of superstition and religion as an embedded topos, see Cameron (2010) 4–7; Martin (2004) 9–20; Santangelo (2013) 38–47. Several works produced in this period drew upon this opposition in their titles, such as the anonymous False Religion worse than No Religion: An Enquiry concerning Superstition, as it affects the Rights and Happiness of Civil Society: Written for the Advancement of True Religion and Virtue, London, 1730. 196 Katherine East 10 Cic. Div. 2.149: quam ob rem, ut religio propaganda etiam est, quae est iuncta cum cognitione naturae, sic superstitionis stirpes omnes eiiciendae (trans. Falconer). 11 On the differing forms superstition could take see Cameron (2010) 29–75; Thomas (1971) 113–50. 12 Hume (1788 [1741–1742]) 69–70. 13 Cameron (2010) 6. 14 Blount (1683) 3. On Blount’s treatment of superstition see Israel (2001) 360–3. There are also numerous examples of this association between superstition and irrationality in the works of John Toland (1696) 44; (1714) 29; (1700) 18. 15 See Harrison (1990) 5–18. 16 Hume (1788 [1741–1742]) 71; emphasis original. 17 Trenchard (1709) 16. 18 Cic. Div. 2.148–9: Nam, ut vere loquamur, superstitio, fusa per gentis, oppressit omnium fere animos atque hominum imbecillitatem occupavit. Quod et in iis libris dictum est, qui sunt de natura deorum, et hac disputatione id maxume egimus. Multum enim et nobismet ipsis et nostris profuturi videbamur si eam funditus sustulissemus. Nec vero – id enim diligenter intellegi volo – superstitione tollenda religio tollitur. Nam et maio- rum instituta tueri sacris caerimoniisque retinendis sapientis est, et esse praestantem aliquam aeternamque naturam, et eam suspiciendam admirandamque hominum generi pulchritudo mundi ordoque rerum caelestium cogit confiteri. Quam ob rem, ut religio propaganda etiam est, quae est iuncta cum cognitione naturae, sic superstitionis stirpes omnes eligendae (trans. Falconer). 19 This distinction is elaborated by Santangelo (2013) 38–47. Its legacy is considered by Martin (2004) 126–9. 20 Cic. Div. 2.81: quid mirum igitur si in auspiciis et in omni divinatione imbecilli animi superstitiosa ista concipiant, verum dispicere non possint? (trans. Falconer). 21 Cic. Div. 2.83: quid quod aliis avibus utuntur, aliis signis, aliter observant, alia respondent? Non necesse est fateri partim horum errore susceptum esse, partim superstitione, multa fallendo? Atque his superstitionibus non dubitasti etiam omina adiungere … iam illa ‘Favete linguis’ et ‘praerogativam, omen comitiorum’. Hoc est ipsum esse contra se copiosum et disertum, Quando enim ista observans quieto et libero animo esse poteris, ut ad rem gerendam non superstitionem habeas, sed rationem ducem? (trans. Falconer). 22 Toland was a prolific writer of political pamphlets and theological treatises, all of which were directed towards defending the Commonwealth and championing the rationalization of religion. On Toland’s contributions to heterodox thought see Champion (2003); Champion (1992); Israel (2001) 599–627; Wigelsworth (2009) 75–86, 143–8; Jacob (1976) 201–50; Hudson (2009) 81–97. 23 Toland (1712: 59): et Tullius profecto prae cunctis mortalibus Superstitionis malleus dici poterat. 24 The debate regarding where (or indeed whether) Cicero’s true voice can be located in De Divinatione and De Natura Deorum has a long history in Ciceronian scholarship, and will probably continue to exercise Ciceronian scholars for a long time to come. Some of the key studies include Beard (1986) 33–46; Schofield (1986) 47–65; Krostenko (2000) 353–91; Santangelo (2013) 10–36. 25 Toland (1712: 37–8): is advertat velim, eum in libris de Divinatione (qui, ipso pluries dicente, horum de Natura Deorum sunt tantummodo continuatio) larvam sibi aperte detrahere, ac eadem omnino suo ipsius nomine affirmare. Sed, ne mentem ejus non caperent tandem Lectores, subverens, satis speciatim subjectam illorum Librorum sen- tentiam in sine secundi de Divinatione. 26 Toland (1709). Typographical emphasis is as it appears on the frontispiece. 27 Toland (1709) 102–3. 28 Toland (1720) 69–70. 29 Anthony Collins was an English philosopher who produced numerous tracts which sought to demonstrate the rational basis of true religion, a focus which has situated Deconstructing divination 197 him within the radical Deist tradition. Due to A Discourse of Free-Thinking, published in 1713, Collins is also strongly associated with the development of Freethought in England, and its demand that all assertions meet the standards of reason to be accepted as ‘true’. On Collins see Wigelsworth (2013) 86–101, 112–23; Hudson (2009) 98–106; Tarantino (2014) 81–100; Berman (1980) 150–4; Berman (1975) 82–102. 30 Collins (1713) 35; emphasis original. 31 Collins (1713) 35–6, in which he quotes Cic. Div. 2.150. 32 Collins (1713) 110–11. 33 Another Deist writer, Matthew Tindal produced works which challenged the power held by the Church and its clergy on the basis of the rational basis of true religion. On Tindal see Stephen (2006); Hudson (2009) 106–13; Wigelsworth (2009) 58–64. 34 Tindal (1730) 148. 35 Tindal (1730) 148–9, quoting Cic. Div. 2.148. 36 Tindal (1730) 360. 37 Bentley (1713). 38 Collins (1713) 35–6, quoting Cic. Div. 2.150. The Latin text presented by Collins reads: Superstitio enim instat and urget, and quocunque te verteris persequitur: sive tu vatem, sive tu omen audieres; sive immolares, sive avem aspexeris, sive Chaldaeum; si haruspicem videris; si fulserit, si tonuerit, si tactum aliquid de coelo erit, si ostenti simile natum factumve quidpiam; quorum necesse est plerumque aliquid eveniat: ut nunquam quieta mente liceat consistere. Perfugium videtur omnium laborum and sollicitudinum esse somnus; at ex ipso plurimae nascuntur curae metusque. 39 Bentley (1713) 34; emphasis original. 40 Bentley (1713) 34–5; emphasis original. Regarding the superstitious practices identified by Bentley here: the Death-watch was the belief that if someone watching over the dying fell asleep in the hour before or after midnight, they would die within the year; Thirteen at one Table refers to the belief that if there are thirteen people seated around a table, one will die within the year, a superstition connected to the number of Jesus’ disciples; the Spilling of Salt refers to the belief that it is unlucky to spill salt, as Judas Iscariot spilled salt at the Last Supper; Childermas-day refers to Holy Innocents’ Day, 28 December, which marks the massacre of the children of Bethlehem ordered by King Herod, a day considered unlucky, particularly for entering agreements such as marriage. 41 Collins (1713) 37. 42 The references to human weakness or hominum imbecillitatem occur at Cic. Div. 2.148, 2.19, 2.81, 2.125. On how Ciceronian superstitio developed across his works, see Santangelo (2013) 37–47. 43 Cic. Div. 2.129: utrum philosophia dignius, sagarum superstitione ista interpretari an explcatione naturae? (trans. Falconer). See also Div. 2.83, 2.85, 2.100. 44 Toland (1720: 70): RESP. Hac Lege institui regique volumus: Haudquaquam mendaci- bus, Et superstitiosis hominum commentis. MOD. Non claræ sunt fictæ Leges, nec uni- versales, Non semper eædem, nec efficaces unquam: RESP. Paucis ergò, aut oppidò nullis sunt utiles, Solis exceptis Interpretibus. MOD. Aures interim advertite. 45 Toland (1720: 70): RESP. Non vigilat Superstitiosus, Non dormitat tranquillus; Neque beatè vivit, Neque securè moritur: Vivus and mortuus, Factus Sacrificulorum præda. On Toland’s use of Ciceronian scepticism against priestcraft in Pantheisticon see East (2016) 245–62. 46 Collins (1713) 110–11. 47 Tindal (1730) 168. 48 Cic. Div. 2.110: Sibyllae versus observamus, quos illa furens fudisse dicitur. Quorum interpres nuper falsa quadam hominum fama dicturus in senatu putabatur eum, quem re vera regem habebamus, appellandum quoque esse regem, si salvi esse vellemus (trans. Falconer). 49 Toland (1695) 31–2. In his Origines Judiciae, Toland again employed examples from De Divinatione to demonstrate priestly exploitation of superstition to facilitate politi- 198 Katherine East cal gain. Toland (1709: 167–8) quotes Cicero (Div. 2.141) on Alexander’s dream of a serpent; Toland (1709: 177–84) quotes Cicero (Div. 2.118) on the Pythian manipulation of prophecies to benefit Philip of Macedon. 50 Cudworth (1678). 51 Cudworth (1678) 283. 14 Prophecy and Paul Kruger Robert Grendon’s appropriation of Graeco-Roman prophets and prophetic devices in his South African epic, Paul Kruger’s Dream

Szerdi Nagy

In his epic poem Paul Kruger’s Dream, Robert Grendon tells the story of the rise and fall of the Boer nation and its leader, Paul Kruger. Grendon does this in the traditional style of the national epic, but his work comes with a twist. While most nationalist heroic epics tell the story of a great hero, in Grendon’s work his ‘hero’ presents himself to be an anti-hero of sorts. Grendon was known for his outspoken dislike for both Paul Kruger and the Boers, and as an educated coloured South African man, he positioned himself firmly in the camp of the British, under whose rule he thought he would have a better life. We know that Grendon studied Classics and took Latin to matric,1 and so had a fair knowledge of the classical world and both Greek and Latin literature. We also know that he took part in the South African War on the side of the British. It is with these things in mind that one must approach any reading of Paul Kruger’s Dream. As a classically trained non-white South African living at the turn of the twentieth century,2 Grendon attempted to legitimize his work with the widest possible audience.3 He did this by employing a number of recognizably classical literary devices and figures. This chapter will explore Grendon’s use and appropriation of prophecy as a device for narrative development. From 1867, Grendon was educated at Zonnebloem College in the Cape, a mission-run school set up to provide non-white young men with a vocational education. Those students who exhibited academic abilities received a matriculation- level education in subjects such as maths, philosophy and literature, which would allow them the opportunity of becoming school teachers. Grendon was one of the students who attended this institution and matriculated after having written a Latin paper in addition to his other subjects.4 Our partial knowledge of his Latin syllabus indicates that Grendon would have been familiar with authors such as Caesar, Cicero, Livy and Vergil. In the construction of his own epic, we can see borrowings from both Livy and Vergil, but it definitely seems that his reliance on the epic is more pronounced than his borrowings from the histories. We can see echoes of the Aeneid in a number of places, some of which will become apparent in this chapter. Paul Kruger’s Dream was published in 1902 and revised in 1904, after Paul Kruger’s death. It chronicles the origin, rise and what Grendon believed to be the fall of the Boer nation, from the arrival of the Dutch mariners at the Cape, to their Boer descendants’ migration into the interior of South Africa and 200 Szerdi Nagy their eventual downfall through defeat in the South African War,5 and the Swiss exile of their leader Paul Kruger. In some sense, then, Grendon’s plot can be seen as cyclical in nature.6 Paul Kruger’s speeches are interspersed by the speeches of various divinities, risen heroes and statesmen, some friendly to Paul Kruger, others not. For these reported sections Grendon employs fixed rhymed verse forms, and on the occasions when Paul Kruger himself speaks in rhymed verse, it is primarily to indicate a temporal space between the speech and his reporting of it.7 These prosodic and temporal variations enhance the narrative, guaranteeing the constant attention of the reader. From the 1903 edition of the Zonnebloem College Magazine, Christison quotes from a review of Paul Kruger’s Dream: ‘Grendon’s metres throughout are very varied, a well-chosen method of relieving the monotony almost necessary in a long Epic Poem’.8 In South Africa at this time there were many perceived nationalities and very little in the way of a homogenous national identity.9 In composing Paul Kruger’s Dream, Grendon made an attempt to forge a new national identity central to whose construction is the ideology of equality that included black and other ‘non- white’ people in a common civil society.10 The moral didacticism, the political overtones in his writing and the national scale of his envisioned readership all show that, like Vergil, Grendon saw epic as the optimum vehicle with which to transmit his ethical, political and nationalistic lessons.11 While Paul Kruger may seem an unlikely candidate for an epic hero, the principles for which he stands (in the context of the poem) are the very substance of the genre. Guérard believed that ‘no man, purely as an individual, is the proper subject for a true epic. A hero does not tower very high above the average human stature: he becomes “epic” only when he represents something greater than himself – a nation, a race, a faith’.12 That Paul Kruger’s Dream implicates the fortunes of an entire ‘race’ through the life of Paul Kruger is spelt out on more than one occasion.

Prophecy in Paul Kruger’s Dream Grendon employs many different prophetic voices in Paul Kruger’s Dream. These serve a number of purposes within the narrative, including the development of the plot, providing the reader with a foreshadowing of events to come, and giving Grendon himself a voice with which to share his own religious beliefs and political opinions. These prophetic voices have the dual purpose of inject- ing Grendon’s work with his own Christian voice, and also offer him a means to legitimize his epic through their various classical allusions, which lend a sense of inherited gravitas. The oracles received by Paul Kruger fall into a variety of categories, the three most prominent of which are adversative oracles, predictive oracles and unsolicited oracles.13 In the epic, Paul Kruger is visited by a number of figures including the deities Fortuna and Mars, and the personification of Truth, who all bring him both prophetic warnings and advice. In addition to these divine prophets, he is also visited by a number of historical figures, including the first governor of the Cape Colony, Simon van der Stel; the ill-fortuned Piet Retief; a group of Voortrekkers who were killed in 1838; the Voortrekker leader Pieter Prophecy and Paul Kruger 201 Uys and his son Dirk; the founder of Potchefstroom, Hendrik Potgieter; creator of the South African Republic, Andries Pretorius; and Johannes Brand, the fourth president of the Orange Free State. Grendon also includes a number of embedded ‘voices’ in the narrative who speak to Paul Kruger, but their identities are never revealed, either to him or to the reader.

The prophetess Fortuna appears to Paul Kruger The first of the prophetic visitors Paul Kruger encounters is the goddess Fortuna. She makes her first appearance to him in Part 2. While the Roman goddess of luck may not be a typical choice for the role of archetypal female prophetess (her role in antiquity is not traditionally that of a prophetess), Grendon clearly intended his readers to define her as such. It is no coincidence that Grendon chose a figure of some substance in the obvious form of the traditional-style prophetess of ancient epic for his first introduction of prophecy in his text. As the narrative unfolds, and the fundamental role of prophecy in the development of the plot becomes more apparent, it makes sense then that Grendon should choose to ground, if you will, his ‘prophetic framework’ in such an obvious classical allusion. He uses this Sibyl-like figure to situate his epic in a classical context, to lend it both credibility and an erudite gravitas. Grendon also manages to inject a distinctly Christian voice into her message. This Christian-influenced message continues throughout Paul Kruger’s interactions with the various prophetic voices that he encounters. Fortuna, who has been sent by Jove, appears to Paul Kruger in the ‘Land of Dreams’, where she provides a substantial and unsolicited prophecy in an iambic tetrameter (2.135). She proclaims: ‘A daughter of the Skies am I! Fortuna is my name; Betwixt this Earth and Heav’n I ply – Jove’s orders to proclaim’ (2.158–61). At this stage in the epic, Paul Kruger is still a young man unsure of his future and destiny. Fortuna warns him of his potentially destructive fate and the ruinous effect his future decisions will have on the Boer nation as a whole. As Paul Kruger’s Dream was only published after its subject’s death, the reader of the epic would be aware that Paul Kruger ultimately ignore this prophecy and that what Fortuna is prophesying will indeed come to pass. Fortuna advises Paul Kruger to ‘Pay goodly heed; be not allur’d By Pride and Tyranny; Lest captive in their chains secur’d Thou perish mis’rably’ (2.186–9). So, while her message is pertinent and vocalizes what many of Grendon’s contemporaries may have thought of Paul Kruger’s decisions, it is ultimately a warning which he ignored, as history can attest.14

The president as a prophet and the Boers as the new Trojans? Our next prophetic appearance is in Part 4, where the ex-Transvaal president, Andries Pretorius, appears to Paul Kruger in a way very similar to the way that Hector appears to Aeneas in Book 2 of the Aeneid (2.268–97). Pretorius, accompanied by some of his kinsfolk, appears to Paul Kruger and advises 202 Szerdi Nagy him to set aside the past troubles of the Boers and to rejoice in his (Pretorius’) accomplishments and those of his son who will succeed him, Marthinus. This idea of the son carrying on his father’s legacy being prophesied bears some notable parallels with the Aeneid, where we see Ascanius’ being groomed to take over the leadership of the Trojans from his father, Aeneas. When reference is made to the founding of a ‘new Boer homeland’ after their defeat at the hands of the British (5.40–50), echoing here the idea of a ‘new Troy’ that was to be established in Italy, there can be little doubt that Grendon is deliberately constructing this correlation between the Boers and Trojans. What makes this an even more interesting comparison is that this alignment of the Boers with the Trojans positions the British, by default, in the role of the Greeks. This perhaps not so subtle allusion would have been readily apparent to his readers, most of whom had been educated in the classical tradition. This equation of Trojan-Boer and Greek-British creates an interesting position that seems to set the tone for both future prophetic engagements and for plot development. Pretorius’ positive prophecy of the Boers’ future, coming so quickly after Fortuna’s warning-laden vision, seems to decrease the immediacy and the threat of Fortuna’s warning to Paul Kruger. The words of Pretorius’ prophecy seem to echo Kruger’s own thoughts whereas Fortuna speaks very much with the voice and knowledge of the author himself. While the readers of Grendon’s work would no doubt be aware of the validity and merit in Fortuna’s words, it would also be clear to them that the author’s intention for Pretorius’ prophesying to be seen as a false oracle was meant to mislead Paul Kruger. The reader would know that it in fact does mislead him, as the actions of Paul Kruger lead to the British defeat of the Boers in the South African War, which signalled the end of both Boer autonomy and the Boer nations. Already at this early point in the epic, and after only two (of the many) prophecies, we get a sense of Grendon’s use of prophecy as a tool to misdirect both the reader and his protagonist, Paul Kruger, while using other prophetic voices to illuminate the ‘Truth’ mentioned in the poem’s title.15 From Part 7 onwards, we see Pretorius’ prophecies come to partial fruition. We see this denouement in the Battle of Boomplaats in 1848, where Pretorius was defeated by the British, and following which Marthinus, his son, was voted in as president of the Transvaal Republic in 1855. Grendon grounds all of his prophecies in historical fact to lend them additional weight and credibility. Paul Kruger appears to be energized and revitalized after Pretorius’ prophecy is realized and his arrogance and treachery (characteristics which both Grendon and the British attributed to him), slowly start to develop and becomes more evident. Both historically and within the text, after the initial positivity of Marthinus’ rule, he is replaced by the incredibly inept Thomas Burgers in 1872. With this transition in leadership, we see our narrative jump ahead some seventeen years. Grendon is not one for smooth time transitions. Under Burger’s rule, we see the annexation of the Transvaal by the British in 1877 and at this point in the epic Paul Kruger is shown appealing for divine assistance to cast aside the British yoke (Part 8: introduction). Prophecy and Paul Kruger 203 Mars and the return of Fortuna It is then at the request of Paul Kruger himself that the next divinity enters the narrative. Mars, the god of war appears to the Boers. As anyone with even a basic classical education would know, in the Trojan War, Ares (Mars) fought on the side of the Trojans (Il. 5.699), so Mars helping the Boer states seems to be Grendon reiterating to his readers this identification of the Boers with the Trojans. Mars warns the Boers of the coming struggle and battles they will fight to regain their independence from the British.16 In Mars’ reiteration of Fortuna’s earlier prophecy, he not only lends it credibility, but also reiterates for the reader the impending doom faced by the Boers. Shortly after Mars’ appearance, Fortuna returns to the narrative. She has returned fifty-five years after her last appearance to Paul Kruger, which means that in the short space of seven parts of the epic, fifty-five years have gone by. As the entire epic consists of thirty-seven parts and takes us up to the end of the South African War in 1902, we can see that Grendon is placing much of his emphasis on the three years of the South African War, which take up the majority of the remaining parts of the epic. Fortuna sternly rep- rimands Paul Kruger for ignoring her previous warnings and admonishes him for blaspheming and turning away from God. Paul Kruger’s disdain for the Christian god is clearly visible:

Of glitt’ring gold! In tyranny I reign! My foot doth rest upon these aliens’ necks, Whose piteous cries of mercy I disdain, And laugh to scorn! Their God in this dread hour Of need is powerless to lend them aid! My rock, and strength is Mammon, who is lord, And king of Heaven’s God! Ye serfs, and dogs, Where now is your Eternal, Mighty, King? (13.8–13)

Fortuna tells him of his inevitable doom and the plagues that will come and affect his people.17 This prophecy begins to come true shortly afterwards with the Uitlander Crisis18 and the Jameson Raid,19 two historical events which Grendon includes at this point in his narrative to lend prophetic weight to Fortuna’s words. In this second and last visit by Fortuna, she has lost much of her Sibyl-like attributes. Her monologue directed at Paul Kruger contains far more Christian allusions and references, and the classical overtones of her first message are replaced by a strongly didactic and Christian message. By prophesying events that contemporary readers would know, or at least believe to be fact, Grendon is lending credibility and weight to the reasons he gives for the downfall of Paul Kruger and the Boers:

Now hearken, haughty tyrant! God hath heard Thy blasphemy, and wicked boastful shouts! 204 Szerdi Nagy Thy days are number’d, and thy doom is seal’d! Like Babylonia’s king, might’st thou behold Thy downfall trac’d upon thy palace wall! Perchance JEHOVAH in his furious ire May smite thee, and consign thee unto worms, Like him of yore, who did His pow’r usurp! Proud, impious chieftain, wherefore hast thou scorn’d, And disobey’d the words, which I to thee Did utter five and fifty years ago? Say, wherefore thou in tyranny dost reign; And wherefore thou dost trample on the necks Of yonder wretchèd serfs, who vainly cry To thee for mercy, and relief? Thy doom Is seal’d! Some day this land from thee must pass! Yon wretches most assuredly to arms Will rise! Rebellion broods within their breasts, And ere this year its closing days beholds, The yoke of serfdom will they strive to cast Aside! But they will fall beneath thy hand! (13.62–87)

The appearance of ‘truth’ and the shades of the dead With the departure of Fortuna, the plot slows down considerably, the remaining twenty-four parts of the epic covering only seven years of historical time. The next five parts detail the growing dissent within the Boer republics and the increasing unhappiness of resident aliens. Britannia appears and encourages the latter to rise up against the Boers and in Parts 20 to 24, we see Paul Kruger prepare himself for war and call the Boers together to meet at Paardekraal (now Krugersdorp). At this historical meeting of some 10,000 Boers, the personification of Truth appears and provides them with an unsolicited prophecy warning of the consequences of declaring war against Britain and of the eventual dispersal of the Boer people as a result of this war, and that even their ally, Mars, will turn against them (24.131–2). The Boers ignore Truth’s warning, preferring to face annihilation over being taken by the British. The remainder of Truth’s monologue degenerates into a sour attack on the Boers, which is far more personal and vicious than anything that has been said before. It almost appears that as his epic progressed, it became harder for Grendon to hide his personal beliefs and that the anger stemming from his personal experiences as a mixed-race Southern African who fought against the Boers became harder to suppress. This most certainly reflects a naïveté in the composition that he is unable to separate his emotion from his text. Unfortunately, this really affects the remainder of the epic, which has now lost its attempted neutrality, and its author’s blatant prejudices, which up until this point have remained at least partially hidden, have risen to the surface. It would be constructive if one could use his outpouring as evidence of an emotive Prophecy and Paul Kruger 205 propaganda, not unlike Vergil’s in the Aeneid, but unfortunately this just does not seem to be the case with Grendon. The last significant example of prophecy in Paul Kruger’s Dream appears in Part 25, when Paul Kruger recounts the visit the Boers were paid at Paardekraal (1880) by the shades of illustrious dead Boers who appeared after the vitriolic monologue of ‘Truth’. These shades appear in a similar fashion to those that appear to Odysseus in the Odyssey Book 11,20 and to Aeneas in Book 6 of the Aeneid:21

The gates of Hades burst, and now the shades Of hero kinsmen, long departed, rise In stature full, and awful, to forewarn Their heedless living kinsmen of their doom. All mad confusion into calm decline. (25.15–19)

While one may expect the appearance of Paul Kruger’s deceased countrymen to bring support and encouragement to the Boers, the shades in fact do the exact opposite. Their message is not particularly different from that of ‘Truth’. The shades admonish both the theft of land by the Dutch in southern Africa 247 years prior, and their abominable treatment of the native black people. The first governor of the Dutch Cape Colony, Simon van der Stel, warns his kinsmen of their impending doom for their wicked ways (25.59–85) and the Boer leader Piet Retief, murdered by the Zulu king Dingane, cautions the Boers that their scorning of God will ultimately lead to their demise (25.86–102). Father and son Piet and Dirk Uys, who were killed during the Battle of Italeni against the Zulus during the Great Trek, advise the Boers that they will have fought the Zulus in vain if they do not repent of their poor treatment of native South Africans (25.130–71). Hendrik Potgieter, referred to as the ‘Prince of the illustrious dead’, after bemoan- ing his loss of Natal, also warns the Boers of their ruin after their defeat at the Battle of Boomplaats (25.175–254).22 The final shade to speak to the assembled Boers is Johannes Brand, the politician and lawyer who became the fourth president of the Orange Free State. Of all the shades to speak, his words are particularly harsh and cutting, calling the Boer leaders Kruger, Steyn, Leyds and Reitz Judases who would ‘hurry the Boers to their bitter defeat’ (25.332). He calls his people a corrupt, greedy and an accursed race. With these climactic words, the shades disappear and the pattern of ignoring prophecies continues as Paul Kruger and the Boers somehow reject the warnings of both Truth and the shades of their forebears, and present the British with an ultimatum for war on 9 October 1899.

Prophecies realized From Part 27 onwards, we see a rapid acceleration of the plot as all the prophecies given to the Boers and Paul Kruger over the preceding twenty-seven parts are realized. We witness through Paul Kruger’s narrative, the Siege (and relief) 206 Szerdi Nagy of Ladysmith,23 the Boers being driven out of Natal, the war turning against the Boers and Paul Kruger beseeching the personification of Britannia to be kind to the Boers. We see the death of General Joubert of peritonitis in Pretoria, and Paul Kruger’s exhortation to the Boers to continue fighting despite this loss. As the war continues to turn against the Boers, we see the transportation of Boer families from their homeland and, in Part 34, we see Paul Kruger himself fleeing on the Dutch warship (the Gelderland) when he realizes that the war is lost. We witness a significant change in Paul Kruger at this point. Gone is the arrogant and greedy leader; in his place, we get to see a tired old man lamenting his terrible misfor- tunes and the waste of his life. The epic continues to take a melancholic tone in Part 35 with the death of Queen Victoria in 1901 and in the penultimate part, we see Paul Kruger despairing in his acknowledgement that he has turned away from God and embraced Mammon. In the final part, we see the close of Paul Kruger’s life, the death of Cecil John Rhodes and the restoration of peace to South Africa. On his deathbed, Grendon has Paul Kruger repent for all his sins and return to God, praising God for returning peace to South Africa. There is an air of optimism at the close of the epic with the hope of a lasting peace between Boer and Briton. It is Grendon’s own hope that we see here. A hope that this peace would usher in an era of peace and equality for all of South Africa’s non-whites who, he believed, would fare better under British rule than they had under the Boers. Grendon constructed the early parts of the narrative in a way which seems to suggest that the story will follow the pattern of the Trojans – the parallels with the Trojans being forced to flee Troy after its fall to the Greeks and the Boer mass migration of the Great Trek after the British come into power at the Cape Colony is conspicuous. Aeneas’ journey with his people to Italy to found a ‘new Troy’ is notably similar to the Boers’ search for a new homeland away from the British. Having studied Vergil, and the earlier books of the Aeneid in particular,24 it would make sense that Grendon would rely heavily on Vergil as a model for his own epic. The Sibyl-like figure of Fortuna and the shades that visit Paul Kruger, reminiscent of Book 6 of the Aeneid, would then seem to lend weight and credibility to Grendon’s epic through his careful use of easily identifiable Classical precursors. This makes it seem rather peculiar that he swerves sharply away from this expected plot line development midway through his epic, choosing to end instead with the Boers losing their homelands and coming out as a people loathed by God, which is quite the opposite of the Trojans’ situation at the end of the Aeneid. Grendon perhaps set up the narrative to deliberately mislead his audience for at the end of his epic, one can draw nothing but negativity from his depiction of the Boers and Paul Kruger. It seems, then, that Grendon appropriates familiar figures and images from classical antiquity to lend essential weight to his attempt at writing a full-scale national epic, but once the narrative develops, he seems to discard these in favour of a more Christian-centred didacticism and a deep-seated dislike for the Boers which he fails to hide. Whether this evolution was deliberate, or merely evidence of a naïve and inexperienced writer, is not immediately clear. At the end of his epic, we find that – if we continue to use Grendon’s alignment of the Boers with the Trojans and the Greeks with the British – the ‘Greeks’ are Prophecy and Paul Kruger 207 victorious over the ‘Trojans’, a clear and complete turnaround from the original plot line: where one felt sympathy for Vergil’s defeated Trojans, in Grendon’s text, the reader rejoices in their counterparts’ defeat. Instead of the reader’s sympathy lying where one might expect, with the Boer/Trojans (as we are led to feel by Vergil), we instead find ourselves rejoicing at the victory of the British/Greeks. It is not immediately clear why Grendon has done this. It may not be fair to say that this construction is used only to illustrate his classical background, and that Grendon is less concerned with the accuracy of his allusions than the overall feel of their ‘presence’, but in many ways this does seem to be the case. Of all the classical allusions and devices Grendon attempts to employ within his epic, none is as developed as that of the prophet and the role of prophecy. Grendon’s attempt to create an identifiably South African epic is grounded in this foundation of his (somewhat limited) classical education and it is also this education that led him, however inaccurately, to believe that once the Boers were overthrown by the British, the non-white people of South Africa would be free. While there is a note of optimism at the close of Paul Kruger’s Dream, one cannot help but feel disappointed for Grendon, knowing that everything he hoped would come from the end of the Boer nations, the freedoms he believed that he and other non-white South Africans would enjoy, would never really materialize as he and many others so dearly hoped. One wonders whether, if he had received some prophecy that showed him the reality of the British victory over the Boers, how they chose to govern the Union, the regimes that would emerge in both 1910 and 1948 and what this would mean for non-whites in South Africa, he would have believed it to be a false prophecy as Paul Kruger had so many times in his poem.

Notes 1 The South African school leaving qualification. 2 Though now considered offensive by some, the term ‘non-white’ encapsulates all black, coloured, Indian and mixed-race people. Grendon felt that he himself fell within this category. This term was used before Apartheid and at this stage did not have the deroga- tory connotations that were subsequently connected with it. 3 ‘Attempted’ here is the operative word since his work had very little impact at the time of its composition and was largely forgotten until uncovered by Grant Christison in 2007. 4 Christison (2007) 210. 5 Previously known as the Second Anglo-Boer War, it was fought primarily between the Boer Republics and the British Empire. The war began in 1899 and ended in 1902 with the defeat of the Boers and the signing of the Peace of Vereeniging on 31 May 1902. 6 Christison (2007) 405. 7 Christison (2007) 415. 8 Christison (2007) 416, quoting from the Zonnebloem College Magazine (December 1903) 4. 9 Lambert (2011) offers a thorough exploration of the multiple identities and voices that existed in South Africa and, to some extent, still do. 10 Christison (2007) 420. 11 Christison (2007) 419. 12 Guérard (1940) 231. 208 Szerdi Nagy 13 Aune (1983) 56–66. 14 A more thorough analysis of the role Grendon assigns to Fortuna in the epic may be found in Nagy (2012). 15 The full title is Paul Kruger’s Dream: The Struggle for Supremacy in South Africa Between Boer and Briton, or the Overthrow of ‘Corruption,’ ‘Falsehood,’ ‘Tyranny,’ ‘Wrong,’ and the Triumph of ‘Justice,’ ‘Truth,’ ‘Liberty,’ ‘Right’. 16 He also predicts the decline of ‘Boerdom’ nineteen years after his appearance. 17 ‘The coming year to thee, and to thy land. Will bring forth woe on woe. Three deadly plagues’ (13.114–15). According to Christison (2007), these three plagues are: (1) Terrible Railway Accident (31 December 1895) in Natal, near Glencoe Junction. The passengers on that ill-fated train, mostly women and children, were fugitives from threatens of trouble in Johannesburg. Forty were killed and many seriously wounded. (2) Dynamite Explosion (February 1896) at Vrededorp, a suburb of Johannesburg. The earth was torn thirty feet deep and 200 feet across, every structure levelled to the ground. Loss of life was terrible and enormous. (3) Rinderpest, which decimated the cattle of the state. 18 Uitlander means ‘foreigner’ in Afrikaans. This was the term used for the influx of migrant workers to the Boer republics after the discovery of gold in 1886. The problem lay in the fact that many of them were British, and after a time they grew to outnumber the Boers and were thus seen as a threat to Boer sovereignty. 19 A failed raid carried out against the Boers by the British statesman Leander Jameson. 20 In Book 11 of the Odyssey we see the hero sail to Oceanus with his men on the instructions of Circe, and there, after making the appropriate sacrifices, he is visited by many shades including the ill-fated Elpenor (Od. 11.51–89), Odysseus’ mother, Anticlea (Od. 11.150–224) and the heroes Agamemnon (Od. 11.385–464), Achilles (Od. 11.465– 540) and Ajax (Od. 11.541–92), to name a few, with whom we can see parallels in Grendon’s episode. The figure of Tiresias in the Odyssey (11.90–149) perhaps influenced Grendon’s portrayal of ‘Truth’. 21 Aeneas, accompanied by the Sibyl, also encounters a number of shades in the Underworld, including his helmsman Palinurus (Aen. 6.337–83), his ill-fated lover Dido (Aen. 6.440–76), his cousin Deiphobus (Aen. 6.477–534) and the future Alban kings (Aen. 6.752–76), Romulus, Caesar, Augustus and Marcellus (Aen. 6.777–885). His father Anchises fulfils the main prophetic role which Tiresias performs in the Odyssey and ‘Truth’ in Paul Kruger’s Dream. 22 TI would like to thank the reviewer, who suggested a correlation between this quote and Livy (24.7.1), referring to Hieronymus king of Syracuse in the Second Punic War. There are a number of parallels in Livy’s War with Hannibal which will be the subject of a future paper. 23 Christison (2007: 402) believes that there is a strong possibility that Grendon himself was present at some stage during the siege. 24 This can be asserted with confidence since I have made a close examination of the Latin textbooks he would have used. 15 Cassandra prophesies back Marion Zimmer Bradley’s Firebrand

Elke Steinmeyer

In the 1980s, two women writers published within a few years of each other revisionist retellings of the myth of the Trojan War and its aftermath from the perspective of Cassandra, the princess of Troy and priestess of Apollo, in order to ‘recuperate a lost female voice from a classic text’.1 In 1983, Christa Wolf, a writer from East Germany (the German Democratic Republic), published her Erzählung Kassandra, using a first-person narrative, whose English translation appeared one year later in 1984. In 1987, the American popular fiction writer Marion Zimmer Bradley published her fantasy novel The Firebrand (translated into German as Die Feuer von Troia in 1988), a third-person narrative, making Cassandra the protagonist. In the following, I shall try to establish the theory underlying both adaptations, but I shall focus mainly on Bradley’s interpretation and shall only occasionally make reference to Wolf’s novel, which has been extensively treated by modern scholarship. I shall explore the intertextual relationships between Firebrand and the ancient sources, Firebrand and Bradley’s previous works, and briefly those between Firebrand and Christa Wolf’s novel. I shall conclude with a short consideration of the title of Bradley’s book and its genre. Christa Wolf’s Kassandra has been classified by Hutcheon as ‘historiographic metafiction’.2 Most of Bradley’s works have generally been subsumed under the umbrella term ‘fantasy’. Like science fiction, ‘fantasy literature has proven tre- mendously difficult to pin down’.3 For Edward James and Farah Mendelsohn, the main difference between these two genres lies in the fact that ‘fantasy is about the construction of the impossible whereas science fiction may be about the unlikely, but is grounded in the scientifically possible’.4 According to this definition, Firebrand is not a simple fantasy novel, since it includes elements from the supernatural, such as visions, and ends on the utopian note that there is a remote possibility that one day matriarchy might survive or be revived, at least in some isolated areas. However, Bradley’s Firebrand has been labelled not only as ‘fantasy’. There are a multitude of subgenres whose characteristics fit or partially fit the novel, such as ‘historical fantasy’, a mixture of fantasy and historic fiction with fantastic elements (magic for instance), or ‘classical fantasy’ with feminist undertones, and also ‘revisionist history’.5 Bradley’s novel deals with an ancient myth and not with historical events, however; therefore, I propose the 210 Elke Steinmeyer term ‘revisionist mythography’, which revises the myth by rewriting its masculine gendered premises. The myth of the Trojan War and its countless subplots is probably one of the most popular subjects in ancient Greek and Roman literature, and has been told by a multitude of ancient authors in various literary genres. All these ancient sources, however, share three main problematic features:

1 The fact that all the authors are men impacts indirectly on their interpretations of the myth and makes them biased, deliberately or inadvertently. Because of the male bias, they depict the narratives and characters – both male and female – through a male lens and not necessarily as a woman would do. 2 Most of the ancient sources focus on the male characters in the myth and only to a lesser extent on the female characters. 3 Most of the ancient sources tell the myth from a Greek perspective and much less through the eyes of the Trojans.

The first and second points – the depiction of women in ‘male-authored literary texts’ – have been identified by Classics scholars.6 In Spoken Like a Woman, Laura McClure observes that ‘literary representations of women may be viewed as male constructs appropriated by men for the purpose of speaking about male concerns rather than as simple reflections of social reality’.7 Nancy Rabinowitz goes a step further when she remarks that ‘the study of classics has not only shaped texts and constituted canons, it has also engendered generations of students and scholars, instilling the masculine views of antiquity. I say masculine advisedly since not only were the authors of epic, tragedy, and philosophy men, but the citizenry of antiquity (even of supposedly democratic Athens) was also male’.8 Rabinowitz’s observations are supported by the historical situation and the subordinate position of women, especially in ancient Athens. McClure’s and Rabinowitz’s viewpoints can be expanded and integrated into a broader concept developed by feminist and film studies scholar Laura Mulvey, who in 1975 coined the term ‘male gaze’. She defines this as follows: ‘the determining male gaze projects its fantasy onto the female figure, which is styled accordingly’.9 Stereotyped female characters are depicted through the lenses of stereotypical heterosexual male agents in order to conform to a stereotyped ‘horizon of expectation’. The active male gaze denies women their own agency and relegates them to the status of passive objects.10 These concerns could provide a possible explanation for a recent trend in literature and theatre: the desire to fill a gap, to ask or answer questions which have not been asked or answered before and to provide alternative solutions by giving a voice to the marginalized or silenced characters of Greek and Roman mythology in order to add another perspective, another viewpoint to the existing canon. Some examples are Katherine Beutner’s novel Alcestis (2010), which retells the traditional myth from Alcestis’ perspective; Adèle Geras’ novel Troy (2000), which narrates the myth of the Trojan War from the perspective of two female servants in the royal palace in Troy; and a theatre production entitled Ismene, Schwester Cassandra prophesies back 211 von (original title: Zus van) by the Dutch theatre producer Lot Vekemans,11 which can be considered a sequel of sorts to Sophocles’ Antigone, imagining the destiny of the survivors (Ismene and Creon) in the form of a monologue presented by Ismene, interspersed with flashbacks from the past. Postmodern and feminist critics have commented on the idea of shifting the focus from the traditionally accepted centre to the neglected margins. One characteristic of the complex notion of postmodernism is, to quote Linda Hutcheon, that ‘The centre no longer completely holds. And, from the decentred perspective, the “marginal” and what I will be calling […] the “ex-centric” […] takes on new significance’.12 Hutcheon later identifies feminism as one of the ‘ex-centrics’.13 In this context, one can observe that the rewriting of ancient myths from a feminist perspective has appealed particularly to female writers choosing a secondary, or invented, female character as their protagonist. In this way, they take the concept of mythography or mythopoiesis, which has been practised since antiquity, to the next level by adding a female authorial voice to the exclusively male-authored canon of ancient texts, and by shifting the focus from the traditional protagonists to the marginalized ones, to ‘the other’. This adds a different dimension which cannot be found in the ancient primary sources and opens up a broad spectrum of new insights into well-known stories. One of the special merits of Bradley’s novel lies in the fact that she rewrites the well-known story of the Trojan War and its aftermath from the viewpoint of the victims, thereby ‘writing back’ with a clear feminist agenda. In her Acknowledgements, she writes:

Readers will be likely to bring challenges; ‘That’s not the way it happened in the Iliad’. Of course not; had I been content with the account in the Iliad, there would have been no reason to write a novel. Besides, the Iliad stops short just at the most interesting point, leaving the writer to conjecture about the end from assorted legends and traditions. If the writers of Greek drama felt free to improvise, I need not apologize for following their excellent example.14

Although Cassandra is not a protagonist or main character in the ancient sources,15 she features in a considerable number of texts after Homer’s Iliad.16 Probably the most important and most influential are the two scenes in tragedy: Aeschylus’ Agamemnon (1072–1330) and Euripides’ The Trojan Women (308–461). Bradley refers to them in her Postscript.17 Lycophron’s Alexandra is the first ancient text which explicitly mentions the rape of Cassandra by the Lesser Ajax in the temple of Athena in Troy, while earlier sources considered Ajax’s violent behaviour towards Cassandra (dragging her forcibly away from the statue of the goddess to which she was clinging) sufficient to provoke Athena’s wrath against the Greeks. Bradley made use of all of these texts, plus some relatively unknown ones, in creating the plot of her novel, which will be discussed below. Bradley’s Firebrand starts with a prologue featuring Cassandra as an old woman and grandmother, surrounded by a large group of women and children in an unnamed location and a simple dwelling. She is depicted as a figure of authority in the household, but also practising traditional women’s tasks such as spinning. 212 Elke Steinmeyer She forbids a passing bard to sing about the Trojan War, since his story is full of lies, while she, as an eyewitness, knows what really happened. Her grandchildren do not know her Trojan past and so together with the bard they all urge her to tell them her story so that the truth can be preserved and passed on. Cassandra refuses at first, but, ultimately, the story is told as a flashback in a third-person narrative in the novel, which is further subdivided into three books and an epilogue. In almost all the ancient sources, Cassandra is killed together with Agamemnon by Clytemnestra upon their arrival in Mycenae. Only Dares Phrygius, a Latin writer from the fifth century ad, at the time of the so-called Third Sophistic, changes her destiny in his Acta diurna belli Troiani (43): Antenor, a counsellor of the Trojan king Priam, but also a close friend of the Greek army leaders, arranges for Cassandra, her twin brother Helenos, her mother Hecabe and her sister-in-law Andromache to be released after the fall of Troy, in order to flee to the Chersonese.18 Bradley uses this very obscure version of the myth, whose purpose will be discussed later. The first part of Bradley’s novel describes in great detail Cassandra’s childhood among the other members of the royal family in Troy. Polyxena is Hecabe’s and Priam’s first-born daughter here, not the youngest, as in the ancient myth, and an extremely vain girl. Hecabe is a sister of the Amazon queen Penthesilea and was brought up as an Amazon, until she decided to marry King Priam, moved to the city of Troy and gave up her previous lifestyle among free women in favour of a much more restricted and traditional life controlled by men. The permanent conflict between the old ways of matriarchy and the new system of patriarchy is one of the central themes of Bradley’s novel. Under matriarchy, queens ruled over cities, women were free to choose their lovers and to practise polyandry, and children belonged to their mothers and not to their fathers, while men had a subordinate position. Patriarchy was imported by the male tribes of the north, who brought iron with them, a set of new rules and a whole pantheon of new gods, and demanded rulership over the cities and the obedience of women. Bradley uses the myth of the Trojan War in order to introduce a second war: her novel features not only the conflict between the Greek and Trojan armies and heroes, but also between the old female civilization and religion, fighting for their survival, and the new male counterpart, which gradually vanquishes the former. I will return to this point later. In all the ancient sources, Cassandra has a twin brother called Helenos. Bradley changes this and makes Paris Cassandra’s twin brother. In the ancient myth, it was Cassandra who interpreted the dream Hecabe has just before giving birth and in which she sees Paris as a flame bringing the downfall of Troy;19 in Bradley’s novel, it is a priestess in Troy who provides the interpretation of the dream. As in the ancient myth, Paris is raised as a shepherd on Mount Ida and returns to Troy only as a young man. Despite their being twins, Paris nurses a strange deep dislike of Cassandra, almost an aversion, and even attacks her physically on one occasion, trying to strangle her.20 In the ancient myth, it is Cassandra who is appre- hensive about Paris and foresees the danger which he will bring to Troy, warning everybody about him in vain. Bradley might have wanted to create a counter- scenario in which Paris’ hatred of his sister adds another negative dimension to his arrogant and vain personality. Cassandra prophesies back 213 Cassandra has her first vision at the age of six, when she and her mother are on their way to the temple of Apollo. In the sunlight, Cassandra sees a fleet of foreign ships in Troy’s harbour and is very confused, because her mother cannot see them.21 During their visit to the temple, Cassandra hears the voice of the god, who asks her whether she wants to become his priestess and claims that she will belong to him, when that day comes.22 She also feels instinctively attracted to the snakes which live in a basket next to the altar and plays with them.23 From this point on, Cassandra will develop a special relationship with snakes throughout the novel, and will even become an expert on snakes. This scene has an intertextual link to a fragment by the Hellenistic historian Anticleides (Fr. 17),24 in which he relates that both Helenos and Cassandra were left overnight in Apollo’s temple and that the next morning, they were found with snakes licking their ears, through which both received the gift of prophecy. Bradley here goes decidedly against the mainstream version of the myth, in which Cassandra received the gift of prophecy from Apollo in exchange for sexual favours but, after her refusal of his advances, he cursed her never to be believed although she always spoke the truth.25 Bradley will provide another reason for Apollo’s curse later. It is interesting to observe Hecabe’s reaction when Cassandra tells her that Apollo has asked her to become his priestess and allowed her to touch his snakes. Long before Apollo’s curse, her mother reacts very harshly, scolding Cassandra for lying and for making up the story. She even threatens to beat her if she ever repeated it.26 This is the first of countless instances in the novel when people think she is mad and react to Cassandra’s prophecies with disbelief, irritation, threats and even physical violence. These reactions are reminiscent of the scene in Euripides’ The Trojan Women in which neither Hecabe nor the women of the chorus understand Cassandra’s ecstasy about her enforced marriage to Agamemnon (341–50) and where all attempts by Cassandra to comfort them through her prophecies about their own and the Greeks’ future have the opposite effect: Hecabe and the chorus feel even worse, because they think that Cassandra is mocking them and their suffering (406–7), which in turn increases Cassandra’s helplessness and desperation (458–61). In the novel, some years after the incident in Apollo’s temple, during a feast Cassandra asks her parents about the identity of the boy she had seen before in a bowl of water, not knowing that he is her twin brother Paris and that she is touching upon a very sensitive issue. Cassandra does not receive an answer, but a slap in the face from Priam, who insists that she be raised somewhere else.27 Her mother arranges for her to live for many years with the Amazons, as she herself did in her childhood and youth. During her teenage years, Cassandra is trained as a warrior and learns about the Amazons’ way of life. Bradley also provides an unusual interpretation of the Centaurs, depicting them as small misshapen men with brown skin and long hair and beards who ride on wild horses so that it could seem from a distance as if their bodies and the bodies of their horses had fused into one.28 Cassandra travels with the Amazons to Colchis, where Queen Imandra, also an Amazon and cousin of Hecabe and Penthesilea, reigns over the city in the old matriarchal tradition. During her time in Colchis, Cassandra is initiated as a 214 Elke Steinmeyer priestess of the Goddess, the old mother goddess of the times before the male gods, and dedicates her life to her.29 She further develops her special relationship with snakes, which are also associated with the Goddess and not only with Apollo, and receives one as a farewell gift from Queen Imandra. Cassandra loves living with the Amazons and does not want to return to Troy, but is told that she cannot be an Amazon for the rest of her life. Therefore, she goes back to Troy with Andromache, Queen Imandra’s daughter, who does not want to live like an Amazon, but is happy to marry Hector. She will remain close friends with Cassandra until the end. Bradley offers an alternative reason for Apollo’s curse, very different from the ancient sources. In the same way as patriarchy and matriarchy are in conflict throughout the narrative, Apollo combats the Goddess. Cassandra is caught in this eternal struggle, since both of the divinities claim ownership over her. She constantly feels that she is betraying the one when she serves the other. She has been initiated as a priestess of the Goddess in Colchis, but was called by Apollo himself to serve in his temple upon her return to Troy, where she was in charge of the god’s snakes. Apollo has reminded her that she belongs to him.30 Suddenly, Aphrodite also begins to claim her service,31 but this time, Cassandra vehemently refuses since, in her opinion, Aphrodite is not a manifestation of the Goddess, but an independent goddess whom she cannot respect. The battle between Apollo and the Goddess escalates when Cassandra fights off a priest who tries to rape her, dis- guised as the god, and both claim that they protected her, because she was chosen by each of them.32 As a result, Apollo tells Cassandra that he loved her, but that she has betrayed him by serving his old enemy, the Goddess. He will not release her, but takes away her gift of prophecy.33 Cassandra weeps because she has lost Apollo’s love, but is nevertheless relieved to be freed from the gift of prophecy. It turns out, however, that Apollo has his revenge on her in the same way as in the Greek myth: nobody will believe her,34 and he will not forgive her as long as she lives.35 The Trojan War comes into the narrative only halfway through the novel. At this point, too, Bradley has made some changes. The Greeks in general are depicted as much more primitive, rude and cruel than in the Homeric epic. Their camp is filthy and stinks,36 and the deplorable hygienic conditions could easily provide a rational explanation for the outbreak of the plague.37 In Bradley’s narrative, Odysseus has for years been a close friend of the Trojan royal family, before being forced to join the Greek army and fight against his former hosts. He attempts to double-cross both sides, and also tries to comfort Cassandra after the fall of Troy, telling her that he will take good care of her mother Hecabe, and that she need not worry. He wants to save Cassandra too, but Agamemnon insists on his claim to her.38 Hector knows immediately that it is Patroclus in Achilles’ armour and not Achilles himself, when he kills him.39 From the beginning, Achilles is presented as being severely mentally disturbed, with sadistic traits, but becomes completely insane after Patroclus’ death. As in Homer’s Iliad, he kills Hector and mutilates his corpse. Nevertheless, Briseis, who is a servant of Hecabe in Troy, becomes infatuated with him and goes voluntarily to the Greek camp.40 Bradley also includes elements of later versions of the myth, in which the Amazons join the Trojan army after Hector’s death, as depicted in the Cassandra prophesies back 215 Posthomerica of Quintus Smyrnaeus, a Greek author of the third or fourth century ad, although she has them arrive just before Hector’s death. This gives Bradley the opportunity to bring the Amazons back into the narrative and to illustrate their extinction. Penthesilea arrives in Troy with only two dozen Amazons – their number has gradually diminished over the years and many have opted for a life in the cities under male rule. When the Amazons attack him, Achilles and his men kill all of them, which signals the end of the old matriarchal order. Bradley chooses an obscure version of Achilles’ reaction to Penthesilea’s death: instead of falling in love with the dead Amazon queen, he strips her of her armour and, in an act of necrophilia, violates her corpse.41 Only the twelfth-century Byzantine scholar Eustathius of Thessalonike mentions this act in his commentaries on Homer’s epics, but it fits Bradley’s earlier characterization of Achilles as a killing machine who runs amok very well. There is no place for a romantic love story or a remorseful Achilles in Bradley’s narrative. She wants to depict the Greeks as cruel and brutish in opposition to Homer’s idealized depiction. Achilles’ perverted and hideous crime is the first of many unspeakable cruelties the Greeks will commit during the conquest of Troy. Cassandra avenges Penthesilea’s death by taking her aunt’s bow and arrows and shooting Achilles in the heel.42 These special arrows, made by the Centaurs, are drenched with poison made from the skin of a toad,43 which causes a slow and painful death. Since Cassandra is wearing a ceremonial dress and a mask of the god Apollo, people think that is was the Sun God himself who killed Achilles. The simultaneous extinction of the Centaurs and the Amazons foreshadows the end of the old world. Bradley provides a new version of the fall of Troy, combining a different myth and a scientific explanation. The Greeks do indeed build a giant wooden horse in honour of the god Poseidon, but this has nothing to do with the conquest of the city. It is Poseidon himself, whom Homer already calls the “earth-shaker” in the Iliad and Odyssey, who destroys the impenetrable walls of Troy with an earthquake, allowing the Greek army to invade the city. Apollo, the protector of Troy, tries to fight off Poseidon in a long duel, but ultimately loses.44 The reason for Poseidon’s anger is Paris’ betrayal of his first wife, Oenone, and their son, whom he married while still living on Mount Ida, but whom he abandoned for the sake of Helen. Oenone is the daughter and priestess of the river god Scamander, another manifestation of Poseidon,45 who then punishes the whole city for Paris’ disrespect and makes Hacabe’s nightmare come true: that Paris has brought with him the downfall of Troy. In Bradley’s narrative, Cassandra is raped by the Lesser Ajax in Athena’s temple during the conquest of Troy. As mentioned above, the Hellenistic writer Lycophron is the first to mention this crime explicitly (357–64), which from then on became part of the myth. In Firebrand, Ajax is depicted as a brutal, under- hand and disgusting person who rapes Cassandra although (or maybe because) he knows that Agamemnon wants her for himself. Agamemnon is very upset when he accuses Ajax: ‘you knew she fell to me in the casting of lots, but you had to try and get ahead of me. You always were spiteful, Ajax’.46 Agamemnon nevertheless takes Cassandra back to Mycenae with him. During the sea passage, Agamemnon 216 Elke Steinmeyer tries genuinely to please Cassandra, to gain her confidence and promises to make her his ‘lawful consort’ and not treat her as a slave.47 The return takes such a long time that Cassandra falls pregnant and gives birth to a baby boy, whom she names Agathon. Agamemnon wants Agathon and not Orestes, his son by his wife Clytemnestra, to one day rule over Mycenae.48 This is reminiscent of a passage in Pausanias’ Graeciae Descriptio (2.16.6) which describes several graves in Mycenae, among them the graves of Agamemnon, Cassandra and their twin sons Teledamus and Pelops, who were all killed by Aegisthus.49 Bradley depicts Mycenae, like Colchis, as a centre of matriarchy. Clytemnestra, dressed in Minoan fashion,50 welcomes Agamemnon and Cassandra politely and invites Agamemnon into the palace. Aegisthus kills the fleeing Agamemnon with an axe,51 but it is clear that Aegisthus is in a subordinate position to Clytemnestra, since he hands the axe to her after the murder. Clytemnestra does not kill either Cassandra or her child, but sets them free and even invites them to stay with her in Mycenae as the Lady’s priestess (another name for the old mother Goddess), since she wants to restore the ‘old ways’52 – an offer which Cassandra declines because she does not entirely trust Clytemnestra. One of the reasons for Clytemnestra’s decision is that Helen told her that Cassandra was her only friend in Troy.53 Ultimately, both women put the Great Goddess above their rivalry and show each other respect. Cassandra decides to go to Colchis to Queen Imandra together with Zakynthia, a female servant who requests to accompany her, since she has been called by a god to go there.54 The Epilogue disproportionately condenses a lot of events into a few pages so that one has the impression that Bradley has already said what she wants to say and that the Epilogue is only an afterthought used to tie up loose ends. After a long and wearisome journey, Cassandra and Zakynthia arrive in Colchis. Cassandra has observed on the way that Zakynthia is in fact a cross-dressing man.55 Queen Imandra, who has in the meantime had a daughter, now seven years old, who will rule after her,56 is only too glad to see Cassandra and suggests that she stay. Cassandra learns the fates of some of the other characters: her own mother died in Troy from a broken heart (and does not transform into a black howling dog, as in the standard version of the myth), Odysseus has been lost for three years and Aeneas has founded a new city. According to Imandra, Clytemnestra has died in childbirth and Aegisthus has been killed by Orestes, who is now ruling over Mycenae. By letting Clytemnestra die naturally, Bradley can bypass the difficult topic of the matricide, but she also shows that Clytemnestra’s vision of reviving matriarchy and the ‘old ways’ was short-lived. It seems that Colchis is the only remaining stronghold of matriarchy in the ancient world. Then Zakynthia reveals herself as a man, Zakynthos, and claims to have been called by the Goddess to found a new city together with an Amazon or a woman with the spirit of the Amazons, ‘a world where Earth Mother will be worshipped in the old ways’. He also claims that he has dreamed ‘of a city where women are not slaves, and where men need not spend their lifetimes in war and fighting’.57 Cassandra agrees to go with him and to help him found this city, ‘to build a world better than Troy’ where the truth about Troy and the Trojan War can be preserved.58 Cassandra prophesies back 217 The Epilogue links the narrative back to the Prologue. Is the anonymous place the city which Cassandra has set off to found? Only women and children are mentioned – are there no men in this place? It seems to be a rather shabby and run-down place – is this the city which Zakynthos and Cassandra dreamed of? Cassandra’s family does not know the truth about the Trojan War, which means that her hope to build a better place which preserved the story has not been fulfilled. Cassandra, who says in the Epilogue that she can no longer spend her life behind walls, seems to live as a rather ordinary woman performing traditional women’s tasks.59 It becomes clear why Bradley decided to follow Dares Phrygius and why Cassandra needed to survive: to offer a glimpse of hope that there might be a new era and a new beginning for a lost religion and civilization. Bradley was a well-known author long before the publication of Firebrand, through her earlier works, particularly the Darkover series and The Mists of Avalon (1983), probably her most famous novel, a fantasy feminist retelling of the Arthurian legend from the perspective of Morgan le Fay. One can find similar questions and situations in The Mists of Avalon and Firebrand. The struggle between the old female religion and the new male religion, represented respectively in Firebrand by the Goddess and Apollo, is mirrored in Mists of Avalon in Morgan le Fay, the representative of the ‘old ways’ of matriarchy and worship of the Mother Goddess, and King Arthur, the future founder of Camelot who brings with him the new male-dominated Christian religion. As in Firebrand, the old civilizations perish and are replaced by the new male-dominated ways of life: the old pagan world is gradually overpowered by Christianity and disappears into the mists in Mists of Avalon. In Firebrand, there is at least the utopian prospect that the old ways might survive in Colchis and in the new city to be founded by Cassandra and Zakynthos. Although I have not found any information about whether Bradley had read the English translation of Christa Wolf’s novel Kassandra (1984) before writing her own (1987), there are too many correspondences between the two novels to consider them coincidences. They allow me to exercise what Michael Riffaterre calls ‘aleatory intertextuality’:

In recent articles Riffaterre has made clear that we must distinguish between aleatory intertextuality (which is not unlike Barthes’ notion of ‘circular memory’ and which allows the reader to read a text through the prism of all and any familiar texts) and obligatory intertextuality which demands that the reader take account of a hypogrammatic origin.60

Probably the strongest intertextual link is the romance between Cassandra and Aeneas, which is not mentioned in any of the ancient sources and which must have been an invention by Christa Wolf herself which Bradley has taken over. In Bradley’s novel, Aeneas escapes from Troy before the Greeks’ invasion (in the German translation, he takes Cassandra’s adopted daughter Honey with him); in Wolf’s text, he escapes alone. In both novels, Cassandra refuses to go with him, albeit for different reasons. In Bradley, his wife Creusa and his two daughters 218 Elke Steinmeyer have already left Troy long before; in Wolf, Creusa does not feature. Wolf and Bradley share a strong interest in matriarchy and the Minoan and Mycenaean civilizations, which Wolf explicitly states in one of her four essays on the novel.61 In Wolf’s novel, there is a hidden parallel world exclusively for women, in which women can act freely, without male constraints, and worship the old mother goddess Cybele. Also in Wolf’s novel, the Amazons join the Trojan War and are van- quished in its course. Like Bradley, Wolf makes use of the obscure version of the myth in which Achilles violates the corpse of the dead Amazon queen Penthesilea after he has killed her in battle, and she also includes the rape of Cassandra by the Lesser Ajax. Interestingly, Wolf also hints at the possibility of a friendship between Cassandra and Clytemnestra in other times and other circumstances – an idea which Bradley develops in greater detail to the extent that Clytemnestra spares Cassandra’s life and the life of her baby son (by Agamemnon) and sets them free. Of particular interest is the title of Bradley’s novel. The term ‘firebrand’ evokes in a non-anglophone speaker the association of a ‘blaze’ or ‘conflagration’. The translators of the German version of the novel must have had the same misleading idea, since the official German title is Die Feuer von Troia (The Fires of Troy). The term firebrand denotes first ‘a person who is passionate about a particular cause, typically inciting change and taking radical action’ and second ‘a piece of burning wood’.62 One might think immediately of Hecabe’s dream before Paris’ birth, in which Paris is a burning torch who sets Troy alight. Bradley allows for different interpretations. Cassandra champions the female and matriarchal cause, which fits the first definition. But the comparison with a torch can also be found. There is a passage in which Cassandra wants to warn the inhabitants of Troy about an impending earthquake: ‘It was as if she were filled with fire; she was driven, burning with the heat of the warning that flamed and raged within her. She fled down the streets, shrieking her warning over and over again’.63 One is reminded of the scene in Euripides’ The Trojan Women (308–463) where Cassandra storms onstage with a torch as part of the traditional marriage ceremony. In this passage from Bradley, both definitions of the word ‘firebrand’ are perhaps alluded to, depicting Cassandra as an eternal flame or reminder of painful truth. Through the ancient figure of the mythological prophetess Cassandra, Marion Zimmer Bradley offers her reader a prophetic vision of an alternative, matriarchal world in her novel Firebrand. It is up to the reader to decide whether Cassandra’s prophecy has been realized or whether Apollo’s curse still prevails.

Notes 1 Miller (2010) 29. Two other famous examples utilizing the same ‘narrative strategy’ (ibidem) are Margaret Atwood’s The Penelopiad (2005) and Ursula Le Guin’s Lavinia (2008). I would like to thank the anonymous readers for their valuable comments and suggestions. 2 Hutcheon (1988) 95. 3 James and Mendelsohn (2012) 1. 4 James and Mendelsohn (2012) 1. Cassandra prophesies back 219 5 https://en.wikipedia.org/wiki/Historical_fantasy 6 McClure (1999) 37. 7 McClure (1999) 5. 8 Rabinowitz (1993) 4. 9 Mulvey (2009) 19. 10 This phenomenon seems to be a global one and is not restricted to antiquity. Parallels are to be found in contemporary literature worldwide and even in South African fiction. For instance, African American and Maori women writers have taken up the idea of ‘talking back’, ‘answering back’ or ‘back chat’. See Smith (2006) 18. Indian women writers in South Africa in turn have faced an even more complex situation because of the racial politics of the Apartheid regime which was abolished only in 1994. Govinden (2008: 3) characterizes their particular position as follows: ‘it is arguable that Indian women’s writings in South Africa have suffered marginalization not only because of gender discrimination, but also because they have been seen as part of a minority culture’. But she also establishes that: ‘one may discover that Indian women’s writings in South Africa also provide a useful lens with which to read and re-read the important issues of our time. As with those of other black women writers in South Africa who were largely excluded and “othered”, autobiographical, fictional and discursive writing produced by Indian women in South Africa constitutes a way of “talking back”’ (2008: 2–3). Govinden is quoting Smith and Watson (2006); Smith (1993) 20. 11 The play premiered in 2005 in Haarlem in the Netherlands and is part of the current repertoire of the Deutsche Theater in Berlin, where its first German production also took place in 2010. 12 Hutcheon (1988) 12. 13 Hutcheon (2002) 69. 14 Bradley (1987 & 2002) 1. 15 Although the title of Lycophron’s work is Alexandra, Cassandra’s other name, she is more of an overarching link than a protagonist. 16 For a complete list of the ancient Greek and Latin sources, see Neblung (1997) 252–255. 17 Bradley (1987 & 2002) 402. 18 See also Neblung (1997) 201. 19 Bradley (1987 & 2002) 9–12. 20 Bradley (1987 & 2002) 275. 21 Bradley (1987 & 2002) 17. 22 Bradley (1987 & 2002) 18. 23 Bradley (1987 & 2002) 18. 24 140 FGrH Fr. 17. See also Neblung (1997) 105. 25 See especially Aeschylus, Agamemnon 1202–1213. 26 Bradley (1987 & 2002) 19–20. 27 Bradley (1987 & 2002) 25–6. 28 Bradley (1987 & 2002) 34. 29 Bradley (1987 & 2002) ch. 13. 30 Bradley (1987 & 2002) 110. 31 Bradley (1987 & 2002) 135. 32 Bradley (1987 & 2002) 146. 33 Bradley (1987 & 2002) 149. 34 Bradley (1987 & 2002) 151, 154. 35 Bradley (1987 & 2002) 4. 36 Bradley (1987 & 2002) 234, 307. 37 Bradley (1987 & 2002) 250. 38 Bradley (1987 & 2002) 376. 39 Bradley (1987 & 2002) 293. 40 Bradley (1987 & 2002) 199–201. 41 Bradley (1987 & 2002) 348. 220 Elke Steinmeyer 42 Bradley (1987 & 2002) 349–350. 43 Bradley (1987 & 2002) 306. 44 Bradley (1987 & 2002) 366. 45 Bradley (1987 & 2002) 361. 46 Bradley (1987 & 2002) 374. The English original of Firebrand also tells that another Geek soldier rapes Cassandra’s adopted daughter, Honey, whom she found as a baby on her way back from Colchis to Troy, and kills her afterwards (Bradley 1987: 372, 374). The German translation has changed the story and some passages in the text. Here Cassandra asks Aeneas to take Honey with him when he prepares to leave Troy just before the invasion of the Greek army and therefore, Honey, who is still a toddler, escapes together with Aeneas into safety. I think that Bradley added this new element in order to further depict the Greeks as inhumane and perverted sadists. In the same way as Achilles defiled Penthesilea’s corpse, the Greek soldier states explicitly his abnormal sexual desire for small children. It was probably too graphic and too offensive for a German readership. 47 Bradley (1987 & 2002) 380, 382–3. 48 Bradley (1987 & 2002) 384, 387. 49 See also Neblung (1997) 191. 50 Bradley (1987 & 2002) 392. 51 Bradley (1987 & 2002) 393. 52 ibidem. 53 ibidem. 54 Bradley (1987 & 2002) 394. 55 Bradley (1987 & 2002) 395–6. 56 Bradley (1987 & 2002) 399. 57 Bradley (1987 & 2002) 400. 58 Bradley (1987 & 2002) 400–401. 59 Bradley (1987 & 2002) 399. 60 Worton and Still (1990) 26. This is the case for Wolf, who explicitly informs the reader about her sources. 61 Wolf (1983). 62 http://www.google.co.za/#hl=en&gs_rn=9&gs_ri=psyab&cp=15&gs_id=1g&xhr=t& q=firebrand+definition&es_nrs=true&pf=p&sclient=psyab&oq=%22firebrand+defi &gs_l=&pbx=1&bav=on.2,or.r_qf.&bvm=bv.45512109,d.d2k&fp=728da8b4b02cd2ff &biw=960&bih=465 63 Bradley (1987 & 2002) 275. Bibliography

Any abbreviations employed here are commonly found elsewhere and therefore should not be a cause of any confusion.

Adams, J.N. 2003. Bilingualism and the Latin Language. Cambridge. Addey, C. 2013. ‘Ecstasy between Divine and Human: Re-assessing Agency in Iamblichean Divination and Theurgy.’ In J.F. Finamore and J. Philips (eds.), Literary, Philosophical and Religious Studies in the Platonic Tradition: Proceedings of the Seventh Annual International Society for Neoplatonic Studies Conference. 7–24. Bonn. Addey, C. 2014a. Divination and Theurgy in Neoplatonism: Oracles of the Gods. Farnham. Addey, C. 2014b. ‘The Daimonion of Socrates: Daimones and Divination in Neoplatonism.’ In H. Tarrant and D. Layne (eds.), The Neoplatonic Socrates: Essays in Late Antiquity and the Reception of Socrates. 51–72. University Park, PA. Addey, C. 2017. ‘Plato’s Women Readers.’ In H. Tarrant, D. Layne, D. Baltzly and F. Renaud (eds.), Brill’s Companion to the Reception of Plato in Antiquity. Leiden. Aicher, P.J. 1995. Guide to the Aqueducts of Ancient Rome. Wauconda. Alföldi, A. 1972. ‘Redeunt Saturnia regna, III: Iuppiter-Apollo und Veiovis.’ Chiron 2 215–30. Bonn. Amandry, P. 1950. La Mantique apollinienne à Delphes: essai sur la fonctionnement de l’Oracle. Paris. Ammianus Marcellinus. 1939–1940. Rerum gestarum libri. Ed. J.C. Rolfe. Volumes 2–3. Cambridge, MA. Anderson, B. 1983. Imagined Communities: Reflections on the Origin and Spread of Nationalism. London. Anon. 1730. False Religion worse than No Religion: An Enquiry concerning Superstition, as it affects the Rights and Happiness of Civil Society: Written for the Advancement of True Religion and Virtue. London. Apuleius. 1991. De Deo Socratis. Ed. C. Moreschini. Apulei Platonici Madaurensis opera quae supersunt. Volume 3. De philosophia libri. Leipzig. Aristotle. 1955. On the Pythagoreans. Ed. W.D. Ross. Aristotelis Fragmenta Selecta. Oxford. Arnold, T.K. 1839. Henry’s first Latin book. London. Ashby, T. 1935. The Aqueducts of Ancient Rome. Oxford. Asheri, D., A. Lloyd and A. Corcella. 2007. A Commentary on Herodotus Books I–IV. Ed. O. Murray and A. Moreno. Oxford. Astin, A.E. 1964. ‘Leges Aelia et Fufia.’ Latomus 23 421–45. Athanassiadi, P. 1991. ‘The Fate of Oracles in Late Antiquity: Didyma and Delphi.’ Deltion Christianikes Archaiologikes Etaireias 15 [1989–1990] 27–8. 222 Bibliography Athanassiadi, P. 1993. ‘Persecution and Response in Late Paganism: The Evidence of Damascius.’ JHS 113 1–29. Athanassiadi, P. 2013. ‘The Divine Man of Late Hellenism: A Sociable and Popular Figure.’ In M. Dzielska and K. Twardowska (eds.), Divine Men and Women in the History and Society of Late Hellenism. 13–27. Kraków. Audollent, A. 1904. Defixionum Tabellae. Paris. Aune, D.E. 1983. Prophecy in Early Christianity and the Ancient Mediterranean World. Michigan. Aupert, P. and D.R. Jordan. 1981. ‘Magical Inscriptions on Talc Tablets from Amathous.’ AJA 85 184 (Abstract). Bagnall, R.S., N. Aravecchia, R. Cribiore, P. Davoli, O.E. Kaper and S. McFadden. 2016. An Oasis City. New York. Banchich, T.M. 1993. ‘Julian’s School Laws: Cod. Theod. 13.3.5 and Ep. 42.’ AW 24 5–14. Barker, E. 2006. ‘Paging the Oracle: Interpretation, Identity and Performance in Herodotus’ History.’ G&R 53 1–28. Bates, O. 1914. The Eastern Libyans: An Essay. London. Bateson, G. 1973. Steps to an Ecology of Mind: Collected Essays in Anthropology, Psychiatry, Evolution and Epistemology. St Albans. Beard, M. 1986. ‘Cicero and Divination: The Formation of a Latin Discourse.’ JRS 76 33–46. Beard, M. 2013. ‘Risk in the Humanities. Alea iacta est.’ In L. Skinns, M. Scott and T. Cox (eds.), Risk, 85–108. Cambridge. Beard, M., J. North and S. Price. 1998. Religions of Rome. Cambridge. Beck, H. 2015. ‘Beyond “Foreign Clienteles” and “Foreign Clans”: Some Remarks on the Intermarriage between Roman and Italian Elites.’ In M. Jehne and F. Pina Polo (eds.), Foreign Clientelae in the Roman Empire: A Reconsideration. 57–72. Stuttgart. Beck, R. 2006. ‘The Religious Market of the Roman Empire: Rodney Stark and Christianity’s Pagan Competition.’ In L.E. Vaage (ed.), Religious Rivalries in the Early Roman Empire and the Rise of Christianity. 233–2. Waterloo. Beiner, R. 2011. Civil Religion: A Dialogue in the History of Political Philosophy. Cambridge. Belayche, N. 2007. ‘Les dieux “nomothètes”. Oracles et prescriptions religieuses à l’époque romaine impériale.’ Revue de l’histoire des religions 224.2 171–91. Bellemore, J. 1989. ‘When Did Valerius Maximus Write the Dicta et Facta Memorabilia.’ Antichthon 23 67–80. Bénabou, M. (1975). La Résistance africaine à la Romanisation. Paris. Bendlin, A. 2000. ‘Looking beyond the Civic Compromise: Religious Pluralism in Late Republican Rome.’ In E. Bispham and C. Smith (eds.), Religion in Archaic and Republican Rome and Italy. Evidence and Experience. 115–35, 167–71. Edinburgh. Bendlin, A. 2011. ‘On the Uses and Disadvantages of Divination: Oracles and their Literary Representations in the Time of the Second Sophistic.’ In J. North and S. Price (eds.), The Religious History of the Roman Empire. 175–250. Oxford. Bentley, R. 1713. Remarks Upon a Late Discourse of Free-thinking. London. Bergemann, J. 1990. Römische Reiterstatuen: Ehrendenkmäler im öffentlichen Bereich. Mainz. Berger, P.L. and T. Luckmann. 1966. The Social Construction of Reality: A Treatise in the Sociology of Knowledge. New York. Berman, D. 1975. ‘Anthony Collins and the Question of Atheism in the Early Part of the Eighteenth Century.’ Proceedings of the Royal Irish Academy 75 85–102. Bibliography 223 Berman, D. 1980. ‘Hume and Collins on Miracles.’ Hume Studies 6 150–4. Berman D.W. 2015. Myth, Literature, and the Creation of the Topography of Thebes. Cambridge. Bernand, A. 1991. Sorciers grecs. Paris. Bernand, A. 2000. ‘La Voix des exclus.’ In A. Moreau and J.-C. Turpin (eds.), La Magie. 4 Volumes. 2.7–18. Montpellier. Bers, V. 2009. Genos Dikanikon. Amateur and Professional Speech in the Courtrooms of Classical Athens. Cambridge, MA. Berthelet, Y. 2011. ‘Le Rôle des pontifes dans l’expiation des prodiges à Rome, sous la République: le cas des “procurations” anonymes’ Cahiers “Mondes anciens” 2, Journées doctorales ANHIMA 2008 et 2009 1–13. Betz, H.D. (ed.). 1992. The Greek Magical Papyri in Translation including the Demotic Spells. 2nd ed. Chicago, IL. Bianchi, E. 2010. Il Rex Sacrorum a Roma e nell’Italia antica. Milan. Bingham-Kolenkow, A. 2002. ‘Persons of Power and Their Communities.’ In L. Ciraolo and J. Seidel (eds.), Magic and Divination in the Ancient World. 133–44. Leiden. Birley, A. 2007. ‘Sejanus: His Fall.’ In N. Sekunda (ed.), Corolla Cosmo Rodewald. 121– 50. Gdańsk. Bloomer, W.M. 1992. Valerius Maximus and the Rhetoric of the New Nobility. Chapel Hill, NC. Blount, C. 1683. Miracles, No Violations of the Laws of Nature. London. Boissevain, U. P. (ed.). 1955. Cassii Dionis Cocceiani historiarum Romanarum quae supersunt. Vol. 3. Berlin. Edited and translated by E. Cary. 1914–1927. Dio’s Roman History VIII. New York. Bonnechere, P. 2003. Trophonios de Lébadée: cultes et mythes d’une cité béotienne au miroir de la mentalité antique. Leiden. Bonnechere, P. 2007. ‘Divination.’ In D. Ogden (ed.), A Companion to Greek Religion. 145–60. Oxford. Bosman, P.R. 2016. ‘The Dodona Bronze Revisited.’ AClass 59 182–94. Bosworth, A.B. 1988. Conquest and Empire. Cambridge. Bouché Leclerq, A. 1882. Histoire de la divination dans l’antiquité. 4 Volumes. Paris. Bourguet, E. 1914. Les Ruines de Delphes. Paris. Bowden, H. 2005. Classical Athens and the Delphic Oracle: Divination and Democracy. Cambridge. Bowden, H. 2013. ‘Seeking Certainty and Claiming Authority: The Consultation of Greek Oracles from the Classical to the Roman Imperial Periods.’ In V. Rosenberger (ed.), Divination in the Ancient World: Religious Options and the Individual. 41–60. Stuttgart. Bowersock, G.W. 1994. Fiction as History: Nero to Julian. Berkeley, CA. Bowra, C.M. 1959. ‘ΕΙΠΑΤΕ ΤΩΙ ΒΑΣΙΛΗΙ.’ Hermes 87 426–35. Bradley, M.Z. 1988. Die Feuer von Troia. Trans. Manfred Ohl and Hans Sartorius. Frankfurt. Bradley, M.Z. 2002. The Firebrand. EBook Design Group digital back-up edition v1 HTML November 27, 2002. (First published: New York: Simon and Schuster 1987). Brandl, L. 1975. ‘Early Christianity in Africa: North Africa, the Sahara, the Sudan, Central and East Africa: A Contribution to Ethnohistory.’ Présence Africaine NS. 96 467–95. Bremmer, J.N. 2002. The Rise and Fall of the Afterlife. London. Brett, M. 1979. ‘The Arab Conquest and the Rise of Islam in North Africa.’ In J.D. Fage and R. Oliver (eds.), The Cambridge History of Africa Volume 2: From c. 500 BC to AD 1050. 490–555. Cambridge. Brett, M. and E. Fentress. 1996. The Berbers. Oxford. 224 Bibliography Briscoe, J. 1993. ‘Some Notes on Valerius Maximus.’ Sileno 19 395–408. Briscoe, J. 1998. Valeri Maximi Facta et Dicta Memorabilia. 2 Volumes. Stuttgart. Briscoe, J. 2008. A Commentary on Livy. Books 38–40. Oxford. Briscoe, J. 2010. ‘Review of Andrea Themann-Steinke, Valerius Maximus: ein Kommentar zum zweiten Buch der Facta et Dicta memorabilia.’ ExClass 14 379–82. Briscoe, J. 2012. A Commentary on Livy. Books 41–45. Oxford. Brogan, O. and D.J. Smith. 1984. Ghirza: A Libyan Settlement in the Roman Period. Tripoli. Brooke, C. 2012. Philosophic Pride: Stoicism and Political Thought from Lipsius to Rousseau. Princeton, NJ. Broughton T.R.S. 1951–1952. The Magistrates of the Roman Republic. 2 Volumes. New York. Brown, P. 1971. ‘The Rise and Function of the Holy Man in Late Antiquity.’ JRS 61 99–100. Brown, T.S. 1978. ‘Aristodicus of Cyme and the Branchidae.’ AJP 99 64–78. Browne, G. (ed.). 1983. Sortes Astrampsychi, volumen I. Ecdosis prior. Leipzig. Bruner, J. 1991. ‘The Narrative Construction of Reality.’ Critical Inquiry 18 1–21. Brunt, P.A. 1965. ‘The Aims of Alexander.’ G&R 12 205–15. Buck, D.F. 1992. ‘Eunapius’ Lives of the Sophists: A Literary Study.’ Byzantion 62 141–57. Buckler, J. 1989. Philip II and the Sacred War. Leiden. Burck, E. 1979. ‘Die Johannis des Corippus.’ In E. Burck (ed.), Das römische Epos. 379–99. Darmstadt. Burkert, W. 1985. Greek Religion. Archaic and Classical. Oxford. Burkert, W. 1987. Ancient Mystery Cults. Cambridge, MA. Burkert, W. 1991. Greek Religion: Archaic and Classical. Trans. J. Raffan. Hoboken, NJ. Burkert, W. 1997. The Orientalizing Revolution: Near Eastern Influence on Greek Culture in the Early Archaic Age. Trans. M.E. Pinder and W. Burket. Cambridge, MA. Burkert, W. 2005. ‘Signs, Commands, and Knowledge: Ancient Divination between Enigma and Epiphany.’ In S.I. Johnston and P.T. Struck (eds.), Mantikê: Studies in Ancient Divination. 29–50. Leiden. Burr, V. 2003. Social Constructionism. 2nd ed. London. Cadotte, A. 2007. La romanisation des dieux. L’interpretatio romana en Afrique du Nord sous le Haut-Empire. Leiden. Cameron, A. 1980. ‘The Career of Corippus Again.’ CQ 30 534–9. Cameron, A. 1982. ‘Byzantine Africa. The Literary Evidence.’ In J. Humphrey (ed.), Excavations at Carthage VII. 29–62. Ann Arbor, MI. Cameron, A. 1984. ‘Corippus’ Iohannis: Epic of Byzantine Africa.’ Proceedings of the Liverpool Latin Seminar 4 167–80. Cameron, A. 1985. Procopius and the Sixth Century. Berkeley, CA. Cameron, A. 2001. ‘Wandering Poets: A Literary Movement in Byzantine Egypt.’ In G. Nagy (ed.), Greek Literature in the Byzantine Period: Greek Literature. 14–55. New York. Cameron, E. 2010. Enchanted Europe: Superstition, Reason, and Religion, 1250–1750. Oxford. Camous, Th. 2004. Le Roi et le fleuve. Ancus Marcius Rex aux origines de la puissance romaine. Paris. Camps, G. 1986. ‘Funerary Monuments with Attached Chapels from the Northern Sahara.’ The African Archaeological Review 4 151–64. Camps, G. 1990. ‘Qui sont les Dii mauri ?’ Antiquités africaines 26 131–53. Camps, G. 2002. ‘Liste onomastique libyque. Nouvelle édition.’ Antiquités africaines 38–9 211–57. Bibliography 225 Canter, H.V. 1922. ‘“Fortuna” in Latin Poetry.’ Studies in Philology 19 64–82. Carastro, M. 2003. ‘Quelle altérité pour les Grecs? Les katάdesmoi et l’invention de la notion de magie.’ Monitor ISH 5 1–14. Carastro, M. 2006. La Cité des mages: penser la magie en Grèce ancienne. Grenoble. Cardauns, B. 1976. Varro. Antiquitates Rerum Divinarum. Mainz. Carey, C. 1994. ‘Legal Space in Classical Athens.’ G&R 41 172–86. Carter, C.J. 1975. ‘Valerius Maximus.’ In T. A. Dorey (ed.), Empire and Aftermath: Silver Latin II 26–56. London. Castrucci, G. 2012. ‘Dodona versus Delphi in Greek Tragedy: The Wanderings of the Hero between Expiation and Ties of γενος.’ Logeion 2 1–25. Cawkwell, G. 2005. The Greek Wars: The Failure of Persia. Oxford. Champion, J. 1992. The Pillars of Priestcraft Shaken: The Church of England and its Enemies, 1660–1730. Cambridge. Champion, J. 2003. Republican Learning: John Toland and the Crisis of Christian Culture, 1696–1722. Manchester. Champlin, E. 2011. ‘Tiberius and the Heavenly Twins.’ JRS 101 73–99. Chaniotis, A. 2010. ‘Megatheism: The Search for the Almighty God and the Competition of Cults.’ In S. Mitchell and P. van Nuffelen (eds.). 2010. One God: Pagan Monotheism in the Roman Empire. 112–40. Cambridge. Chantraine, P. 1968–1980. Dictionnaire étymologique de la langue grecque: histoire des mots. 4 Volumes. Paris. Christ, M. 2013. ‘Herodotean Kings and Historical Enquiry.’ In R. Munson (ed.), Herodotus. Vol. 1: Herodotus and the Narrative of the Past. 212–50. Oxford. Christidis, A.-P., S. Dakaris and I. Vokotopoulou. 1999. ‘Magic in the oracular tablets from Dodona.’ In D.R. Jordan, H. Montgomerymand and E. Thomassen (eds.), The World of Ancient Magic. 67–72. Bergen. Christison, G. 2007. African Jerusalem: The Vision of Robert Grendon. PhD Dissertation. University of KwaZulu Natal. Christison, G. 2012. ‘Grendon, Robert.’ In E.K. Akyeampong and H.L. Gates (eds.), Dictionary of African Biography. 6 Volumes. 2.503–6. Oxford. Ciholas, P. 2003. The Omphalos and the Cross: Pagans and Christians in Search of a Divine Center. Macon. Clark, E. 1998. ‘The Lady Vanishes: Dilemmas of the Feminist Historian after the “Linguistic Turn.”’ Church History 67 1–31. Clark, G. 1993. Women in Late Antiquity: Pagan and Christian Lifestyles. Oxford. Clay, J.S. 2009. ‘The Silence of the Pythia.’ In L. Athanassaki, R.P. Martin and J.F. Miller (eds.), Apolline Politics and Poetics. 5–15. Athens. Clinton, K. 1974. The Sacred Officials of the Eleusinian Mysteries. Philadelphia, PA. Coarelli, F. 1985. Il Foro romano II. Periodo repubblicano e augusteo. Rome. Coarelli, F. 1988. Il campo Marzio. Dalle origine alla fine della repubblica. Rome. Cobet, J. 1986. ‘Herodotus and Thucydides on War.’ In I.S. Moxon, J.D. Smart and A.J. Woodman (eds.), Past Perspectives: Studies in Greek and Roman Historical Writing. 1–18. Cambridge. Cohen, D. 1995. Law, Violence and Community in Classical Athens. Cambridge. Colonna, A. (ed.). 1938. Heliodorus: Aethiopica. Rome. Collins, A. 1713. A Discourse of Free-Thinking, Occasion’d by The Rise and Growth of a Sect call’d Free-Thinkers. London. Collins, D. 2001. ‘Theoris of Lemnos and the Criminalization of Magic in Fourth-Century Athens.’ CQ 51 477–93. 226 Bibliography Collins, D. 2008a. Magic in the Ancient Greek World. Oxford. Collins, D. 2008b. ‘Mapping the Entrails: The Practice of Greek Hepatoscopy.’ AJPh 129 319–45. Combès, R. 1995. Valère Maxime: Faits et dits mémorables. Volume I. Paris. Constantelos, D.J. 1964. ‘Paganism and the State in the Age of Justinian.’ The Catholic Historical Review 50 372–80. Cornell, T.J. 2013. The Fragments of the Roman Historians. 3 Volumes. Oxford. Courby, M.F. 1927. Fouilles de Delphes: Tome II: Topographie et architecture: La ter- rasse du temple. Paris. Couzens, T. ‘Sol T. Plaatje and the First South African Epic.’ English in Africa 14 41–65. Couzens, T. 1988. ‘Robert Grendon: Irish Traders, Cricket Scores and Paul Kruger’s Dreams.’ English in Africa 15 49–91. Cox, P. 1983. Biography in Late Antiquity: A Quest For the Holy Man. Berkeley, CA. Cox Miller, P. 2000. ‘Strategies of Representation in Collective Biography.’ In T. Hägg and P. Rousseau (eds.), Greek Biography and Panegyric in Late Antiquity. 209–54. Berkeley, CA. Cramer, F.H. 1954. Astrology in Roman Law and Politics. Philadelphia. Crawford, M.H. 1974. Roman Republican Coinage. 2 Volumes. Cambridge. Croke, B. and J. Harries. 1982. Religious Conflict in Fourth-century Rome: A Documentary Study. Sydney. Crosby, D.J. 2015. ‘“Arrows Fletched from Our Own Wings”: The Early Church Fathers and the “Delphi of the Mind.”’ In W.M. Johnston and D.J. Crosby (eds.), A Dangerous Mind: In Celebration of the Ideas of Delbert L. Wiens. 150–63. Eugene, OR. Csordas, T.J. 1990. ‘Embodiment as a Paradigm for Anthropology.’ Ethos 18 5–47. Cudworth, R. 1678. The True Intellectual System of the Universe: The First Part; wherein, All the Reason and Philosophy of Atheism is Confuted; and Its Impossibility Demonstrated. London. Cumont, F. 1912. Astrology and Religion among the Greeks and Romans. London Curbera, J.B. 2012. ‘The Curse Tablets of Richard Wünsch Today.’ In M. Piranomonte and F. Marco Simón (eds.), Contesti Magici – Contextos Mágicos. 193–4. Rome. Curbera, J.B. 2015. ‘From the Magician’s Workshop: Notes on the Materiality of Greek Curse Tablets.’ In D. Boschung and J. Bremmer (eds.), The Materiality of Magic. 97–122. Cologne. Dakaris, S., A.-Ph. Christides and I. Vokotopoulou. 1993. ‘Les Lamelles oraculaires de Dodona et les villes de l’Épire dans l’antiquité.’ In P. Cabanes (ed.), L’Illyrie méridion- ale et l’Épire dans l’antiquité. 2. 55–60. Paris. Dakaris, S. 2000. Dodona. Athens. Dakaris, S., I. Vokotopoulou and A.-Ph. Christides. 2013. Ta christeria elesmata tis Dodonis ton anaskafon D. Evangelidi. 2 Volumes. Athens. Dakoronia, F. 1993. ‘Homeric Towns in East Lokris.’ Hesperia 62 115–27. D’Amato, S. and C. Letta 1975. Epigrafia della regione dei Marsi. Milan. Dascalakis, A.P. 1966. Alexander the Great and Hellenism. Thessalonica. David, J.-M. 1995. ‘Le Tribunal du préteur: contraints symboliques et politiques sous la République et le début de l’Empire.’ Klio 77 371–85. Davies, J. (ed.). 1718. M. Tullii Ciceronis de Natura Deorum libri tres. Cambridge. Davies, J. (ed.). 1721. M. Tullii Ciceronis libri de Divinatione et de Fato. Cambridge. Davies, J.P. 2004. Rome’s Religious History: Livy, Tacitus and Ammianus on their Gods. Cambridge. De Boer, J.Z., J.R. Hale and J. Chanton. 2001. ‘New Evidence for the Geological Origins of the Ancient Delphic Oracle (Greece).’ Geology 29 707–11. Bibliography 227 Delatte, A. (ed.). 1927. Anecdota Atheniensia. Vol. 1. Liège. Dempsey, T. 1918. The Delphic Oracle: Its Early History, Influence, and Fall. Oxford. Denti, M. 1991. ‘Il Marsia di Paestum.’ AIONArch 13 133–88. Denzey Lewis, N. 2014. ‘Living Images of the Divine: Female Theurgists in Late Antiquity.’ In K.B. Stratton and D.S. Kalleres (eds.), Daughters of Hecate: Women and Magic in the Ancient World. 274–97. Oxford. Deutsch, M.E. 1928. ‘I am Caesar, not Rex.’ CP 23 394–98. Dick, B.F. 1963. ‘The Technique of Prophecy in Lucan.’ TAPA 94 37–49. Dick, B.F. 1965. ‘The Role of the Oracle in Lucan’s de Bello Civili.’ Hermes 93 460–6. Dickie, M.W. 2001. Magic and Magicians in the Greco-Roman World. London. Diels, H. and W. Kranz. 1951. Die Fragmente der Vorsokratiker griechisch und deutsch. Berlin. Dieterle, M. 2007. Dodona. Religionsgeschichtliche und historische Untersuchungen zur Entstehung und Entwicklung des Zeus-Heiligtums. Hildesheim. Dillon, J. 1996. The Middle Platonists, 80 B.C. to A.D. 220. Rev.ed. Ithaca, NY. Dixon, S. 2001. Reading Roman Women: Sources, Genres and Real Life. London. Dodds, E.R. 1951. The Greeks and the Irrational. Berkeley, CA. Dodds, E.R. 1965. Pagan and Christian in an Age of Anxiety. Cambridge. Dodds, E.R. 1973. The Ancient Concept of Progress and other Essays on Greek Literature and Belief. Oxford. Downey, G. 1957. ‘The Emperor Julian and the Schools.’ CJ 53 97–103. Drijvers, J.W. 1996. ‘Ammianus Marcellinus 15.13.1–2: Some Observations on the Career and Bilingualism of Strategius Musonianus.’ CQ 46 532–7. East, K.A. 2014. ‘Superstitionis Malleus: John Toland, Cicero, and the War on Priestcraft in Early Enlightenment England.’ History of European Ideas 40 965–83. East, K.A. 2016. ‘Cicero the Pantheist: A Radical Reading of Ciceronian Scepticism in John Toland’s Pantheisticon (1720).’ Intellectual History Review 26 245–62. Edlund-Berry, I. 2006. ‘Hot, Cold, or Smelly: The Power of Sacred Water in Roman Religion, 400–100 BCE.’ In Schultz, C.E. and P.B. Harvey (eds.), Religion in Republican Italy. Yale Classical Studies 33 162–80. Edwards, M.J. 2000. Neoplatonic Saints: The Lives of Plotinus and Proclus by their Students. Liverpool. Edwards, M.J. 2015. Religions of the Constantinian Empire. Oxford. Ehlers, W. 1980. ‘Epische Kunst in Coripps Johannis.’ Philologus 124 109–35. Eidinow, E. 2007. Oracles, Curses and Risk among the Ancient Greeks. Oxford. Eidinow, E. 2013. ‘Oracular Consultation, Fate, and the Concept of the Individual.’ In V. Rosenberger (ed.), Divination in the Ancient World. Religious Options and the Individual. 21–39. Stuttgart. Elm, S. 2012. Sons of Hellenism, Fathers of the Church: Emperor Julian, Gregory of Nazianzus, and the Vision of Rome. Transformation of the Classical Heritage. Berkeley, CA. Engels, D. 2007. Das römische Vorzeichenwesen (753–27 v. Chr.). Quellen, Terminologie, Kommentar, historische Entwicklung. Stuttgart. Erdkamp, P. (ed.). 2007. A Companion to the Roman Army. Malden, MA. Etiope, G., G. Papatheodorou, D. Christodoulou, M. Geraga and P. Favali. 2006. ‘The Geological Links of the Ancient Delphic Oracle (Greece): A Reappraisal of the Natural Gas Occurrence and Origin.’ Geology 34 821–4. Eunapius. 1956. Vitae Sophistarum. Ed. J. Giangrande. Eunapii vitae sophistarum. 1–101. Rome. English Translation ed. W.C. Wright. 1921. Eunapius, Lives of the Philosophers and Sophists. Cambridge, MA. 228 Bibliography Evans, J.A.S. 1972. Procopius. New York. Evans, J.A.S. 1982. Herodotus. Boston. Evans, J.D. 1992. The Art of Persuasion. Political Propaganda from Aeneas to Brutus. Ann Arbor, MI. Evans, R. 2013. ‘The End of Sybaris (510 BC).’ In Fields of Death: Retracing Ancient Battlefields. 1–26. Barnsley. Evans, R. 2015. ‘New Perspectives on the War in Ionia (499–493 BC).’ In Fields of Battle: Retracing Ancient Battlefields. 1–39. Barnsley. Falconer, W.A. (trans.). 1923. Cicero: On Old Age, On Friendship, On Divination. Cambridge, MA. Faraone, C.A. 1991a. ‘Binding and Burying the Forces of Evil: The Defensive Use of “Voodoo Dolls” in Ancient Greece.’ CA 10 165–205. Faraone, C.A. 1991b. ‘The Agonistic Context of Early Greek Binding Spells.’ In C.A. Faraone and D. Obbink (eds.), Magika Hiera: Ancient Greek Magic and Religion. 1–32. Oxford. Faraone, C.A. 1999. Ancient Greek Love Magic. Cambridge, MA. Fauth, W. 1975. ‘Die Bedeutung der Nekromantie-Szene in Lucans Pharsalia’. RhM 118 325–44. Fehling, D. 1989. Herodotus and His ‘Sources’: Citation, Invention, and Narrative Art. Trans. J.G. Howie. Leeds. Fentress, E. 1981. ‘African Building, Money, Politics and Crisis in Auzia.’ In A.C. King and M. Henig (eds.), The Roman West in the Third Century. Contributions from Archaeology and History. Vol. 1. 199–210. Oxford. Finamore, J. F. 1985. Iamblichus and the Theory of the Vehicle of the Soul. Chico, CA. Finamore, J. F. 2012. ‘Iamblichus on the Grades of Virtue.’ In E. Afonasin, J. Dillon and J. F. Finamore (eds.), Iamblichus and the Foundations of Platonism. 113–32. Leiden. Finamore, J.F. and S.I. Johnston. 2010. ‘The Chaldaean Oracles.’ In L. Gerson (ed.), The Cambridge History of Philosophy in Late Antiquity. Volume 1. 161–73. Cambridge. Flacelière, R. 1965. Greek Oracles. Trans. D. Garman. New York. Flower, M.A. 2000. ‘Alexander and Panhellenism.’ In A.B. Bosworth and F.J. Baynham (eds.), Alexander the Great in Fact and Fiction. 96–135. Oxford. Flower, M.A. 2008. The Seer in Ancient Greece. Berkeley, CA. Fontenrose, J.E. 1952. ‘Review of P. Amandry. La Mantique Apollinienne à Delphes: essai sur le fonctionnement de l’oracle.’ AJP 73 445–8. Fontenrose, J.E. 1978. The Delphic Oracle: Its Responses and Operations with a Catalogue of Responses. Berkeley, CA. Forsythe, G. 2012. Time in Roman Religion. One Thousand Years of Religious History. New York. Fosl, P.S. 1994. ‘Doubt and Divinity: Cicero’s Influence of Hume’s Religious Skepticism.’ Hume Studies 20 103–20. Foster, J. and D. Lehoux. 2007. ‘The Delphic Oracle and the Ethylene-Intoxication Hypothesis.’ Clinical Toxicology 45 85–9. Fowden, G. 1977. ‘The Platonist Philosopher and his Circle in Late Antiquity.’ Philosophia 7 359–82. Fowden, G. 1982. ‘The Pagan Holy Man in Late Antique Society.’ JHS 102 33–59. Franke, P.R. 1956. ‘Das Taubenorakel zu Dodona und die Eiche des Zeus als der heilige Baum des Zeus Naois.’ Athener Mitteilungen 71 60–5. Frankfurter, D. 1998. Religion in Roman Egypt: Assimilation and Resistance. Princeton, NJ. Bibliography 229 Frankfurter, D. 2000. ‘“Things unbecoming Christians”: Violence and Christianization in Fifth-century Panopolis.’ JECS 8 273–95. Frankfurter, D. 2005. ‘Voices, Books, and Dreams: The Diversification of Divination Media in Late Antique Egypt.’ In S.I. Johnston and P.T. Struck (eds.), Mantikê. Studies in Ancient Divination. 233–52. Leiden. Fraser, P.M. and E. Matthews. 1987. Lexicon of Greek Personal Names. (5+ vols) Oxford. Fuhrmann, M. 1990. ‘Die Antike Mythen im christlich-heidnischen Weltanschauungskampf der Spätantike.’ Antike und Abendland 36 67–81. Gagé, J. 1955. Apollon romain. Rome. Gager, J.G. (ed.). 1992. Curse Tablets and Binding Spells from the Ancient World. Oxford. Gagné, R. 2009. ‘Mystery Inquisitors: Performance, Authority, and Sacrilege at Eleusis.’ CA 28 211–47. Garstad, B. 2005. ‘The “Tyche” Sacrifices in John Malalas: Virgin Sacrifice and Fourth- century Polemical History.’ Illinois Classical Studies 30 83–135. Gärtner, T. 2008. Untersuchungen zur Gestaltung und zum historischen Stoff der “Johannis” Coripps. Berlin. Gartziou-Tatti, A. ‘L’Oracle de Dodone. Mythe et rituel.’ Kernos 3 175–84. Gawlick, G. 1963. ‘Cicero and the Enlightenment.’ Studies on Voltaire and the Eighteenth Century 25 657–82. Gayet, A. 1902a. Antinoë et les sculptures des Thais et Serapion. Paris. Gayet, A. 1902b. ‘L’Exploration des nécopoles gréco-byzantines d’Antinoë et les sarcophages de tombes pharaoniques de la ville antique.’ Annales du musée Guimet 30 25–46. Gayet, A. 1903. ‘L’Exploration des nécopoles de la montagne d’Antinoë (fouilles exécu- tées en 1901–1902).’ Annales du musée Guimet 30 115–40. Gélard, M.-L. 2006. ‘Une Cuiller à pot pour demander la pluie. Analyse de rituels nord- africains contemporains.’ Journal des Africanistes 76 81–102. George, L. 2005. ‘Listening to the Voice of the Fire: Theurgical Fitness and Esoteric Sensitivity.’ In R.M. Berchman and J.F. Finamore (eds.), History of Platonism: Plato Redivivus. 287–303. New Orleans. Georges, P.B. 2000. ‘Persian Ionia under Darius: The Revolt Reconsidered.’ Historia 49 1–39. Geschke, H. 1968. ‘Die Reiterstatue der Aemilier und Marcier. Ein Beitrag zur monumen- talen Rekonstruktion der frühen Geschichte in Rom.’ JhbNumGeld 18 25–42. Ghazal, A.H. 1986. ‘New Lights on the Distinction between Ammon of Libya and Zeus of Cyrene.’ In Libya Antiqua. Part 11 of the General History of Africa: Studies and Documents. 173–7. Paris. Gillam, R. 2015. ‘The Egyptian Oracle.’ In R. Gillam and J. Jacobsen (eds.), The Egyptian Oracle Project: Ancient Ceremony in Augmented Reality. 17–107. London. Glava, Z.A. 1937. A Study of Heliodorus and his Romance, the Aethiopica. PhD Dissertation, New York University, New York. Goldie, M. 1993. ‘Priestcraft and the Birth of Whiggism.’ In N. Phillipson and Q. Skinner (eds.), Political Discourse in Early Modern Britain. 209–31. Cambridge. Govinden, D. 2008. ‘Sister Outsiders’: The Representation of Identity and Difference in Selected Writings by South African Indian Women. Pretoria. Graf, F. 1993. ‘Bois sacrés et oracles en Asie Mineure.’ In Les bois sacres. Actes du Colloque international organisé par le Centre Jean Bérard et l’École Pratique des Hautes Études. Naples. Graf, F. 1997. Magic in the Ancient World. Trans. F. Philip. Cambridge, MA. 230 Bibliography Graf, F. 1999. ‘Magic and Divination.’ In D.R. Jordan, H. Montgomery and E. Thomassen (eds.), The World of Ancient Magic. 283–98. Athens. Graf, F. 2005a. ‘Rolling the Dice for an Answer.’ In S.I. Johnston and P.T. Struck (eds.), Mantikê: Studies in Ancient Divination. 51–98. Leiden. Graf, F. 2005b. ‘Fluch und Verwünschung.’ In V. Lambrinoudakis et al. (eds.), Thesaurus Cultus et Rituum Antiquorum. Volume 3. 247–70. Los Angeles, CA. Graf, F. and S.I. Johnston. 2017. Brill’s New Pauly, s.v. Magic & Magi III.C.2. Gransden, K.W. 1973–1974. ‘Typology, Symbolism an Allegory in the Aeneid.’ Proceedings of the Virgilian Society 13 14–27. Gransden, K.W. 1976. Virgil. Aeneid Book VIII. Cambridge. Greatrex, G. 2014. ‘Perceptions of Procopius in Recent Scholarship.’ Histos 8 76–121. Greaves, A.M. 2012. ‘Divination at Archaic Branchidai-Didyma.’ Hesperia 81 177–206. Green, S.J. 2014. Disclosure and Discretion in Roman Astrology: Manilius and his Augustan Contemporaries. Oxford Greer, R.A. and M.M. Mitchell (eds.). 2007. The “Belly-Myther” of Endor: Interpretations of I Kingdoms 8 in the Early Church. Atlanta, GA. Gregory, T.E. 1983. ‘Julian and the Last Oracle at Delphi.’ GRBS 24 355–66. Grendon, R. 1904. Paul Kruger’s Dream, The Struggle for Supremacy in South Africa between Boer and Briton; or the Overthrow of ‘Corruption’, ‘Falsehood’, ‘Tyranny’, ‘Wrong’, and the Triumph of ‘Justice’, ‘Truth’, ‘Liberty’, ‘Right’: A Poem. (Acquired via personal correspondence with Grant Christison, Grendon’s biographer). Gronovius, J. (ed.). 1692. M. Tullii Ciceronis Opera quae Extant Omnia, ex Mss. Codicibus Emendata Studio atque Industria Jani Gulielmii et Jani Gruteri, Additis eorum Notis Integris. Leiden. Grottanelli, C. 2005. ‘Sorte Unica Pro Casibus Pluribus Enotata. Literary Texts and Lot Inscriptions as Sources for Ancient Kleromancy.’ In S.I. Johnston and P.T. Struck (eds.), Mantikê. Studies in Ancient Divination. 129–46. Leiden. Gruterus, J. (ed.). 1618. M. Tullii Ciceronis Opera Omnia quae Exstant. Hamburg. Guérard, A.L. 1940. Preface to World Literature. New York. Guzzo, P.G. 2015. ‘Marsia, le compedes e Charax di Pergamo.’ RM 121 115–24. Haack, M.-L. 2003. Les Haruspices dans le monde romain. Bordeaux. Hägg. T. 2012. The Art of Biography in Antiquity. Cambridge. Hale, J.R., J.Z. de Boer, J.P. Chanton and H.A. Spiller. 2003. ‘Questioning the Delphic Oracle: Overview/An Intoxicating Tale.’ Scientific American 67–73. Hammond, N.G.L. 1967. Epirus. Oxford. Hammond, N.G.L. 1980. Alexander the Great: King, Commander and Statesman. Bedminster. Hammond, N.G.L. 1998. ‘The Branchidae at Didyma and in Sogdiana.’ CQ 48 339–44. Hannoum, A. 2001. Colonial Histories, Post-colonial Memories: The Legend of the Kahina, a North African Heroine. Westport, CT. Hansen, M.H. 2006. The Shotgun Method: The Demography of the Ancient Greek City- State Culture. Columbia, MO. Hansen, W.F. (ed.). 1998. Anthology of Ancient Greek Popular Literature. Bloomington, IN. Hardy, B.C. 1968. ‘The Emperor Julian and his School Law.’ Church History 37 131–43. Harries, J. and I.N. Wood. 2010. The Theodosian Code: Studies in the Imperial Law of Late Antiquity. 2nd ed. London. Harris, J.A. 2015. Hume: An Intellectual Biography. Cambridge. Harris, W.V. 1991. Ancient Literacy. Cambridge, MA. Harrison, P. 1990. ‘Religion’ and the Religions in the English Enlightenment. Cambridge. Harrison, T. 2000. Divinity and History: The Religion of Herodotus. Oxford. Bibliography 231 Hawley, R. 1994. ‘The Problem of Women Philosophers in Ancient Greece.’ In L.J. Archer, S. Fischler and M. Wyke (eds.), Women in Ancient Societies: An Illusion of the Night. 70–87. London. Heineman, K.M. 2012. The Decline of Delphi. PhD Dissertation. The University of Newcastle. Hendrickx, B. 2013. ‘Al-Kahina: The Last Ally of the Roman-Byzantines in the Maghreb against the Muslim Arab Conquest?’ Journal of Early Christian History 3 47–61. Herbert-Brown, G. 2002. ‘Ovid and the Stellar Caldendar.’ In G. Herbert-Brown (ed.), Ovid’s Fasti: Historical Readings at its Bimillenium. 101–28. Oxford. Herman, C. 2006. Prophets and Profits: Managerialism and the Restructuring of Jewish Schools in South Africa. Cape Town. Herodotus. 1920–1981. Histories. Ed. A.D. Godley. 4 Volumes. Cambridge, MA. Hilton, J.L. 2012. ‘The Cult of Neoptolemos at Delphi in Heliodoros’ Aithiopika.’ A Class. 55 57–68. Hilton, J.L. 2017. ‘Myth and Narrative Fiction in the Works of the Emperor Julian.’ Listy filologické 140 forthcoming. Hinds, S. 1998. Allusion and Intertext: Dynamics of Appropriation in Roman Poetry. Cambridge. Hodgson, J. 1975. A History of Zonnebloem College 1858–1870: A Study of Church and Society. MA Dissertation. University of Cape Town. Hofmann, H. 1989. ‘Corippus as a Patristic Author?’ Vigiliae Christianae 42 361–77. Hölkeskamp, K.-J. 1987. Die Entstehung der Nobilität. Studien zur sozialen und politischen Geschichte der römischen Republik im 4. Jhdt. v. Chr. Stuttgart. Hölkeskamp, K.-J. 2010. Reconstructing the Roman Republic: An Ancient Political Culture and Modern Research. Rev. ed. of 2004. Rekonstruktionen einer Republik. Trans. H. Heitmann-Gordon. Princeton, NJ. Hölkeskamp, K.-J. 2014. ‘In Defense of Concepts, Categories, and Other Abstractions: Remarks on a Theory of Memory (in the Making).’ In K. Galinsky (ed.), Memoria Romana: Memory in Rome and Rome in Memory. 63–81. Ann Arbor MI. Hölkeskamp, K.-J. 2016. ‘In the Web of (Hi-)stories: Memoria, Monuments, and their Myth-Historical Interconnectedness.’ In K. Galinsky (ed.), Memory in Ancient Rome and Early Christianity. 169–213. Oxford. Holland, L.B. 1933. ‘The Mantic Mechanism at Delphi.’ AJA 37 201–14. Homer. 1931. Iliad. Ed. T.W. Allen. Oxford. Hopkins, K. 1991. ‘Conquest by Book.’ In M. Beard, A.K. Bowman and M. Corbier (eds.), Ancient Literacy in the Roman World. 133–58. Ann Arbor, MI. Hopkins, K. 1999. A World Full of Gods: Pagans, Jews and Christians in the Roman Empire. London. Hornblower, S. 2013. Commentary on Herodotus Book 5. Cambridge. Hornblower, S. 2015. Lykophron: Alexandra. Oxford. Horsley, G.H.R. and S. Mitchell. 2000. The Inscriptions of Central Pisidia. Bonn. Horster, M. 2007. ‘Living on Religion: Professionals and Personnel.’ In J. Rüpke (ed.), A Companion to Roman Religion. 331–42. Oxford. Hoskins, N.A. 2004. The Coptic Tapestry Albums and the Archaeologist of Antinoé, Albert Gayet. Seattle, WA. How, W.W. and J. Wells. 1913–1928. A Commentary on Herodotus. 2 Volumes. Oxford. Hudson, W. 2009. The English Deists: Studies in Early Enlightenment. London. Hume, D. 1788 [1741–1742]. ‘Of Superstition and Enthusiasm.’ In Essays and Treatises on Several Subjects. In Two Volumes. Vol. 1. Containing Essays, Moral, Political, and Literary. 69–74. London. 232 Bibliography Humm, M. 2014. ‘Numa and Pythagoras: The Life and Death of a Myth.’ In J.H. Richardson and F. Santangelo (eds.), The Roman Historical Tradition: Regal and Republican Rome. 35–51. Oxford. Hutcheon, L. 1988. A Poetics of Postmodernism. History, Theory, Fiction. New York. Iamblichus. 1973. Fragments. Edited and translated by J.M. Dillon. Reprinted 2009. Iamblichi Chalcidensis In Platonis Dialogos Commentariorum Fragmenta. Leiden (Reprint Westbury). Iamblichus. 1991. Iamblichus: On the Pythagorean Way of Life. Edited and translated by J. Dillon and J. Hershbell. Atlanta, GA. Iamblichus. 2002. De Anima. Ed. J.M. Dillon and J.F. Finamore. Leiden. Iamblichus. 2003. De mysteriis. Ed. E.C. Clarke, J.M. Dillon and J.P. Hershbell. Atlanta, GA. Iamblichus. 2009. Letters. Ed. J. Dillon and W. Polleichtner. Atlanta, GA. Ingold, T. 2000. The Perception of the Environment: Essays in Livelihood, Dwelling and Skill. London. Ingold, T. 2011. Being Alive: Essays on Movement, Knowledge and Description. London. Isnardi, M. 1955. ‘TEXNH e HΘΟΣ nella metodologia storiografica di Polibio.’ SCO 3 102–10. Isnardi Parente, M. and T. Dorandi (eds.). 2012. Senocrate e Ermodoro. Testimonianze e frammenti. Pisa. Israel, J. 2001. Radical Enlightenment: Philosophy and the Making of Modernity, 1650– 1750. Oxford. Itgenhorst, T. 2005. Tota illa pompa: Der Triumph in der römischen Republic. Göttingen. Jacob, M.C. 1976. The Newtonians and the English Revolution, 1689–1720. Hassocks. Jacoby, F. 1926. Die Fragmente der griechischen Historiker. Berlin. Jaeger, M. 1997. Livy’s Written Rome. Ann Arbor, MI. James, E. and F. Mendelsohn (eds.). 2012. The Cambridge Companion to Fantasy Literature. Cambridge. Jameson, M.H., D.R. Jordan and R.D. Kotansky. 1993. A Lex Sacra from Selinus. Durham, NC. Jerard, N. 1995. ‘A Discussion on the Effects of Christianity on the Western Psyche.’ Women-Church: An Australian Journal of Feminist Studies in Religion 16 16–22. Jim, S.F.T. 2011. ‘Review of A. Missiou, Literacy and Democracy in Fifth-century Athens, Cambridge.’ BMCR November. Johnson, W.A. 2009. ‘Introduction.’ In W.A. Johnson and H.N. Parker (eds.), Ancient Literacies: The Culture of Reading in Greece and Rome. 3–10. Oxford. Johnston, S.I. 1999. Restless Dead: Encounters between the Living and the Dead in Ancient Greece. Berkeley, CA. Johnston, S.I. 2005. ‘Introduction: Divining Divination.’ In S.I. Johnston and P.T. Struck (eds.), Mantikê: Studies in Ancient Divination. 1–28. Leiden. Johnston, S.I. 2008. Ancient Greek Divination. Oxford. Johnston, S.I. 2012. ‘Sosipatra and the Theurgic Life: Eunapius’ Vitae Sophistarum 6.6.5– 6.9.24.’ In J. Rüpke and W. Spickermann (eds.), Reflections on Religious Individuality: Greco-Roman and Judaeo-Christian Texts and Practices. 99–117. Berlin. Joleaud, L. 1933. ‘Gravures rupestres et rites de l’eau en Afrique du Nord.’ Journal de la Société des Africanistes 3 197–282. Jones, M. 2005. ‘Base and Heavenly Magic in Heliodoros’ Aithiopika.’ Ancient Narrative 4 79–98. Jordan, D.R. 1975. ‘A Curse Tablet from a Well in the Athenian Agora.’ Hesperia 44 245–248. Bibliography 233 Jordan, D.R. 1985a. ‘A Survey of Greek Defixiones not Included in the Special Corpora.’ GRBS 26 151–97. Jordan, D.R. 1985b. ‘Defixiones from a Well near the Southwest Corner of the Athenian Agora.’ Hesperia 54 205–55. Jordan, D.R. 1985–1988. ‘New Archaeological Evidence for the Practice of Magic in Classical Athens.’ In Praktika tou XII Diethnous Synedriou Klasikês Archaiologias 1983–4. 4 Volumes. 4 273–7. Athens, GA. Jordan, D.R. 2000. ‘New Greek Curse Tablets (1985–2000).’ GRBS 41 5–46. Jouguet, P. 1928. Alexander the Great and the Hellenistic World. Chicago, IL. Kahane, A. 1999. ‘Blood for Ghosts? Homer, Ezra Pound, and Julius Africanus.’ New Literary History 30 815–36. Kaldellis, A. 2004. Procopius of Caesarea: Tyranny, History, and Philosophy at the End of Antiquity. Philadelphia, PA. Kambitsis, S. 1976. ‘Une nouvelle tablette magique d’Égypte.’ BIFAO 76 213–23. Kapferer, B. 1997. The Feast of the Sorcerer: Practices of Consciousness and Power. Chicago, IL. Keller, M. 2002. The Hammer and the Flute: Women, Power and Spirit Possession. Baltimore, MD. Keskiaho, J. 2013. ‘Re-visiting the libri Sibyllini: Some Remarks on their Nature in Roman Legend and Experience.’ In M. Kajava (ed.), Studies in Ancient Oracles and Divination. 145–72. Rome. Kindt, J. 2006. ‘Delphic Oracle Stories and the Beginning of Historiography: Herodotus’ Croesus logos.’ CP 101 34–51. Kittelä, S.-I. 2013. ‘Dodona and Neoptolemus: Heroic Genealogies and Claims of Ethnicity.’ In Kajava, M. (ed.), Studies in Ancient Oracles and Divinatio. 29–48. Rome. Klingshirn, W.E. 2006. ‘Inventing the Sortilegus: Lot Divination and Cultural Identity in Italy, Rome, and the Provinces.’ In C.E. Schultz and P.B. Harvey (eds.), Religion in Republican Italy. 137–61. Cambridge. Kondratieff, E. 2010. ‘The Urban Praetor’s Tribunal in the Roman Republic.’ In F. de Angelis (ed.), Spaces of Justice in the Roman World. 89–126. Leiden. Koraes, A. 1804–1806. Heliodorus: Aethiopica Biblia Deka. Paris. Kotila, H. 1992. Memoria Mortuorum: Commemoration of the Departed in Augustine. Rome. Krostenko, B.A. 2000. ‘Beyond (Dis)belief: Rhetorical Form and Religious Symbol in Cicero’s De Divinatione.’ TAPA 130 353–91. Kroll, J.H. 1977. ‘An Archive of the Athenian Cavalry.’ Hesperia 46 83–140. Kuhlmann, K.P. 1988. Das Ammoneion: Archäologie, Geschichte und Kultpraxis des Orakels von Siwa. Mainz. Kuhlmann, K.P. 1998. ‘Roman and Byzantine Siwa: Developing a Latent Picture.’ In O.E. Kaper (ed.), Life on the Fringe: Living in the Southern Egyptian Deserts during the Roman and Early Byzantine Periods. 159–80. Leiden. Lambert, M. 2011. The Classics and South African Identities. London. Lambert, P.-Y. 2004. ‘Defining Magical Spells and Particularly Defixiones of Roman Antiquity: A Personal Opinion.’ In K. Brodersen and A. Kropp (eds.), Fluchtafeln. Neue Funde und neue Deutungen zum antiken Schadenzauber. 71–80. Frankfurt. Lane Fox, R. 1986. Pagans and Christians in the Mediterranean World from the Second Century AD to the Conversion of Constantine. Harmondsworth. Lang. M.L. 1976. The Athenian Agora, Volume 21: Graffiti and Dipinti. Princeton, NJ. Lang, M.L. 1982. ‘Writing and Spelling on Ostraka.’ Studies in Attic Epigraphy, History and Topography Presented to E. Vanderspoel. Hesperia Supplement 19 75–87. 234 Bibliography Langlands, R. 2011. ‘Roman Exempla and Situation Ethics: Valerius Maximus and Cicero de Officiis.’ JRS 101 100–22. Lanni, A. 2006. Law and Justice in the Courts of Classical Athens. Cambridge. Lanzi, S. 2004. ‘Sosipatra, La Teurga: Una “Holy Woman” iniziata Misteri Caldaici.’ Studi e Materiali di Storia delle Religioni 70 275–94. Larson, J.L. 2016. Understanding Greek Religion: A Cognitive Approach. London. Lee, A.D. 2016. Pagans and Christians in Late Antiquity: A Sourcebook. Oxford. Lehoux, D. 2007. ‘Drugs and the Delphic Oracle.’ CW 101 41–56. Leone, A. 2013. The End of the Pagan City: Religion, Economy and Urbanism in Late Antique North Africa. Oxford. Lepelley, C. 2010. ‘The Use of Secularized Latin Pagan Culture by Christians.’ In D.M Gwynn and S. Bangert (eds.), Religious Diversity in Late Antiquity. 477–93. Leiden. Letta, C. 1972. I Marsi e il Fucino nell’antichità. Milan. Levene, D.S. 1993. Religion in Livy. Leiden. Lewis, R. 2014. The Role of the Pythia in the Delphic Oracle: Ancient and Modern Perspectives. Unpublished MA Dissertation. University of South Africa. Lewy, H. 1956. Chaldaean Oracles and Theurgy: Mysticism, Magic and Platonism in the Later Roman Empire. Ed. M. Tardieu. Paris. (Revised edition published 1978.). Lhôte, E. 2006. Les Lamelles oraculaires de Dodone. Genève. Lieu, S.N.C. 1986. The Emperor Julian: Panegyric and Polemic. Translated Texts for Historians. Liverpool. Lillington-Martin, C. 2013. ‘Procopius on the Struggle for Dara in 530 and Rome in 537–38: Reconciling Texts and Landscapes.’ In A. Sarantis and N. Christie (eds.), War and Warfare in Late Antiquity: Current Perspectives. 599–630. Leiden. Lintott, A.W. 1999. The Constitution of the Roman Republic. Oxford. Lipsey, R. 2001. Have You Been to Delphi: Tales of the Ancient Oracle for Modern Minds. Albany, NY. Lisdorf, A. 2007. The Dissemination of Divination in Roman Republican Times: A Cognitive Approach. PhD Dissertation. University of Copenhagen. Littleton, C.S. 1986. ‘The Pneuma Enthusiastikon: On the Possibility of Hallucinogenic “Vapors” at Delphi and Dodona.’ Ethos 14 76–91. Littré, E. (ed.). 1853. Oeuvres complètes d’Hippocrate. Volume 8. Paris. MacBain, B. 1982. Prodigy and Expiation: A Study in Religion and Politics in Republican Rome. Brussels. McClure, L. 1999. Spoken Like a Woman. Speech and Gender in Athenian Drama. Princeton, NJ. MacMullen, R. 1966. Enemies of the Roman Order: Treason, Unrest and Alienation in the Empire. Cambridge, MA. MacMullen, R. 1984. Christianizing the Roman Empire A.D. 100–400. New Haven, CT. Malaise, M. 1990. ‘Bès et les croyances solaires.’ In S. Israelit-Groll (ed.), Studies in Egyptology Presented to Miriam Lichtheim. 680–729. Jerusalem. Manuwald, G. 2007. Cicero, ‘Philippics’ 3–9. Ed. with introduction, translation and commentary. Berlin. Marincola, J. 1996. Herodotus: The Histories. Trans. A. de Sélincourt. Revised with introductory matter and notes. Harmondsworth. Marincola, J. 2001. Greek Historians. Cambridge. Marinus. 2001. Proclus ou Sur le Bonheur. Ed. H.D. Saffrey and A.-P. Segonds. Paris. Marshall, E. 1998. ‘Constructing the Self and Other in Cyrenaica.’ In R. Laurence and J. Berry (eds.), Cultural Identity in the Roman Empire. 49–63. London. Bibliography 235 Martin, D.B. 2004. Inventing Superstition: From the Hippocratics to the Christians. Cambridge, MA. Martin, T. 2001. ‘The Development of Winged Angels in Early Christian Art.’ Espacio, Tiempo y Forma. Serie vii. Historia del Arte 14 11–29. Martinez, D.G. 1991. Michigan Papyri XVI: A Greek Love Charm from Egypt (P.Mich. 757). Atlanta, GA. Mattingly, D.J. 1983. ‘The Laguatan: A Libyan Tribal Confederation in the Late Roman Empire.’ Libyan Studies 14 96–108. Mattingly, D.J. 1994. Tripolitania. Ann Arbor, MI. Mattingly, D.J. 2008. ‘Laguatan.’ In S. Chaker and Y. Modéran (eds.), Encyclopédie ber- bère, 28–29: Kirtēsii–Lutte. 4314–18. Aix-en-Provence. Maurizio, L. 1995. ‘Anthropology and Spirit Possession: A Reconsideration of the Pythia’s Role at Delphi.’ JHS 69–86. Maurizio, L. 1997. ‘Delphic Oracles as Oral Performances: Authenticity and Historical Evidence.’ ClAnt 16 308–34. Maurizio, L. 2013. ‘Interpretative Strategies for Delphic Oracles and Kledons: Prophecy Falsification and Individualism.’ In V. Rosenberger (ed.), Divination in the Ancient World: Religious Options and the Individual. 61–79. Stuttgart. Mazurek, T. 2004. ‘The decemviri sacris faciundis: Supplication and Prediction.’ In C. F. Konrad (ed.), Augusto augurio. Rerum humanarum et divinarum commentationes in honorem Jerzy Linderski. 151–68. Stuttgart Meeks, D. 1992. ‘Le Nom du dieu Bès et ses implications mythologiques.’ In U. Luft (ed.), The Intellectual Heritage of Egypt: Studies Kákosy. 423–36. Budapest. Menn, S. 2013. ‘Plato’s Soteriology?’ In V. Adluri (ed.), Philosophy and Salvation in Greek Religion. 191–216. Berlin. Merleau-Ponty, M. 1962. The Phenomenology of Perception. Trans. C. Smith. London. Merrills, A. (ed.). 2004. Vandals, Romans and Berbers. New Perspectives on Late Antique North Africa. Aldershot.Merrills, A. and R. Miles. 2014. The Vandals. Malden. Meyer, E.A. 2013. The Inscriptions of Dodona and a New History of Molossia. Stuttgart. Miller, T.S. 2010. ‘Myth-Remaking in the Shadow of Vergil: The Captivel(ated) Voice of Ursula K. Le Guin’s Lavinia.’ Mythlore 29 29–51. Migne, J.-P. (ed.). 1857–1904. Patrologiae cursus completus. Series Graeca. Paris. Missiou, A. 2011. Literacy and Democracy in Fifth-Century Athens. Cambridge. Mitchell, L.G. 2001. ‘Euboean Io.’ CQ 51 339–52. Mitchell, M.M. 2005. ‘Patristic rhetoric on allegory: Origen and Eustathius put 1 Samuel 28 on trial.’ Journal of Religion 85 414–45. Mitford, T.B. 1971. The Inscriptions of Kourion. Philadelphia, PA. Modéran, Y. 1986. ‘Corippe et l’occupation byzantine de l’Afrique: pour une nouvelle lecture de la Johannide.’ Antiquités africaines 22 195–212. Modéran, Y. 2003. Les Maures et l’Afrique Romaine (IVe–VIIe siècle) Rome. Modéran, Y. 2008. ‘Des Maures aux berbères. Identité et ethnicité en Afrique du nord dans l’antiquité tardive.’ In V. Gazeau, P. Baudin and Y. Modéran (eds.), Identité et eth- nicité: concepts, débats historiographiques, exemples (IIIe–XIIe siècle). 91–134. Caen. Momigliano, A. (ed.). 1963. The Conflict between Paganism and Christianity in the Fourth Century. Oxford. Momigliano, A. 1971. The Development of Greek Biography. Cambridge, MA. Moore, C.H. 1921. ‘Prophecy in the Ancient Epic.’ HSCP 32 99–175. Moreschini, C.O.T. 2002. ‘Un testimonium sul Saturno Africano nella Iohannis di Corippo?’ Studi Classici e Orientali 47 329–349. 236 Bibliography Moreschini, C.O.T. 2007. ‘Exegesis by Distorting Pagan Myths in Corippus’ Epic Poetry.’ In W. Otten and K. Pollmann (eds.), Poetry and Exegesis in Premodern Latin Christianity: The Encounter between Classical and Christian Strategies of Interpretation. 173–98. Supplements to Vigiliae Christianae, 87. Leiden. Morgan, J.R. 1996. ‘Heliodorus.’ In G. L. Schmeling (ed.), The Novel in the Ancient World. 417–56. Leiden. Morgan, J.R. 2005. ‘Le Blanc et le noir: perspectives païennes et chrétiennes sur l’Éthiopie d’Héliodore.’ In B. Pouderon (ed.), Lieux, décors et paysages de l’ancien roman des origines à Byzance. 309–18. Lyon. Morgan, M.G. 1978. ‘The Introduction of the Aqua Marcia into Rome, 144–140 B.C.’ Philologus 122 25–58. Morstein-Marx, R. 2004. Mass Oratory and Political Power in the Late Roman Republic. Oxford.Moussy, C. 1966. Gratia et sa famille. Paris. Mueller, H.-F. 2002. Roman Religion in Valerius Maximus. London. Mulvey, L. 2009. ‘Visual Pleasure and Narrative Cinema.’ In Visual and Other Pleasures. 2nd ed. 14–27. Basingstoke. Mzamane, M.V. 1983. ‘Colonial and Imperial Themes in South African Literature 1820–1930.’ The Yearbook of English Studies 13 (Colonial and Imperial Themes Special). 181–95. Naber, S.A. 1873. ‘Observationes criticae in Heliodorum.’ Mnemosyne 2nd series 1 145–69, 313–53. Naether, F. 2010. Die Sortes Astrampsychi: Problemlosungsstrategien durch Orakel im römischen Ägypten. Tübingen. Nagy, S. 2012. ‘Fortuna in the Freestate: An Analysis of Robert Grendon’s Use of Fortuna in his Epic Paul Kruger’s Dream. ’ Akroterion 57 115–27. Neblung, D. 1997. Die Gestalt der Kassandra in der antiken Literatur. Stuttgart. Nesselrath, H.-G. 1999. ‘Dodona, Siwa und Herodot – ein Tesfall für den Vater der Geschichte.’ Mus. Helv 56 1–14. Nice, A.T. 1999. Divination and Roman Historiography. PhD Dissertation. University of Exeter. Nice, A.T. 2001. ‘Ennius or Cicero? The Disreputable Diviners at Cic. Div. 1.132.’ AClass 44 153-66. Nice, A.T. 2003. ‘The Persona of Umbricius and Divination in Juvenal, Satires Three and Six.’ In C. Deroux(ed.), Studies in Latin Literature and Roman History XI. 401–18. Brussels. Nicholas, F.-J. 1954. ‘Les Notions D’âme Et De Divinité En Afrique Occidentale’. Anthropos 51 551–94. Nilsson, M.P. 1941. Geschichte der griechischen Religion: Erster Band bis zur griechischen Weltherrschaft. Munich. Niżyńska, J. 2001. ‘Marsyas’s Howl: The Myth of Marsyas in Ovid’s Metamorphoses and Zbigniew Herbert’s “Apollo and Marsyas.”’ Comparative Literature: Journal of the American Comparative Literature Association 53 151–70. Nodzynska, A. 2006. ‘The Libyan Oracle of Zeus Ammon.’ Meander 61 113–21. North, J. 1990. ‘Diviners and Divination at Rome.’ In M. Beard and J. North (eds.), Pagan Priests. 51–71. London. North, J. 2000a. Roman Religion. Greece and Rome New Surveys in the Classics 30. Oxford. North, J. 2000b. ‘Prophet and text in the third century BC.’ In E. Bispham and C. Smith (eds.), Religion in Archaic and Republican Rome and Italy. 92–107. Edinburgh. Oakley, S. 2007. A Commentary on Livy. Books VI–X. Oxford. Bibliography 237 Oduyoye, M.A. 2009. Hearing and Knowing: Theological Reflections on Christianity in Africa. Eugene, OR. Ogden, D. 1999. ‘Binding Spells: Curse Tablets and Voodoo Dolls in the Greek and Roman Worlds.’ In B. Ankarloo and S. Clark (eds.), Witchcraft and Magic in Europe Volume 2: Ancient Greece and Rome. 1–90. London. Ogden, D. 2001. Greek and Roman Necromancy. Princeton, NJ. Ogden, D. 2007. In Search of the Sorcerer’s Apprentice: The Traditional Tales of Lucian’s Lover of Lies. Swansea. Ogden, D. 2009. Magic, Witchcraft and Ghosts in the Greek and Roman Worlds: A Sourcebook. 2nd ed. New York. Ogden, D. 2013. Drakōn: Dragon Myth and Serpent Cult in the Greek and Roman Worlds. Oxford. Olshewsky, T.M. 1991. ‘The Classical Roots of Hume’s Skepticism.’ Journal of the History of Ideas 52 269–87. Oppé, A.P. 1904. ‘The Chasm at Delphi.’ JHS 24 214–40. Opper, T. 2008. Hadrian: Empire and Conflict. London. Orlin, E.M. 2010. Foreign Cults in Rome: Creating a Roman Empire. Oxford. Osborne, R. 1985. ‘Law in Action in Classical Athens.’ JHS 105 40–58. Pachoumi, E. 2011. ‘Resurrection of the Body in the Greek Magical Papyri.’ Numen 58 729–740. Pachoumi, E. 2015. ‘The Religious and Philosophical Assimilations of Helios in the Greek Magical Papyri.’ GRBS 391–413. Pack, R. 1952. ‘A Romantic Narrative in Eunapius.’ TAPA 83 198–204. Pandey, N.B. 2013. ‘Caesar’s Comet, the Julian Star, and the Invention of Augustus.’ TAPA 143 405–49. Parke, H.W. 1967. The Oracles of Zeus. Dodona, Olympia, Ammon. Oxford. Parke, H.W. 1985a. The Oracles of Apollo in Asia Minor. London. Parke, H.W. 1985b. ‘The Massacre of the Branchidae.’ JHS 105 59–68. Parke, H.W. and D.E.W. Wormell. 1956. The Delphic Oracle. 2 Volumes. Oxford. Parker, R.C.T. 1985. ‘Greek States and Greek Oracles.’ In P.A. Cartledge and F.D. Harvey (eds.), Crux. Essays in Greek History Presented to G.E.M. de Ste. Croix on his 75th Birthday. 298–326. London. Parker, R.C.T. 2000. ‘Greek States and Greek Oracles.’ In R. Buxton (ed.), Greek Religion. 76–108. Oxford. Parker, R.C.T. 2005. Polytheism and Society at Athens. Oxford. Parker, R.C.T. 2016. ‘Seeking Advice from Zeus at Dodona.’ G&R 63 69–90. Parkin, D. 1991. ‘Simultaneity and Sequencing in the Oracular Speech of Kenyan Diviners.’ In P. Peek (ed.), African Divination Systems: Ways of Knowing. 173–89. Bloomington, IN. Pearson, L. 1939. Early Ionian Historians. Oxford. Pearson, L. 1960. The Lost Histories of Alexander the Great. New York. Peek, P. 1991. ‘Introduction: The Study of Divination, Present and Past’ & ‘Part Four: Divination, Epistemology, and Truth.’ In P. Peek (ed.), African Divination Systems: Ways of Knowing. 1–22, 133–5. Bloomington, IN. Penella, R.J. 1990. Greek Philosophers and Sophists in the Fourth Century A.D. Leeds. Perry, B.E. 1964. Secundus the Silent Philosopher: The Greek Life of Secundus, Critically Edited and Restored so far as Possible together with Translations of the Greek and Oriental Versions, the Latin and Oriental Texts, and a Study of the Tradition. Ithaca, NY. 238 Bibliography Persson, A.W. 1942. New Tombs at Dendra near Midea. Lund. Petersen, W. (director) 2004. Troy. Warner Bros. Pharr, C. 1932. ‘The Interdiction of Magic in Roman Law.’ TAPA 63 269–95. Philostratus. 2005. Life of Apollonius of Tyana. Ed. C.P. Jones. 2 Volumes. Cambridge, MA. Piccinini, J. 2016. ‘Renaissance or Decline? The Shrine of Dodona in the Hellenistic Period.’ In M. Melfi and O. Bobou (eds.), Hellenistic Sanctuaries: Between Greece and Rome. 152–69. Oxford. Pina Polo, F. 1995. ‘Procedures and Functions of Civil and Military Contiones in Rome.’ Klio 77 203–16. Plato. 1901. Phaedrus. Trans. I. Burnet (Reprinted 1979). Oxford. Plato. 1903. Euthydemus. Trans. I. Burnet (Reprinted 1977). Oxford. Plato. 1980. Symposium. Ed. K. Dover. Cambridge. Plato. 1995. Apology. Ed. E.A. Duke, W.F. Hicken, W.S.M. Nicoll, D.B. Robinson and J.C.G. Strachan. Oxford. Plato. 1995. Theaetetus. Ed. E.A. Duke, W.F. Hicken, W.S.M. Nicoll, D.B. Robinson and J.C.G. Strachan. Oxford. Plato. 2003. Republic. Ed. S.R. Slings. Oxford. Plotinus. 1966. Enneads. Volume 1. Ed. A.H. Armstrong. Cambridge, MA. Plutarch. 1939. De garrulitate = Plutarch’s Moralia VI. Ed. W.C. Helmbold. Cambridge, MA. Plutarch. 1959. On the Sign of Socrates = Plutarch’s Moralia VII. Trans. P.H. De Lacy and B. Einarson. Cambridge, MA. Popkin, R.H. and M. Goldie. 2006. ‘Scepticism, Priestcraft, and Toleration.’ In M. Goldie and R. Wokler (eds.), The Cambridge History of Eighteenth-century Political Thought. 79–109. Cambridge. Porphyry. 1966. Vita Plotini = Plotinus, Enneads I. Ed. A.H. Armstrong. Cambridge, MA. Porphyry. 1886. Vitae Pythagorae = Porphyrii philosophi Platonici opuscula selecta. Ed. A. Nauck. 2nd ed. Leipzig. Reprinted 1963. Hildesheim. Pötscher, W. 1966. ‘Zeus Naios und Dione in Dodona.’ Mnemosyne 19 113–47. Potter, D. 1994. Prophets and Emperors: Human and Divine Authority from Augustus to Theodosius. Cambridge, MA. Preisendanz, K. and A. Heinrichs (eds.). 1973–1974. Papyri Graecae Magicae. Die griechischen Zauberpapyri. Second Rev. ed. 2 Volumes. Stuttgart. Price, S. 1985. ‘Delphi and Divination.’ In P.E. Easterling and J.V. Muir (eds.), Greek Religion and Society. 128–54. Cambridge. Price, S. 2008. ‘Displacing the Gods? Agency and Power in Adaptations of Ancient History and Myth.’ Journal of Adaptation in Film and Performance 1 117–32. Proclus. 1954. In Platonis Alcibiadem I. Ed. L.G. Westerink. Amsterdam. Trans. W. O’Neill. 1st ed. The Hague 1965. Westbury. (Reprinted in 2011.) Proclus. 1903–1906. In Platonis Timaeum Commentaria. Ed. E. Diehl. 3 Volumes. Leipzig. Ps.-Plato. 1903. Theages. Trans. I. Burnet. Oxford. (Reprinted in 1977.) Ps.-Plato. 1927. Alcibiades I. Trans. W.R.M. Lamb. Cambridge, MA. (Reprinted in 2011.) Rabinowitz, N.S. and A. Richlin. 1993. Feminist Theory and the Classics. London. Radet, G. 1950. Alexandre Le Grand. Paris. Ramsey, J.T. and A.L. Licht. 1997. The Comet of 44 B.C. and Caesar’s Funeral Games. Atlanta. Rasmussen, S.W. 2003. Public Portents in Republican Rome. Rome. Rattenbury, R. M. 1938. ‘Review of Glava.’ A Study of Heliodorus. CR 52 145. Rattenbury, R. M., T. W. Lumb, and J. Maillon, J. (eds.). 1935. Héliodore: Les Éthiopiques (Théagéne et Chariclée). 3 volumes. Paris. Translated by J. R. Morgan. 1989. An Ethiopian Story. In B. P. Reardon (ed.), Collected Ancient Greek Novels, 349–588. Berkeley, CA. Bibliography 239 Rawson, E. 1973. ‘Scipio, Laelius, Furius and The Ancestral Religion.’ JRS 63 161–74. Rawson, E. 1974. ‘Religion and Politics in the Late Second Century B.C. at Rome.’ Phoenix 28 193–212. Rawson, P.B. 1987. The Myth of Marsyas in the Roman Visual Arts. Oxford. Reinach, T. 1914. ‘L’Origine du Marsyas du forum.’ Klio 14 321–37. Riess, W. 2012. Performing Interpersonal Violence: Court, Curse, and Comedy in Fourth- century BCE Athens. Berlin. Richardson, J.H. 2016. ‘Virum mihi, Camena, insece: The Cult of the Camenae and the Commemoration of Achievement in Early Rome.’ In M. García Morcillo, J.H. Richardson and F. Santangelo (eds.), Ruin or Renewal? Places and the Transformation of Memory in the City of Rome, 73–94. Rome. Richardson, L. 1992. A New Topographical Dictionary of Ancient Rome. Baltimore MD. Richlin, A. 2014. Arguments with Silence: Writing the History of Roman Women. Ann Arbor, MI. Rickenbach, J. 1999. ‘Dodona – eine der ältesten Orakelstätten in der Antieke.’ In A. Langer and A. Lutz (eds.). Orakel. Der Blick in die Zukunft. 44–9. Zürich. Riedlberger, P. 2010. Philologischer, historischer und liturgischer Kommentar zum 8. Buch der Johannis des Goripp nebst kritischer Edition und Übersetzung. Groningen. Rike, R.L. 1987. Apex Omnium: Religion in the Res Gestae of Ammianus. Berkeley, CA. Ripat, P. 2011. ‘Expelling Misconceptions: Astrologers at Rome.’ CP 106 115–54. Robinson, E.W. 1997. The First Democracies: Early Popular Government outside Athens. Stuttgart. Rodgers, R.H. 1982. ‘What the Sibyl Said: Frontinus, Aq. 7.5.’ CQ 32 174–77. Rodgers, R.H. 2004. Frontinus: De Aquaeductu Urbis Romae. Cambridge. Rohde, E. 1914. Der griechische Roman und seine Vorläufer. 3rd ed. Darmstadt. Rosán, L.J. 1949. The Philosophy of Proclus: The Final Phase of Ancient Thought. New York. Rosenbloom, D. 1993. ‘Shouting Fire in a Crowded Theater: Phrynichos’s Capture of Miletos and the Politics of Fear in Early Attic Tragedy.’ Philologus 137 159–96. Rosenberger, V. 2001. Griechische Orakel: Eine Kulturgeschichte. Darmstadt. Rosenberger, V. 2005. ‘Prodigien aus Italien: geographische Verteilung und religiöse Kommunikation.’ CCGG 16 235–57. Rosenberger, V. 2012. Religion in der Antike. Darmstadt. Rous, G. 1976. Delphes. Son oracle et ses dieux. Paris. Rubinstein, L. 2000. Litigation and Cooperation: Supporting Speakers in the Courts of Classical Athens. Stuttgart. Ruiz-Montero, C. 2005. ‘Magic in the Ancient Novels.’ In S. Frangoulidis, S. Harrison, S. Paschalis and M. Zimmermann (eds.), The Greek and the Roman Novel: Parallel Readings. 38–56. Groningen. Rüpke, J. 1995. Kalender und Öffentlichkeit. Die Geschichte der Repräsentation und religiösen Qualifikation von Zeit in Rom. Berlin Rüpke, J. 2008. Fasti Sacerdotum. Oxford. Rüpke, J. 2012. Religion in Republican Rome. Rationalization and Ritual Change. Philadelphia, PA. Russell, D.A. (ed.). 1964. Longinus on the Sublime. Oxford. Russo, F. 2005. ‘I carmina Marciana e le tradizioni sui Marcii.’ Parola del Passato 340 5–32. Ryan, F.X. (Unpublished Paper). ‘Another Unnoticed Preface in Valerius Maximus (V.M. 1.1.1).’ Said, E. 1978. Orientalism. New York. 240 Bibliography Santangelo, F. 2013. Divination, Prediction and the End of the Roman Republic. Cambridge. Santangelo, F. 2016. ‘The Statue of Marsyas.’ In M. García Morcillo, J. H. Richardson and F. Santangelo (eds.), Ruin or Renewal? Places and the Transformation of Memory in the City of Rome. 49–71. Rome. Sartorio, G.P. and A.L. Silverio (eds.). 1986. Il trionfo dell’acqua: acque e acquedotti a Roma, IV sec. a.C.–XX sec.: mostra organizzata in occasione del 16 Congresso ed esposizione internazionale degli adquedotti, 31 ottobre 1986–15 gennaio 1987, Museo della civiltà romana. Rome. Scheid, J. 2013. Les Dieux, l’État et l’individu. Réflexions sur la religion civique à Rome. Paris. Scheid, J. 2016. The Gods, the State, and the Individual. Reflections on Civic Religion in Rome. Trans. C. Ando. Philadelphia, PA. Scheidweiler, F. 1953. ‘Zu der Schrift des Eustathius von Antiochien über die Hexe von Endor.’ RhM 96 319–29. Schertz, P.J.M. 2005. Seer or Victim? The Figure of Marsyas in Roman Art, Religion, and Politics. PhD Dissertation. University of Southern California, Los Angeles. Schibli, H.S. 1993. ‘Xenocrates’ Daemons and the Irrational Soul.’ CQ 43 143–67. Schindler, C. 2009. Per carmina laudes. Untersuchungen zur spätantiken Verspanegyrik von Claudian bis Coripp. Berlin. Schofield, M. 1986. ‘Cicero for and against Divination.’ JRS 76 47–65. Scott, M. 2014. Delphi: A History of the Center of the Ancient World. Princeton, NJ. Scullard, H.H. 1981. Festivals and Ceremonies of the Roman Republic. London. Sehlmeyer, M. 1999. Stadtrömische Ehrenstatuen der republikanischen Zeit. Stuttgart. Shackleton Bailey, D.R. 2000. Valerius Maximus: Memorable Doings and Sayings. 2 Volumes. Cambridge. Shaw, G. 1985. ‘Theurgy: Rituals of Unification in the Neoplatonism of Iamblichus.’ Traditio 41 1–28. Shaw, R. 1991. ‘Splitting Truths from Darkness: Epistemological Aspects of Temne Divination.’ In P. Peek (ed.), African Divination Systems: Ways of Knowing. 137–52. Bloomington, IN. Shea, G.W. 1973. ‘Myth and Religion in an Early Christian Epic.’ Medieval Studies 35 118–28. Shea, G.W. 1998. The Iohannis or de Bellis Libycis of Flavius Cresconius Corippus (Studies in Classics 7). Lewiston. Sheppard, A. 1982. ‘Proclus’ Attitude to Theurgy.’ CQ 32 212–24. Sissa, G. 1990. Greek Virginity. Trans. A. Goldhammer. Cambridge. Slater, N.W. 2007. ‘Posthumous Parleys: Chatting up the Dead in the Ancient Novel.’ In S. Frangoulidis, S. Harrison, S. Paschalis and M. Zimmermann (eds.), The Greek and Roman Novel: Parallel Readings. 57–69. Groningen. Small, P.J. 1982. Cacus and Marsyas in Etrusco-Roman Legend. Princeton, NJ. Smardz, K.E. 1979. The Sacred Tree in Ancient Greek Religion. MA Dissertation. McMaster University. Smelik, K.A.D. 1979. ‘The Witch of Endor: I Samuel 28 in Rabbinic and Christian Exegesis till 800 A.D.’ Vigiliae Christianae 33 160–79. Smith, A. 1974. Porphyry’s Place in the Neoplatonic Tradition: A Study in Post-Plotinian Neoplatonism. The Hague. Smith, R. L. 1995. Julian’s Gods: Religion and Philosophy in the Thought and Action of Julian the Apostate. London. Bibliography 241 Smith, R. 2003. ‘What Happened to the Ancient Libyans? Chasing Sources across the Sahara from Herodotus to Ibn Khaldun.’ Journal of World History 14 459–500. Smith, S. (1993). Subjectivity, Identity and the Body: Women’s Autobiographical Practices in the Twentieth Century. Bloomington, IN. Smith, S. and J. Watson (eds.). 2006. Before they Could Vote: American Women’s Autobiographical Writing, 181–1919. Madison. Smith, W. 1876. Smith’s First Latin Course. Toronto. Sogno, C. 2005. ‘Astrology, Morality, the Emperor, and the Law in Firmicus Maternus’ Mathesis.’ Illinois Classical Studies 167–76. Sommerstein, A.H. (ed. and trans.). 2008. Aeschylus III: Fragments. Cambridge, MA. Spiller, H.A., J.R. Hale and J.Z. De Boer. 2002. ‘The Delphic Oracle: A Multi-Disciplinary Defense of the Gaseous Vent Theory.’ Journal of Toxicology: Clinical Toxicology 40 189–96. Stauner, K. 2012. ‘Review of A. Missiou, Literacy and Democracy in Fifth-Century Athens, Cambridge.’ Gephyra 9 152–5. Steinby, E.S. (ed.). 1993–2000. Lexicon Topographicum Urbis Romae. 5 Volumes. Rome. Stephen, L. 2006. Matthew Tindal, Freethinker: An Eighteenth-Century Assault on Religion. London. Stephens, S.A. and J.J. Winkler. 1995. Ancient Greek Novels. The Fragments. Princeton, NJ. Stewart, R. (ed.). 2001. Sortes Astrampsychi, volumen 2. Ecdosis altera. Munich. Stillwell, R., W.L. MacDonald and M.H. McAllister. 1976. The Princeton Encyclopedia of Classical Sites. Princeton, NJ. Stoneman, R. 2011. The Ancient Oracles: Making the Gods Speak. New Haven, CT. Storchi, A.M. 1992. ‘C. Marcio Censorino, la lotta politica intorno al pontificato e la formazione della tradizione liviana su Numa.’ AION 14 105–47. Stratton, K.B 2007. Naming the Witch: Magic, Ideology, & Stereotype in the Ancient World. New York. Stroud, R.S. 2013. The Sanctuary of Demeter and Kore: The Inscriptions. Princeton, NJ. Stuart, M. 1943. ‘Pliny, Historia Naturalis XXI.41.’ AJP 64 440–44. Stuart, M. 1944. ‘P. Oxyrhynchus 668, 188–190.’ CP 39 40–44. Stuart, M. 1945. ‘The Denarius of M.’ Aemilius Lepidus and the Aqua Marcia.’ AJA 49 226–51. Sumner, G.V. 1963. ‘Lex Aelia, Lex Fufia,’ AJP 84 337–58. Swan, P.M. 2004. The Augustan Succession: An Historical Commentary on Cassius Dio’s Roman History Books 55–56 (9 B.C.–A.D. 14). Oxford. Swain, S. 1997. ‘Biography and Biographic in the Literature of the Roman Empire.’ In M.J. Edwards and S. Swain (eds.), Portraits: Biographical Representation in the Greek and Latin Literature of the Roman Empire. 1–37. Oxford. Szemler, G.J. 1972. The Priests of the Roman Republic: A Study of Interactions between Priesthoods and Magistracies. Brussels. Tacitus. 1963. Annals. Ed. J. Jackson. Cambridge, MA. Tagliaferro, E. 2000. ‘Linguaggi oracolari a confronto e la satira di Luciano.’ In P. Radici Colace and A. Zumbo (eds.), Atti del Seminario Internazionale di Studi Letteratura scientifica e tecnica greca e latina. 11–62. Messina. Tanaseanu-Döbler, I. 2013a. Theurgy in Late Antiquity: The Invention of a Ritual Tradition. Göttingen. Tanaseanu-Döbler, I. 2013b. ‘Sosipatra – Role Models for ‘Divine’ Women in Late Antiquity.’ In M. Dzielska and K. Twardowska (eds.), Divine Men and Women in the History and Society of Late Hellenism. 123–47. Kraków. 242 Bibliography Tarn. W.W. 1922. ‘The Massacre of the Branchidae.’ CR 36 63–6. Tarn, W.W. 1948. Alexander the Great. Cambridge. Tarantino, G. 2007. Lo Scrittoio di Anthony Collins (1676–1729): I libri e i tempi di un libero pensatore. Milan. Tarantino, G. 2014. ‘Collins’s Cicero, Freethinker.’ In J.R. Wigelsworth, W. Hudson and D. Lucci (eds.), Atheism and Deism Revalued: Heterodox Religious Identities in Britain, 1650–1800. 81–100. London. Taylor, L.R. and R. Scott. 1969. ‘Seating Space in the Roman Senate and the senatores pedarii.’ TAPA 100 529–82. Themann-Steinke, A. 2008. Valerius Maximus: Ein Kommentar zum Zweiten Buch der Facta et Dicta Memorabilia. Trier. Thomas, K. 1971. Religion and the Decline of Magic: Studies in Popular Beliefs in Sixteenth and Seventeenth Century England. London. Thomas, R. 2009a. ‘The Origins of Western Literacy: Literacy in Ancient Greece and Rome.’ In D. Olson and N. Torrance (eds.), The Cambridge Handbook of Literacy. 346–61. Cambridge. Thomas, R. 2009b. ‘Writing. Reading, Public and Private “Literacies”: Functional Literacy and Democratic Literacy in Greece.’ In W.A. Johnson and H.N. Parker (eds.), Ancient Literacies: The Culture of Reading in Greece and Rome. 13–45. Oxford. Thompson, D.J. 1994. ‘Literacy and Power in Ptolemaic Egypt.’ In A.K. Bowman and G. Woolf (eds.), Literacy and Power in the Ancient World. 67–83. Cambridge. Threatte, L. 1980. The Grammar of Attic Inscriptions. Volume 1. Berlin. Timpanaro, S. 1988. Cicerone. Della divinazione. Milan. Tindal, M. 1730. Christianity as Old as the Creation: or, The Gospel, a Republication of the Religion of Nature. London. Tishken, J.E. 2007. ‘The History of Prophecy in West Africa: Indigenous, Islamic and Christian.’ History Compass 5 1468–1482. Toland, J. 1695. Two Essays sent in a Letter from Oxford. London. Toland, J. 1696. Christianity Not Mysterious. London. Toland, J. 1700. Clito: A Poem on the Force of Eloquence. London. Toland, J. 1709. Adeisidaemon, sive Titus Livius a superstitione vindicatus. Annexae sunt ejusdem Origines Judiciae. The Hague. Toland, J. 1712. Cicero Illustratus, Dissertatio Philologico-Critica: sive consilium de toto edendo Cicerone, alia plane methodo quam hactenus unquam factum. London. Toland, J. 1714. The Art of Restoring, or, the Piety and Probity of General Monk. London. Toland, J. 1720. Pantheisticon, sive formula Celebrandae Sodilitatis Socraticae, in tres particulas divisa. Cosmopolis [London]. Tomlin, R.S.O. 1988. `The curse tablets.’ In B. Cunliffe (ed.), The Temple of Sulis Minerva at Bath. Volume 2. The Finds from the Sacred Spring. 59–277. Oxford. Toner, J.P. 2009. Popular Culture in Ancient Rome. Malden, MA. Torelli, M. 1982. Typology and Structure of Roman Historical Reliefs. Jerome Lectures 14. Ann Arbor, MI. Torelli, M. 2011. ‘The Haruspices of the Emperor: Tarquitius Priscus and Sejanus’ Conspiracy.’ In J.H. Richardson and F. Santangelo (eds.), Priests and State in the Roman World. 137–59. Stuttgart. Treadgold, W. 2007. The Early Byzantine Historians. Basingstoke. Trenchard, J. 1709. The Natural History of Superstition. London. Tsetskhladze, G.R. (ed.). 2008. Greek Colonisation: An Account of Greek Colonies and Other Settlements Overseas. Mnemosyne, Supplements Series 193. Leiden. Bibliography 243 Urbano, A.P. 2013. The Philosophical Life: Biography and the Crafting of Intellectual Identity in Late Antiquity. Washington, DC. Valpy, A.J. 1819. Delectus: Sententiarum et Historiarum ad usum Tironum Accomodatus. Boston, MD. Valpy, A.J. 1836. The Second Latin Delectus. Self-published. Vandenberg, P. 2007. Mysteries of the Oracles: The Last Secrets of Antiquity. London. Van Haeperen, F. 2002. Le Collège pontifical (3ème s. a.C.–4ème s. p.C.). Contribution à l’étude de la religion publique romaine. Brussels. Van Ooteghem, J. 1961. Lucius Marcius Philippus et sa famille. Brussels. Van Wees, H. 2011. ‘Demetrius and Draco: Athens’ Property Classes and Population in and before 317 BC.’ JHS 131 95–114. Vasaly, A. 1993. Representations: Images of the World in Ciceronian Oratory. Los Angeles, CA. Verboven, K. 2007. ‘The Associative Order. Status and Ethos among Roman Businessmen in Late Republic and Early Empire.’ Athenaeum 95 861–93. Verburg, I. (ed.). 1724. M. Tullii Ciceronis Opera, Quae Supersunt, Omnia. Amsterdam. Versnel, H.S. 1991. ‘Beyond Cursing: The Appeal to Justice in Judicial Prayers.’ In C.A. Faraone and D. Obbink (eds.), Magika Hiera: Ancient Greek Magic and Religion. 60–106. Oxford. Versnel, H.S. 2010. ‘Prayers for Justice, East and West: New Finds and Publications since 1990.’ In F. Marco Simón and R.L. Gordon (eds.), Magical Practice in the Latin West. 275–354. Leiden. Versnel, H.S. 2017. Brill’s New Pauly, s.v. Defixio. Leiden. Vilaca, A. 2005. ‘Chronically Unstable Bodies: Reflections on Amazonian Corporalities.’ Journal of the Royal Anthropological Institute 11 445–64. Viveiros de Castro, E. 1998. ‘Cosmological Deixis and Amerindian Perspectivism.’ Journal of the Royal Anthropological Institute 4 469–88. Volk, K. 2009. Manilius and his Intellectual Background. Oxford. Von Albrecht, M. 1999. Roman Epic: An Interpretative Introduction. Leiden. Wardle, D. 1997. ‘“The Sainted Julius”: Valerius Maximus and the Dictator.’ CP 92 323–45. Wardle, D. 1998. Valerius Maximus: Memorable Deeds and Sayings, Book I. Oxford. Wardle, D. 2000. ‘Valerius Maximus on the Domus Augusta, Augustus, and Tiberius.’ CQ 50 479–93. Wardle, D. 2002 ‘The Heroism and Heroisation of Tiberius: Valerius Maximus and his Emperor.’ In P. Defosse (ed.), Hommages à Carl Deroux. Tome II, Prose et linguis- tique, Médecine. 433–40. Brussels. Wardle, D. 2006. Cicero: On Divination. Book 1. Oxford. Warwick, P. (ed.). 1980. The South African War: The Anglo-Boer War 1899–1902. London. Watson L.C. 1991. Arae: The Curse Poetry of Antiquity. Leeds. Watts, E. 2005. ‘Orality and Communal Identity in Eunapius’ Lives of the Sophists and Philosophers.’ Byzantion 75 334–61. Weigel, R. 1986. ‘Meetings of the Roman Senate on the Capitoline.’ L’Antiquité Classique 55 333–40. Weis, A. 1992. ‘Marsyas I.’ LIMC 6 366–78. Werbner, R. 1973. ‘The Superabundance of Understanding: Kalanga Rhetoric and Domestic Divination.’ American Anthropologist 75 1414–40. White, H. 1997. ‘An Interpretive Problem in Lycophron’s Alexandra.’ Habis 28 49–51. Whittaker, C.R. 1965. ‘The Delphic Oracle: Belief and Behaviour in Ancient Greece and Africa.’ Harvard Theological Review 58.1 21–47. 244 Bibliography Whyte, S. 1991. ‘Knowledge and Power in Nyole Divination.’ In P. Peek (ed.), African Divination Systems: Ways of Knowing. 153–72. Bloomington, IN. Wigelsworth, J. 2009. Deism in Enlightenment England: Theology, Politics, and Newtonian Public Science. Manchester. Wilburn, A.T. 2012. Materia Magica: The Archaeology of Magic in Roman Egypt, Cyprus and Spain. New Texts from Ancient Cultures. Ann Arbor, MI. Will, E. 1942. ‘Sur la nature de pneuma delphique.’ BCH 56 161–75. Wilson, P. 2003. ‘The Sound of Cultural Conflict: Kritias and the Culture of Mousike in Athens.’ In C. Dougherty and L. Kurke (eds.), The Cultures within Ancient Greek Culture: Contact, Conflict, Collaboration. 181–206. Cambridge. Wiseman, T.P. 1985. ‘Competition and Co-operation.’ In T.P. Wiseman (ed.), Roman Political Life 90 BC–AD 69, 3–19. Exeter. Wiseman, T.P. 2004. The Myths of Rome. Exeter. Wiseman, T.P. 2014. ‘Popular Memory.’ In K. Galinsky (ed.), Memoria Romana. Memory in Rome and Rome in Memory. 43–62. Ann Arbor, MI. Wolf, C. 1983a. Kassandra. Erzählung. Darmstadt. Wolf, C. 1983b. Voraussetzungen einer Erzählung: Kassandra. Frankfurter Poetik- Vorlesungen. Darmstadt. Wolf, C. 1984. Cassandra: A Novel and Four Essays. Trans. J. van Heurck. New York. Worthington, I. (ed.). 2003. Alexander the Great: A Reader. London. Worton, M. and J. Still (eds.). 1990. Intertextuality: Theories and Practices. Manchester. Wright, W.C. (trans.). 1921. Eunapius: Lives of the Philosophers/Lives of the Sophists. Cambridge, MA. Wright, W.C. (trans.). 1923. The Works of the Emperor Julian. Volume 3. Cambridge, MA. Wünsch, R. 1897. Defixionum Tabellae [Atticae], IG iii.3 Appendix. Berlin. [NB ‘Atticae’ is not part of the published title of this volume, but conventionally added to distinguish it from Audollent 1904.] Wünsch, R. 1898. Sethianische Verfluchungstafeln aus Rom. Leipzig. Wysner, G.M. 2013. The Kabyle People. Redditch. Zarini, V. 2000. ‘Brèves considérations sur le prologue du De Raptu Proserpinae de Claudien (1.1–31).’ Vita Latina 158 51–8. Zarini, V. 2003. Rhétorique, poétique, spiritualité: la technique épique de Corippe dans la Johannide. Turnhout. Zeitlin, F. 1990. ‘Thebes: Theater of Self and Society in Athenian Drama.’ In J.J. Winkler and F.I. Zeitlin (eds.), Nothing to Do with Dionysus? Athenian Drama in Its Social Context. 130–67. Princeton, NJ. Zetzel, J.E.G. 1972. ‘Cicero and the Scipionic Circle.’ HStClPh 76 173–9. Zieliński, T. 1929. Cicero im Wandel der Jahrhunderte. Leipzig. Zonnebloem Native College, Report and Class Lists for 1884. Cape Town. (= USPG Pamphlet 590, (micro- film), Bodleian Library, University of Oxford. USPG = Society for the Propagation of the Gospel.) Zosimus. 1887. New History. Ed. I. Mendelssohn. Leipzig. Index

Abydum 134–5 Ammon 11, 162, 164–76 Acarnanians 69 Ammonian 11, 169, 174–5 Acheron, River 5 Ammonium see Siwah Achilles 71, 164, 214–15, 218 Amphiaraus 144 acropolis 17, 56, 67 Amphictionic League 20 Addey, Crystal 9–10, 144–61 Anatolia 83 Adeimantos 36–7 Anaxagoras 81 ἀδίκημα 36–7 Anaxarchus 108 Admetus (Molossian king) 71 Anaxilas (tyrant of Rhegium) 23 Adriatic 72 Ancus Marcius (Roman king) 7, 91, 93, Aeacus 71 95, 99, 101 Aedesius 147–9, 151 Anderson, Benedict 116 Aegisthus 216 Anderson, Ralph 4–5, 50–64 Aelian 23 Andromache 212, 214 Aelius Sejanus, L. 110 Anio River 87–9 Aemilius Lepidus, M. 98 Antalas 163, 165 Aeneas 12, 87, 93, 164–5, 175, 201–2, Antenor 212 205–6, 216–17 Anteros 154 Aeolians 17–18, 21 Anthes 3 Aeschylus 24, 211; Agamemnon 211; The Anticleides 213 Persians 24 Antinopolis 136 Aetolians 69 Antinous 136 Africans 162–3, 165–6, 168, 171, 175, Antioch 2, 135, 138 205, 207 Antoninus (Sosipatra’s son) 9–10, 147, 149 Agamemnon 55, 212–16, 218 Antonius Diogenes 9; Wonders beyond Agathon 216 Thule 9 ἀγύρται 36–7 Aphrodite 214 ἀγωγή 42 Apollo 2, 8, 16, 18, 21, 26, 52, 56–7, 61, Ajax (Lesser) 211, 215, 218 66, 69, 91–4, 115, 117–21, 123, 166–7, Alba (Longa) 99 174, 209, 213–15, 217–18 Alexander of Jerusalem 138 Apollo Thymbraeus 13 Alexander the Great 23, 25–6, 108, 164, Apollonius of Tyana 9, 152 170–71, 174 Appian 76–7, 88 Alexandria 117, 119–20, 122–3, 134–5, Appius Claudius see Claudius Pulcher, Ap. 138, 147, 174 Apuleius 9, 35, 44, 153; Metamorphoses Amalekites 137 9, 35 Amathous 43–4 Aqua Anio (Vetus) 95 Amiternum 89 Aqua Marcia 6, 87–105 Ammianus Marcellinus 133–5, 139, 147 Aquae Sulis 41–2, 44 246 Index Arcadia 55 Besa (Bes) 134–6 Archippe 93 Bessa 8, 130–1, 135–6, 139 Arete 145, 147 Beutner, Katherine 210 arête 154 Birley, Anthony 111 Argive 16 Bithynia 136 Argos 16 Bithynium (Claudiopolis) 136 Aristagoras 2–3, 16, 19–22, 25–6 Blount, Charles 185–6 Aristodicus 17–19 Boeotian 116 Arnobius 134 Boers 12, 199, 201–7 Artaphernes 20 Book of Samuel 8 Arthur (King of the Britons) 217 Boomplaats (Battle of) 202, 205 Ascanius 93, 202 Boreium 170 Asclepigeneia 145, 155–6 Bosman, Philip 5, 65–75 Asclepiodotus of Alexandria 174 Bouleuterion 67 Asia Minor 2, 6, 16–19, 23, 148 Bradley, Marion Zimmer 13, 209, Asiaticus see Cornelius Scipio Asiaticus, L. 211–18; The Firebrand 13, 209, 211, Astrampsychus 83 215, 217–18 Atarneus 18 Branchidae (priestly clan) 16, 19–20, Athanasius of Alexandria 134 23–6 Athena 211, 215 Branchidae see Didyma Athenian Ceramicus 3–4 Brand, Johannes 201, 205 Athenians 2–3, 21, 24, 31–7, 39–42, 44, Brett, Michael 169 53–4, 56, 59–60, 116, 145, 147, 155 Brown, Peter 146 Athens 4, 31–4, 37, 39–40, 53, 56–7, 66–7, Brown, Truesdell 19, 26 116, 155, 210 Bruner, Jerome 54, 58, 60 Attica 31–4, 36, 38, 40 Bruttium 39 auctoritas 89, 100 Burgers, Thomas 202 Audollent, Auguste 32, 38 Burkert, Walter 71 Augila (Awjilah) 168–75 Byzantines 11, 162–3, 166–7, 171, Augilae 169 173–5, 215 augury 7, 78, 93–4, 100 Augustine 122 Cabaon (Laguatan chief) 162 Augustus (Roman emperor) 7, 95, 109–11, Caecilius Metellus Macedonicus, Q. 88 121, 134 Caecilius Metellus Pius, Q. 90 Aures Mountains 173 Caelian Hill 87 Caesar see Julius Caesar, C. Babylonian 10 Caesarea 117, 134, 163 Bacchanalia 91, 107 Calasiris 130–3, 140–1 Baebius Tamphilus, Cn. 94 Calidius, Q. 90 Bana, Eric 50 callida iunctura 84 Barker, Elton 52 Callisthenes 30, 68, 164 Basil of Caesarea 134 Calpurnius Piso, Cn. 106 Bath see Aquae Sulis Camelot 217 Baucis 146 Camenae 87, 95 Beard, Mary 83, 97 Cameron, Averil 170–2, 175–6 Beck, Hans 96 Canna River (Aufidus) 91 Belisarius 162–3 Canon Law 134 Bellemore, Jane 111 Capitoline Hill 87, 92, 96, 98 Bendlin, Andreas 66 Carcasan (leader of second Syrtic Bentley, Richard 190–2 insurrection) 168, 171 Berbers 11, 169 Carians 3, 21–3 Berlin 67 carmina Marciana 97, 100 Berman, Daniel 116 Carthage 98–9, 171 Index 247 Carthaginians 91–2 Constantelos, Demetrios 171 Cassandra 13, 209, 211–18 Constantine the Great 133, 140 Cassius, Longinus, C. 95 Constantius II (Roman emperor) 7, 133, Cassius Dio 96, 99, 107, 110, 136 135, 140 Castor 90, 110 contorniati 43 Castrucci, Greta 71 Corippus (Flavius Cresconius) 11, 162–9, Catiline see Sergius Catilina 171–6; Johannis (De Bellis Libycis) Cato the Elder 77–8 162, 164–6, 175 Cedrenus, George 174 Cornelius Cethegus, M. 94 Celsus 140 Cornelius Cinna, L.99 Centaurs 213, 215 Cornelius Hispalus, Cn. 7, 106 Chaldaeans 7, 10, 97, 107–8, 155 Cornelius Lentulus, L. 98 Champlin, Edward 110 Cornelius Lentulus, P. 99 Chariclea 130–1 Cornelius Scipio Aemilianus, P. Chersonese 212 Cornelius Scipio Africanus, P. 94 Chians 18, 22 Cornelius Scipio Asiaticus, L. 94 Chios 18, 21 Cornelius Sulla, L. 78 chrēsmologoi 1 Cramer, Frederick 106 Christ 12, 121, 138–40 Creon (Theban king) 211 Christ, Matthew 52 Creusa 217 Christian 7–8, 11–12, 67–8, 114–15, Croesus (Lydian king) 8, 17, 19–20, 52, 117–18, 120–3, 131, 133–5, 137–41, 61, 117–18, 171 146, 149, 165–71, 173–4, 184, 188–9, Crosby, Daniel 7–8, 114–29 194, 200–1, 203, 206, 217 Cudworth, Ralph 193–4 Christianity 12, 115, 121, 133, 140, 164, cultus publicus 82 168, 173–4, 176, 190, 194, 217 Cumont, Franz 108 Christison, Grant 200 Cunaxa (Battle of) 57 Chrysanthius 147–8 Curbera, Jaime 32, 35, 38, 42 Cicero see Tullius Cicero, M. curse tablets 3–4, 31–45 Cimmerians 9, 131 Curtius Rufus 23, 25–6 Cinna see Cornelius Cinna, L. Cybele 93, 97, 218 Circe 131 Cymaeans 18–19, 25 Claros (also Clarus) 2, 135, 144, 155 Cyme (Aeolia) 17–18 Claudia (Vestal Virgin) 96 Cyprus 16, 43, 108 Claudian 162, 165, 167 Cyrenaica 162, 168–9, 173 Claudius Caecus, Ap. 96 Cyrenaicans 166, 169 Claudius (Roman emperor) 6, 81–2 Cyrus (Persian king) 17, 19, 26 Claudius Pulcher, Ap. (consul 185 BC) 94 Cyrus (The Younger) 57–8 Claudius Pulcher, Ap. (consul 143 BC) 88, 96, 100 daemonology 118 Claudius Pulcher, Ap. (consul 52 BC) 167 Danube River 20 Clearchus 57–8 Daphne 2 cledonomancy 57 Dares Phrygius 212, 217; Acta diurna belli Clement of Alexandria 117–20, 122–3 Troiani 212 cleromancy 79–81 Darius (Persian King) 16, 19–20, 23, 26 Clitumnus River 87 David (Jewish king) 137 Cluentius 90 David, Jean-Michel 93 Clytemnestra 212, 216, 218 De Marre, Martine 11, 162–82 cognomen 91 Decemviri sacris faciundis 6, 88–9, 91, 94, Colchis 13, 213–14, 216–17 96–101 Collins, Anthony 184, 189–93 Defixionum Tabellae Atticae 32 Columella 78 Dehia 172 concordia 89, 92, 94 Delian League 22 248 Index Delphi 7–8, 16, 19–22, 26, 52, 55, 57, 61, Epirote 67, 69, 72 66–9, 71, 114–24, 135, 144, 155, 167, 174 Epirus 67–8, 71–2 Delphic Oracle 2, 22, 52–4, 57, 66, ἐργασία 39 114–24, 131, 167–8 ἐργαστήρια 39 Demetrius 4 Eretria 16 Demon 2 Eretrians 16 denarii 39 Erictho 4, 9, 168 Denzey-Lewis, Nicola 146 Eros (god/spirit) 154–5 Dercyllis 9 Eros (son of Isigeneia) 42 Derna 168 Erythrae 21 Deucalion 66, 70 Etruria 6, 78 Dickie, Matthew 133 Etruscan 6, 57, 78–80, 82 Didyma 2, 16–30 Eunapius 9–10, 144–55; Lives of the Dingane (Zulu king) 205 Philosophers and Sophists 9, 145 Dio Cassius see Cassius Dio Euripides 56, 211, 213, 218; Women of Dio Chrysostom 118 Troy 211, 213, 218 Diodorus 23, 164, 170 Eusebius 117–22 Diogenes the Cynic 118 Eustathius of Antioch 138–41; On the Dion 60 Belly-Myther 138 Dione 5, 67–8, 70–2 Eustathius of Cappadocia 9, 147–9, 153 Dionysius II (Syracusan tyrant) 60 Eustathius of Thessalonike 215 Dionysus 97 Evans, Richard 2, 16–30 Diotima 144, 154–5 extispicy 76, 80, 82 Dis 98 disciplina 78 fama 110 Dodona 5, 34–5, 65–75, 131, 135 Fama (deity) 12 Dorians 53 Faraone, Christopher 3 Dorios 5 Faustus the Manichean 122 drachuma 80 Fenestella 88–9 drakōn 10 Fentress, Elizabeth 169 Drusus Caesar 110 Fidustius 134 Dufault, Olivier 3–4, 31–49 Firmicus Maternus 134 duovir aedis dedicandae 91 Fonteius Flavianus, L. 82 Dutch Cape Colony 205 Fontenrose, Joseph 52–3, 123 Fortuna 12, 200–4, 206 East, Katherine 11, 183–98 Fortuna Primigenia 81, 91, 97 Ecbatana 17 Fowden, Garth 146 Egeria 87, 95 Frankfurter, David 176 Egypt 9, 36, 133, 135–6, 139, 141, 164, Frontinus 87, 89 168–9 Fucine Lake 7, 93, 100 Egyptian magic 9, 140 Fulvius, M. 94 Eidinow, Esther 38–40 Fulvius Flaccus, C. 95 El Kahina 11, 168, 172 Fulvius Flaccus, M. (Q.) 95 Eleusinian mysteries 37 Fulvius Flaccus, Q. 89 Eleusis 155 Fulvius Flaccus, Ser. 95 Encheles 61 Fulvius Nobilior, M. 95 Endor 8, 130–1, 137–40 Furius Purpurio, L. 89 English Enlightenment 11, 183–4, 186, 194 Ennius 79–80; Telamo 79 Gadara 154 Ephesus 9, 22, 146 Gager, John 41 Ephrem 139 Galileans 114 Epidauros 67 Garama 169 Epimenides 3 Gardens of Pallas 87 Index 249 Gellius, Cn. 93, 100 Hilton, John 8–9, 130–43 Geras, Adèle 210 Histiaeus 3, 19–21 German Democratic Republic 209 Hölkeskamp, Karl-Joachim 94 Ghirza 174 Homer 35–6, 50, 69–71, 131–3, 139, 146, Gracchus see Sempronius Gracchus 211, 214–15; Iliad 12, 211, 214–15; Graf, Fritz 134 Odyssey 131–2, 205, 215 gratia 6, 88–90, 94, 96, 98–100 Honorius (Roman emperor) 167 Great Trek 205–6 Hornblower, Simon 13 Greek magical papyri (Papyri Graecae Huet, Pierre Daniel 189 Magicae) 3, 9, 26 Hume, David 184–6 Gregory of Nazianzus 122 Hutcheon, Linda 209, 211 Gregory of Nyssa 134, 141 Hypatia 147 Grendon, Robert 12, 199–207; Paul Kruger’s Dream 199–201, 203–5, 207 Iamblichus 9, 144–5, 147–55; Grion (Mount) 23 Babyloniaca 9; De mysteriis 150; Guenfan 165–7 Letter to Arete 147; On the Pythagorean Way of Life 147 Hades 121, 131, 205 Ibn ‘Abdal al-Hakam, Abdullah 163 Hadrian (Roman emperor) 136 Ibn al-Athir, Ali 163 Halys River 117–18 Ibn Khaldûn, Muhammad 163, 173 Hammond, Nicholas 22–3, 26, 70 Ida (Mount) 212, 215 Hannibalic War 82 Illyrians 61 Haruspex 77–8, 82, 191 Imandra 213–14, 216 haruspices 6, 78–82, 97–8, 133 Ingold, Tim 54, 58–60 haruspicium 6 Ioannina 67 haruspicy 78–80 Ionian War (Revolt) 2, 16, 20–4 Hawley, Richard 145 Ionians 17–23, 25–6 Hebrew 120–2, 137 Isigeneia 42 Hecabe 212–15, 218 Isis 130, 169, 174 Hecataeus 2, 20–1, 24–6 Ismene 210–11 Hector 50, 201, 214–15 Israel 137–8 Helenos (Helenus, brother of Cassandra) Italeni (Battle of) 205 13, 212–13 Italy 7, 72, 79, 91, 93, 98, 100, 106, 202 Heliodorus 8–9, 130–6, 139–40; Aethiopica 8–9, 130–6, 139–40 James, Edward 209 Helios 131, 133, 141 Jameson Raid 203 Hellanicus 24 Januarius Nepotianus 106 Hellenes 72 Jews 106, 108, 138 Hellenistic 2, 31–6, 38–41, 43, 67, 70–1, John Chrysostom 8, 120 119, 123, 213, 215 John of Euboea 121 Hellopia 71 John Troglita 11, 162–3, 165–6, 172 Hellos 71 Johnston, Sarah 9, 69, 145–6, 149, 150–1 Heraclitus 118 Jonathan 139–40 Herbert of Cherbury 185–6 Jones, Meriel 132 Herbert-Brown, Geraldine 107 Jordan, David 32, 40, 42–3 Herodotean 3, 56 Joubert, Piet 206 Herodotus 2, 8, 16–19, 21–6, 50, 52–7, 61, Jovian (Roman emperor) 134 69–70, 163, 169, 171–3 Julian (Roman emperor) 8, 114, 121–2, Heruli 42 131, 134, 139–41, 174 Hesiod 65, 139; Catalogue of Women 65 Julius Africanus 133; Kestoi 133 hiera oikia 68 Julius Caesar, C. 78, 96, 99–100, 109–11, hieroscopic 1 193, 199 Hilarius 134 Julius Caesar, Sex. 96 250 Index

Julius Obsequens 89 λοιμός 53 Julius Paris 106 London 67 Junonia 99 Lucan 4, 9, 11, 131–2, 162, 165, 167–8, Jupiter 92, 100, 146, 164–6, 169, 171; see 171, 175; Pharsalia 11, 131 also Zeus (Ammon) Lucian 10, 118; Philopseudes 10 Justin Martyr 138 Lucius Aelius Sejanus see Aelius Sejanus, L. Justinian (Roman emperor) 162, 164, ludi Apollinares 91 169–70, 173–4 Lutatius Catulus, Q. 109 Lycophron 13, 93, 211,215; Alexandra 13, Kahina see El Kahina 93, 211 κάπηλοι 32, 39 Lysander 164 καταδεῖται 36 Lysias 3, 40 κατάδεσις 37, 44 κατάδεσμος (katadesmos) 36–7 Mammon 206 καταδἑω 35–6, 44 Manichaeans 173 κάτοχος 36 Manlius, M. 92, 94 Kenyan 51 Manlius, T. 92 Kindt, Julia 52 mantis 56, 58, 60–1 Kittelä, Sanna-Ilaria 72 Marathon (Battle of) 24–5 Kondratieff, Eric 93 Marcii 7, 90–3, 95–6, 98–101 Kremna 83 Marcius, Cn. 91 Kruger, Paul 12, 199–207 Marcius, L. 92 Marcius, P. 94 Lade (Battle of) 16, 22–3, 25 Marcius Censorinus, L. (censor 294 BC) 93 Ladysmith (Siege of) Marcius Censorinus, L. (consul 149 BC) 95 Laelius, C. 88, 98 Marcius Censorinus, L. (moneyer) 92 Laguatan 162–3 Marcius Figulus, C. 95 lamella 5, 67, 70 Marcius Figulus, T. 97 Lang, Mabel 34 Marcius Philippus, L. 95 Langlands, Rebecca 109 Marcius Philippus, Q. (censor 164 BC) 91–2 Larson, Jennifer 5 Marcius Ralla, M. 91–2 Latitudinarians 184 Marcius Rex, Q. 6, 87–96, 99–101 Latmian Gulf 22 Marcius Rutilus Censorinus, C. 91–2 Lazarus 139 Marcius Tremulus, Q. 90, 95 Lebadeia 5, 66 Mardonius 61 Leges Aeliae Fufiae 98 Marinus 155 Lepcis Magna 172, 174 Marmaridans 168 Leros 21 Mars 12, 200, 203–4 Lesbos 18, 21 Marsi 7, 79, 93, 100 lex Didia 98 Marsyas 7, 92–4, 100 lex Domitia 98 Martyrdom of Pionius 138 lex Fannia 98 Mastiman 164 lex Rubria 99 Mattingly, David 168, 173, 175–6 Leyds, Willem 205 Maurians (Mauri, Moors) 11, 162, 166–8, Libanius 147; Oration One 147 172, 174 libertas 92 Maurizio, Lisa 53–4, 57 Libo Drusus see Scribonius Libo Drusus Maximus of Ephesus 146 Lichas 2, 55–6 McClure, Laura 210 Licinius Crassus, C. 90 Medea 13 λιμός 53 Megara 39 Lipsey, Roger 122 Mela see Pomponius Mela Liternum 109 Memphis 130 Livy 6–7, 82, 89–92, 107–8, 189, 199 Mendelsohn, Farah 209 Index 251 Mercury 146 Numantines 77 Mesopotamian 57 Nyole 51 Metellus Pius see Caecilius Metellus Pius Midas 10 oblative 1 Milesians 3, 18–22, 25–6 Octavian see Augustus Miletus 2, 16–30, 81 Octavius, Cn. (consul 87 BC) 99 Miltas 60 Odysseus 131, 133, 205, 214, 216 Minoan 5, 70, 216, 218 Oenone 215 Molossian 67–8, 71 Ogden, Daniel 1–15, 44 Morgan, Gwyn 94, 96 Olympia 67 Morgan, John 9, 140 Olympian 141 Morgan le Fay 217 Orange Free State 201, 205 Mosaic Law 137 Orestes 2, 55–6, 60, 216 Moses 139, 189 Oribasius 174 Mueller, Hans-Friedrich 108 Origen of Alexandria 8, 119–20, 138–40 Mulvey, Laura 210 Orlin, Eric 7, 106 Münzer, Friedrich 94, 96 ornithomancy 78 Murray, Jeffrey 7, 106–13 Orpheus 36 Musaeus 36 Osiris 136 Mycale 21–4, 26 Ostia 91 Mycenae 68, 212, 215–16, 218 ostracism 33–4, 187 Myrcinus 21 ostraka 34 Myrto 9 Oxyrhynchus Epitome 89, 107 Mytilene 18 Myus 21 Paapis 9 Paardekraal (Krugersdorp) 204–5 Nagy, Szerdi vi, viii, 12, 199–208 Pachoumi, Eleni 133 Naia (festival) 67 Pactyes 17–19, 25 ναίω 70, 73 Paelignian Mountains 93 ναός 18 Palestine 135 Nasamones 169 Palladius 134 Natal 205–6 Pamphile 35 Naxos 20 Papyri Graecae Magicae 36 necromancy 8–9, 130–3, 135–41 Paris 212–13, 215, 218 Nemea 67 Parke, Herbert 52–3, 66, 71, 115, 123, 164 Neolithic 68, 70 Parkin, David 51 Neo-Platonism 9, 141, 145–8, 153, 155 Parnassus 120 Neoptolemus 71 paroemographer 2 Neo-Pythagorean 9, 152 Patricius 134 Nero (Roman emperor) 7, 87 patricius 94 Nestorius 145, 155 Patroclus 214 Nicaeans 121 Paul 139 Nice, Alex 6, 87–105 Paul the Chain 133, 135 Nicias 60 Paulina 136 Nicocreon (tyrant of Cyprus) 108 Pausanias 5, 23, 69, 71, 120, 216 Nile Delta 135–6 Pedasa 23 Nisibis (Siege of) 9 Pelasgians 70–1 North, John 97 Peleiai/Peleiades 69 North Africa 11, 162–3, 166, 168–9, Pella 34 172–4, 176 Peloponnese 16 Numa (Roman king) 87, 90–2, 95, 100 Peloponnesian War 2, 57 Numa Marcius 90–1 Pelops 216 Numantia 76 Penthesilea 212–13, 215, 218 252 Index performative efficacy 4, 51, 54, 61–2 Porta Capena 87, 95 Pergamon 144, 148, 152, 155 Porta Maggiore 87 Perge 83 Poseidon 215 Periander (tyrant of Corinth) 18 potestas 89 Persephone 131 Potgieter, Hendrik 201, 205 Perseus (Macedonian king) 97 Praeneste 81 Persia 17, 19–21, 23–6, 59, 109 Pretoria 206 Persians 2–3, 16–26, 52, 56, 59–61, 97, Pretorius, Andries 201–2 109, 117, 130 Pretorius, Marthinus Wessel 202 Perugia 4 Priam (Trojan king) 212–13 Peteline Wood 92 Price, Simon 97 Peterson, Wolfgang 50; Troy 50 Priene 21 φάρμακα 36–7 priestcraft 183–4, 192, 194 Pharsalus 131 princeps 81, 87 Philae 169 Priscillian of Avila 134 Philemon 146 Proclus 145, 153, 155 Philip II (Macedonian king) 21 Procopius 11, 162–4, 169–76; De Aedificiis Philistines 137 163, 169–71; De Bello Vandalico 11, Philometor 10, 151–3 163, 172 Philostorgius 121 prodigy 1, 79, 97–8, 191, 213 Philostratus 9; Life of Apollonius of Tyana 9 Proserpina 98 Phocaea 18, 21 prytaneion 67 Phocian 20 Pseudo-Longinus 120 Phoebus 121, 166; see also Apollo Psilli 168 Phrygia 93 psuchagōgos 5 Phrygian 100, 146 Ptolemy XII Auletes (Egyptian king) 99 Phrynichus 24; Fall of Miletus 24 Pydna (Battle of) 109 Pietermaritzburg 12 Pyrrhus 67 Plataea (Battle of) 61 Pythagoras 92, 108, 146–7, 152, 155 Plato 35–7, 41, 43–4, 69, 118–19, 139, Pythia 1, 8, 11, 52–4, 56–7, 61, 69, 144, 154–5; Gorgias 37; Laws 37, 44; 114–15, 117–21, 123, 144, 155, 167–8 Phaedo 36; Republic 36, 44; Timaeus 36 Platonic 132, 145, 148, 152–5, 193 quaestus 79, 82 Pliny the Elder 88, 91, 93, 99, 108 Quinctius Flamininus, L. 94 Plotinus 144, 147–8, 153, 155 Quinctius Flamininus, T. 94 Plutarch 8, 23, 56, 60, 101, 120, 123, 153, quindecimviri sacris faciundis 99, 193–4 164; On the Delay of Divine Vengeance Quirinal Hill 91 120; On the Obsolescence of Oracles 123 Plutarch of Athens 155 Rabinowitz, Nancy 210 Pollux 110 Regia 91 Polyaenus 59; Stratagemata 59 Reitz, Deneys 205 Polydamas 50 res publica 81 Polyxena 212 Retief, Piet 200, 205 Pompeius, Sex. 131 rex sacrorum 91 Pompeius Magnus, Cn. 99 Rhodes 107 Pompey see Pompeius Magnus Rhodes, Cecil John 206 Pompilia 90 Rhousopoulos, Athanasios 32 Pomponius, L. 78; fabula Atellana 78 Riffaterre, Michael 217 Pomponius Mela 169, 172–3 Ripat, Pauline 107 pontifex maximus 90–1, 97 rivus Herculaneus 87 Popillius Laenas, M. 90 Rodgers, Robert 94 Popillius Laenas, M. (consul 139 BC) 106 Roman Empire 10, 111, 123, 131–4, Porphyry 118, 144, 147–8, 153 137–8, 141, 149 Index 253 Roman Republic 6, 94 Sosimenes 39 Rome 4, 6–7, 42, 76, 78, 80, 82, 87, 90–4, Sosipatra 9–10, 144–61 96–101, 106–8, 110–11, 153, 171 South Africa 12, 199–203, 205–7 Sparta 16, 55 Sabines 92–3, 101 Spartans 2, 52, 55, 164 sacrificuli 82, 107 Speusippus 118 Said, Edward 116 Spinoza, Baruch 186 Salamis (Battle of) 56 Statius 162, 165 Salassi 89, 96 Steinmeyer, Elke 13, 209–20 Samians 22–3 Stephanus Byzantinus 169 Samos 21 Steyn, Marthinus 205 Samuel (Hebrew prophet) 8, 137–9 Storchi Marino, Alfredina 92 Santangelo, Federico 6, 76–86 Strabo 23, 120 Sardis 8, 17, 21–2 Stuart, Meriwether 94–5 Satan 8 Suetonius 107, 110 Saul (Jewish king) 137–40 Sulla see Cornelius Sulla, L. Scamander 215 Sulpicius Galba, C. 95 Scipio see Cornelius Scipio Africanus Sulpicius Galba, Ser. 95–6 Scipio Aemilanius see Cornelius Scipio Sulpicius Galus, C. 97, 109 Aemilianus superstitio 12, 107, 184, 187–90, 192–4 Scott, Michael 123 Susa 19–20 Scribonius Libo Drusus, M. 107, 110–11 Syene (Siege of) 9 Scythopolis 135 Syracuse 60 Secundus (the Silent Philospher) 10 Syrtis 169 Seleucus I 2 Selinus 3 Tabalus 17 Selloi 69, 71 Tacitus 44, 87, 107 Sempronius Gracchus, C. 78, 99 Tanaseanu-Döbler, Ilinca 145–6 Seneca 87 Tarn, William 23 Sepia (Battle of) 16 Tarquin (Roman king) 6, 77 Sergius Catilina, L. (Catiline) 92 Tegea 52, 55 Severans 82 Teledamus 216 Shackleton Bailey, David 111 Teos 18, 21 Shaw, Rosalind 51 Termessos 83 Shea, George 165, 168 Tertullian 8, 138; Apology 8 Sibyl 6, 12, 77, 93, 167, 171, 193, 201, 203, 206 Thales of Miletus 81 Sibylline Books 6–7, 12, 77, 88–9, 91, 97, Theagenes 130 99–100, 193–4 Thebaid 135 Sicily 22 Thebes 26, 116–17 Silius Italicus 162 theios anēr 9, 146 Simonides 24 Thelyphron 9 Sinifere 164 Themis 120 Siwah (Ammonium, Siwa) 164, 166, Themistocles 56, 71 168–71, 173–4 Theodoret of Cyrus 8, 114 Smyrna 164 Theodosius (Roman emperor) 11, 173 Smyrnaeus, Q. 215 Theophrastus 118 Socrates 9, 36–7, 144, 151, 153–5 Thermopylae (Battle of) 24 Socrates of Constantinople 174 Thesaurus Defixionum Magdeburgensis 45 Solomon (Byzantine general) 163 Thesprotia 5 Sopater 149 Thessalian 9, 131, 168 Sophia 42 theurgy 144–5, 148–51, 155 Sophocles 211; Antigone 211 Thrace 20–1 Sortes Astrampsychi 6, 83 Thrasybulus (tyrant of Miletus) 18 254 Index Thrasyllus 7, 107, 111 Van der Stel, Simon 200, 205 Thucydides 2, 53–4, 60, 71 Vandals 162–3, 168, 172 Tiber River 91 Veiovis 91 Tiberius (Roman emperor) 6–7, 107–11 Vekemans, Lot 211 Tigris River 58 veneficium 7 Timotheus 59 Vergil 162, 164–5, 167, 171, 199–200, Tindal, Matthew 184, 186, 190, 193 205–7; see also Virgil Toland, John 12, 184, 186, 188–9, 192–4; Vestal Virgins 87, 96 Pantheisticon 189, 192 Via Appia 42 Tomaros (Mount) 67 Victoria (British monarch) 206 Tomlin, Roger 41, 44 Viminal Hill 87 Toner, Jerry 83 Virgil 12; Aeneid 12, 171, 199, 201–2, Torelli, Mario 93 205–6 Trenchard, John 186 Vitellius (Roman emperor) 7 Trier 134 Vitruvius 87 Tripolitania 163, 173 voodoo dolls 3; ‘Mnesimachus doll’ 3–4 Trojans 12, 201–3, 206–7, 210 Voortrekkers 200 Trophonius 5, 66 Troy 50, 202, 206, 209–18 Wardle, David 106, 109 Truth 12, 200, 202, 204–5 Watts, Edward 147 Tuareg 172 Whyte, Susan 51 Tuder (Todi) 4 Wolf, Christa 209, 217–18; Kassandra Tullius Cicero, M. 6, 11–12, 69, 79–80, 209, 217 82, 89, 92, 99, 109, 118, 122–3, 183–5, ‘wooden wall(s)’ 2, 56 187–94, 199; de Divinatione 6, 11, Wormell, Donald 52–3, 115, 123 183–5, 187–94; De Natura Deorum Wünsch, Richard 32, 34–5, 38, 42–3 188–9, 193; De Republica 109, 189, 192 Tullius Cicero, Q. 78, 183 Xanthus 19, 24; Lydiaca 19 Turnus 165 Xenocrates 118 Tyche 42 Xenophon 57–9, 146; Anabasis 57 Typhon 42 Xerxes (Persian king) 16, 23, 25

Uganda 51 Zakynthia 216 Uitlander Crisis 203 Zakynthos 216–17 Uys, Dirk 201, 205 Zancle 22–3 Uys, Pieter 201, 205 Zatchlas 9 Zeitlin, Froma 116 Valens (Roman emperor) 134–5 Zeus 5, 50, 66, 68–72, 110; Zeus (Jupiter) Valentinian I (Roman emperor) 134, 140 Ammon 164, 171; Zeus Naios 67, 70, Valerian Way (Via Valeria) 87 72; Zeus Soter 59 Valerius Maximus 7, 97, 106–13; Facta et Zonnebloem College 12, 199–200 Dicta Memorabilia 106, 108 Zulu 205