Varronian Moments

Citation preview

BICS-60-2 2017

This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 1

14/11/2017 14:56:09

BULLETIN OF THE INSTITUTE OF CLASSICAL STUDIES 60-2 DIRECTOR & EDITOR: GREG WOOLF

This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 2

14/11/2017 14:56:09

BICS-60-2 2017

BULLETIN OF THE INSTITUTE OF CLASSICAL STUDIES

VARRONIAN MOMENTS EDITED BY VALENTINA ARENA AND FIACHRA MAC GÓRÁIN

INSTITUTE OF CLASSICAL STUDIES SCHOOL OF ADVANCED STUDY UNIVERSITY OF LONDON

This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 3

14/11/2017 14:56:09

Disclaimer: The Publisher, Institute of Classical Studies, and Editors cannot be held responsible for errors or any consequences arising from the use of information contained in this journal; the views and opinions expressed do not necessarily reflect those of the Publisher, Institute of Classical Studies, and Editors, neither does the publication of advertisements constitute any endorsement by the Publisher, Institute of Classical Studies, and Editors of the products advertised. Copyright and Photocopying: © 2017 Institute of Classical Studies, School of Advanced Study, University of London. All rights reserved. No part of this publication may be reproduced, stored or transmitted in any form or by any means without the prior permission in writing from the copyright holder. Authorization to photocopy items for internal and personal use is granted by the copyright holder for libraries and other users registered with their local Reproduction Rights Organisation (RRO), e.g. Copyright Clearance Center (CCC), 222 Rosewood Drive, Danvers, MA 01923, USA (www.copyright. com), provided the appropriate fee is paid directly to the RRO. This consent does not extend to other kinds of copying such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Special requests should be addressed to: permissions@ wiley.com Information for subscribers: Bulletin of the Institute of Classical Studies is published in two issues per year. Institutional subscription prices for 2017 are: Print & Online: US $483 (US), US $483 (Rest of World), €309 (Europe), £263 (UK). Prices are exclusive of tax. Asia-Pacific GST, Canadian GST/HST and European VAT will be applied at the appropriate rates. For more information on current tax rates, please go to www.wileyonlinelibrary.com/tax-vat. The price includes online access to the current and all online back files to January 1st 2010, where available. For other pricing options, including access information and terms and conditions, please visit www.wileyonlinelibrary.com/access Delivery Terms and Legal Title: Prices include delivery of print journals to the recipient’s address. Delivery terms are Delivered at Place (DAP); the recipient is responsible for paying any import duty or taxes. Title to all issues transfers FOB our shipping point, freight prepaid. We will endeavour to fulfil claims for missing or damaged copies within six months of publication, within our reasonable discretion and subject to availability. Back issues: Single issues from current and recent volumes are available at the current single issue price from [email protected]. Earlier issues may be obtained from Periodicals Service Company, 11 Main Street, Germantown, NY 12526, USA. Tel: +1 518 537 4700, Fax: +1 518 537 5899, Email: [email protected] BULLETIN OF THE INSTITUTE OF CLASSICAL STUDIES, (0076-0730), is published in two issues per year. US mailing agent: Mercury Media Processing, LLC, 1850 Elizabeth Avenue, Suite #C, Rahway, NJ 07065 USA. Periodical postage paid at Rahway, NJ. Postmaster: Send all address changes to BULLETIN OF THE INSTITUTE OF CLASSICAL STUDIES, Journal Customer Services, John Wiley & Sons Inc., C/O The Sheridan Press, PO Box 465, Hanover, PA 17331. Publisher: Bulletin of the Institute of Classical Studies is published by John Wiley & Sons Ltd., 9600 Garsington Road, Oxford OX4 2DQ and 250 Main Street, Malden, MA 02148, USA. Journal Customer Services: For ordering information, claims and any enquiry concerning your journal subscription please go to www.wileycustomerhelp.com/ask or contact your nearest office. Americas: Email: [email protected]; Tel: +1 781 388 8598 or +1 800 835 6770 (toll free in the USA & Canada). Europe, Middle East and Africa: Email: [email protected]; Tel: +44 (0) 1865 778315. Asia Pacific: Email: [email protected]; Tel: +65 6511 8000. Japan: For Japanese speaking support, email: [email protected]; Tel: +65 6511 8010 or Tel (tollfree): 005 316 50 480. Visit our Online Customer Get-Help available in 6 languages at www.wileycustomerhelp.com Production Editor: Ananya Basu (email: [email protected]). View this journal online at wileyonlinelibrary.com Designed and typeset at the Institute of Classical Studies, London. ICS Publications and Web Manager: Elizabeth Potter ([email protected]). Printed in Singapore by Ho Printing Pte Ltd. ISSN 0076-0730 (Print) ISSN 2041-5370 (Online)

This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 4

14/11/2017 14:56:09

TABLE OF CONTENTS Acknowledgements

vi

Abstracts

vii

Valentina Arena and Fiachra Mac Góráin Giorgio Piras Grant A. Nelsestuen Duncan MacRae Elisabetta Todisco

Foreword

1

Dicam dumtaxat quod est historicon: Varro and/ on the past

8

Varro, Dicaearchus, and the history of Roman res rusticae ‘The laws of the rites and of the priests’: Varro and late Republican Roman sacral jurisprudence Varro’s writings on the senate: a reconstructive hypothesis

21 34

R. M. A. Marshall

Varro, Atticus, and Annales

49 61

Daniel Hadas

St Augustine and the disappearance of Varro

76

Daniel Vallat

Varro in Virgilian commentaries: transmission in fragments

92

Wolfgang D. C. de Melo

A typology of errors in Varro and his editors: a close look at selected passages in the De lingua Latina

Bibliography

108 123

General Index

141

Index Locorum

143

BICS-60-2  2017

v © 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:39 +00:00 All use subject to https://about.jstor.org/terms

BICS 60-2 Nov14-2 FINAL-FINAL.indd 5

14/11/2017 14:56:09

ACKNOWLEDGEMENTS This special issue began life as a colloquium held at University College London in January 2015. The event could not have taken place without the generous financial support of the Institute of Classical Studies, the Institut Français, and University College London. We would like to thank all the contributors to this event. In addition to those whose work is printed here, we would like to thank Yves Lehmann and Diana Spencer, who gave papers and contributed generously to our discussions. We also thank Marie-Karine Lhommé for her intellectual input as the project was taking shape, even though in the end she could not come to the colloquium. Many others helped us at the event in various capacities, especially Gesine Manuwald, John North, Christopher Smith, and Edwin Shaw. We have incurred further debts of gratitude in preparing the papers for publication. We are especially grateful to Liz Potter at the ICS for her advice and support. The Department of Greek and Latin at University College London provided financial assistance towards the cost of publication. We also thank Greg Woolf, Christopher Smith, Giorgio Piras, and the anonymous readers for their input. We would like to thank Alex Balciunas for sourcing the cover image. Finally, we would like to thank Francesco Busti for translating Elisabetta Todisco’s paper into English and for compiling the bibliography. Valentina Arena and Fiachra Mac Góráin

BICS-60-2  2017

vi © 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:48 +00:00 All use subject to https://about.jstor.org/terms

BICS 60-2 Nov14-2 FINAL-FINAL.indd 6

14/11/2017 14:56:09

ABSTRACTS Dicam dumtaxat quod est historicon: Varro and/on the past

Giorgio Piras

Varro’s approach to his subjects is usually systematic and synchronic, but there are frequent diachronic digressions and observations on time and the past, often divided into three stages (remote past, near past, and present). I discuss Rust. 2.1, with a progressive concept of three successive stages in human history from Dicaearchus, and a fragment from Censorinus, where Varro distinguishes tria discrimina temporum. A significant affinity emerges between etymological research and the study of origins: both involve the study of antiquitas or the origo, and both use the genealogical-reconstructive method. The same image of gradus descendere indicates the sequence of logical and chronological steps in describing human history (Rust. 2.1.3–5) and etymological research (Ling. 5.7–9). Varro is fully aware of the difficulties in reconstructing the ancient past and the origins of language, because uncertainty is a characteristic of the origo of human history and of words. Keywords: Varro, De lingua Latina; Res rusticae; etymology; antiquarian; history; Dicaearchus; Censorinus; antiquitas; vetustas

Grant A. Nelsestuen

Varro, Dicaearchus, and the history of Roman res rusticae

This article reconstructs the historical narrative underlying the account of Roman farming in De re rustica, thereby shedding new light on Varro’s intellectual engagements in composing his agricultural treatise and methodology as a scholar of Rome. Central to this account is Dicaearchus of Messana’s three-stage theory of human development, which not only provides an anthropological pedigree for agriculture (in book 1) and animal husbandry (in book 2), but also establishes the theoretical framework for Varro’s innovative treatment of so-called ‘animal husbandry of the villa’ (in book 3). By presenting this last sphere of rustic production as a newly viable mode of farming contingent on the vast wealth and decline of morality in contemporary Rome, Varro uses it to emblematize ‘urban life’ as a fourth stage of human existence. The history of Roman farming thus offers a profoundly ambivalent narrative, entailing technological and material progress at the cost of moral decline. Keywords: Varro, Dicaearchus, farming, morality, Roman history, technology, anthropology, Porphyry, city

Duncan MacRae

‘The laws of the rites and of the priests’: Varro and late Republican Roman sacral jurisprudence

Starting from Cicero’s famous panegyric on Varro’s Antiquitates and attempting to look past the image of the book provided by Augustine, this article proposes a new reading of that work and its place in late Republican intellectual culture. Cicero’s specific claim that Varro opened up ‘the laws of the rites and of the priests’ for his readers allows us to contextualize the Antiquitates within a contemporary jurisprudence. The rise of Roman legal studies in general in the first century bc extended to the laws of the priestly colleges: there are signs of lively debate over their nature and the production of texts on the details of these iura. By BICS-60-2  2017

vii

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:25 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 7

14/11/2017 14:56:09

viii

BICS-60-2  2017

re-reading the fragments from the Antiquitates alongside the evidence for this sacral-legal turn, we can gain both a new appreciation for the place of law (ius) in Varro’s textualization of Roman religion and a fuller understanding of Republican legal thinking. Keywords: Roman law; pontifical law; pontifical college; Varro; Antiquitates Rerum Divinarum; Cicero; Q. Mucius Scaevola

Elisabetta Todisco

Varro’s writings on the Senate: a reconstructive hypothesis

On first becoming consul in 70 bc, Pompey asked his friend Varro to provide him with a manual on how to conduct a session of the senate. The manual was later lost. Varro returned to the subject decades later in one or more letters to Oppianus. Aulus Gellius reports on both stages of Varro’s composition, political assistance and literary composition, summarizing the contents of the letter to Oppianus. Here I will attempt to reconstruct the work, putting it in its context and discussing its implications. Keywords: Varro, Pompey, senate, Oppianus, Gellius

R. M. A. Marshall

Varro, Atticus, and Annales

This paper explores the scholarly relationship between Varro and Atticus by focusing on the attribution of a Varronian fragment (Gell. 17.21.24). Taking into account Gellius’s citation practices, it is argued that the fragment (concerning the execution of the tyrant Manlius) was mediated via Atticus, not taken directly from Varro. The evidence for Varro’s and Atticus’s friendship and scholarly cooperation undermines traditional attempts to identify any ultimate written source for this material, and prompts a re-evaluation of the relationship between the two friends’ antiquarian interests to outline the consequences of scholarly and social exchange. This investigation suggests that, like many apparent points of AtticoVarronian overlap (including Rome’s foundation-date), the fragment’s origins might owe more to personal acquaintance than secluded book-learning. Their antiquarian choices had the potential to be politically charged, and the version of Manlius’s death endorsed by Varro may thus indicate his political stance towards Caesar. Keywords: Varro, Atticus, Annales, Cicero, Caesar, Manlius, Gellius, 753 BC, antiquarianism, scholarship

Daniel Hadas

St Augustine and the disappearance of Varro

This paper argues that St Augustine’s presentation of Varro’s Antiquitates rerum divinarum in City of God was pivotal to the latter text’s disappearance. It shows how Augustine used the Antiquitates’ tripartite theology (poetic, civil, and natural) to destroy Varro’s authority on traditional Roman religion. In Augustine’s reading, Varro’s open criticism of the gods of myth and poetry implied an equal rejection of the civil cult. This left the natural gods of the philosophers, but Augustine derided Varro’s attempts at philosophical theology. The result was that, to readers of the City of God, the Antiquitates rerum divinarum appeared as a failure: Varro had been incapable of justifying traditional Roman religion, while lacking the courage to attack it openly. Readers could then turn to the City of God itself as a better guide to the Roman gods, and there was no further need to read or copy the Antiquitates. Keywords: Augustine, Varro, civil, apologetics, theology, transmission, Christianization, conversion

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:25 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 8

14/11/2017 14:56:09

BICS-60-2  2017

Daniel Vallat

ix

Varro in Virgilian commentaries: transmission in fragments

This paper analyses the transmission of Varro in late antique Virgil commentaries. Various problems are identified and discussed: the reliability of authors’ names and titles of works in citations and testimonies; different forms of quotation; complications entailed by manuscript transmission; the delimitation of the fragments; the indirect transmission of Varro already in antiquity; the status and function of Varro in a Virgil commentary. Finally I suggest that Varro had a special if implicit status in fourth-century ideological debates, in the tacit rivalry of grammarians with Christian polemicists. Keywords: Virgilian commentaries; grammarians; Servius; manuscripts; textual transmission; quotation; fragments

Wolfgang D. C. de Melo A typology of errors in Varro and his editors: a close look at selected passages in the De lingua Latina Varro’s De lingua Latina, our first grammatical treatise of any length written in Latin, is a problematic text: Varro’s linguistic theory and practice are often at odds with what later grammarians do, and the text has come down to us in a very poor state. This article examines how modern scholars have often approached the transmitted text with preconceived notions, and how this has influenced editorial choices and subsequent interpretations. The piece also looks at Varro’s own linguistic practices and predilections, and how he sometimes reaches conclusions that are logically inconsistent and indefensible. Keywords: Archaism, emendation, etymology, morphology, orthography, pronunciation, textual corruption

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:25 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 9

14/11/2017 14:56:09

FOREWORD VALENTINA ARENA AND FIACHRA MAC GÓRÁIN For we were wandering around and lost like visitors in our own city, and it was as if your books led us home, so that we could finally recognize who and where we were. You revealed the age of our homeland, the chronology of our history, you revealed the laws of the rites and the laws of the priests, you revealed our civic and military institutions, you revealed the topography of our districts and places, you revealed the names, kinds, functions, and origins of all things divine and human. Indeed you have shed a flood of light on our poets and in fact on the whole of Latin language and literature. (Cic. Acad. 1.9: editors’ translation) Cicero’s celebrated praise of Varro constructs a key moment in the intellectual history of the late Roman Republic. By neglecting and forgetting their past, the Romans had lost their sense of identity. This was recovered — and this was what Cicero saw as the truly innovative event — thanks to the writings of Varro, who revealed and ordered the names (nomina), kinds (genera), functions or duties (officia), and origins (causas) of all Roman divine and human things. Only then, so the narrative runs, had the Romans been able to find their own identity again.1 Varro’s exceptional contribution to the history of the late Republic still awaits a comprehensive treatment.2 This said, Varronian studies have recently been witnessing a renaissance after relative neglect, especially in Anglophone scholarship. In 2015 David Butterfield published an edited volume Varro Varius: The Polymath of the Roman World, Antonino Pittà a new edition of Varro’s De vita populi Romani, and Grant Nelsestuen a monograph entitled Varro the Agronomist: Political Philosophy, Satire, and Agriculture in the Late Republic. In the same year a panel at the Society for Classical Studies in New Orleans focused on The Intellectual Legacy of M. Terentius Varro: Varronian Influence on Roman Scholarship and Latin Literature (organized by Anthony Corbeill and Christopher van der Berg). Two monographs on different aspects of Varro’s work are shortly to be published;3 and an exciting new project by Katharina Volk on the sociology and politics of knowledge in late Republican Rome, which also contextualises Varro’s writing in his time and place, is currently under way. There are also a number of new editions of Varro’s texts in preparation: Robert Rodgers’s edition of De re rustica for the Oxford Classical Text series; Giorgio Piras’s edition of De lingua Latina for Teubner; Wolfgang de Melo’s (with commentary) for Oxford University Press; and Joseph McAlhany’s edition of the collected fragments for the Loeb Classical Library. New PhD dissertations on Varro’s notion of the 1

For a judicious analysis of the passage see MacRae in this volume at 34–37.

For an excellent overview of Varronian studies see Butterfield 2014; see also Riposati and Marastoni 1974 and Cardauns 1982. For the most detailed treatment of Varro’s biography and scholarly output, see Dahlmann 1935. For a recent, concise treatment focused on Varro’s ‘historical’ works, see Drummond 2013b: 412–23.

2

3

Rolle 2017; Spencer (forthcoming).

BICS-60-2  2017

1 © 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:44 +00:00 All use subject to https://about.jstor.org/terms

BICS 60-2 Nov14-2 FINAL-FINAL.indd 1

14/11/2017 14:56:10

2

BICS-60-2  2017

past (Irene Leonardis, Paris–Rome) and Varro’s De lingua Latina (Steven Lundy, Texas) have been successfully completed. It is within this scholarly context that we at University College London we organized a colloquium in 2015 with the aim of gauging the temperature (however partial it might be) of current Varronian studies. The papers collected together in the present volume grew out of this event. Contributors were invited to reflect on Varro and his work from the vantage point of their own research. Two major strands emerged in response: the first concerns Varro’s reconstruction as well as his use of the past; the second concerns ways in which later authors received him, and consequently the work which needs to be done to reconstruct and situate Varro’s own writings, as most of it survives only in fragments. While we have placed first in this collection the essays which deal with Varro’s reconstruction of the past and his ordering of knowledge, followed by those concerning the survival and reconstruction of Varro’s texts, we nonetheless firmly believe that there can be no analysis of Varro’s works, not to mention of their relationship with related works or their significance in political or cultural history, without the scholarly labour of constituting Varro’s text in the first place. All of these papers, then, strive to examine the evidence for Varro as much as Varro himself; it emerges throughout that no use or reconstruction of a previous author’s text is a neutral, objective exercise, any more than was Varro’s own reconstruction of the past.4 Whether investigating Roman history, language, genealogy, jurisprudence, religious lore, or political procedure, Varro’s main tools were historical research and etymology, a genealogical-reconstructive method which was substantially inductive and aimed to work back from the present to the past. In his notorious predilection for systems of classification and subdivisions,5 his main structures were not chronological, but rather, as Piras discusses, synchronic and organized by dimensional categories, which, however, do not preclude diachronic organization within them.6 The final aim of his inductive-reconstructive research was to uncover as much as possible that was ‘buried by the lapse of time’ (Ling. 6.2). The attempt proceeded step by step following the sometimes unreliable testimonies of the veteres, to reach the principium, which coincided with the essence, the principle of unchanging truth, or in Garcea’s words, the ‘principe universel, dont l’application s’étend bien au-delà de la grammaire’.7 In the case of language, this can be investigated via the morpho-etymological analysis of Latin words, but in fact, Varro followed an epistemological strategy most likely derived from his teacher Antiochus of Ascalon. This etymological method, onto which his genealogical research could be mapped, was not so much concerned with words themselves and their origins, but rather with ‘the things from which and for which things are named’ and especially the relation between these things, mediated by a name, which could give access to the truth.8 Departing, however, from Stoic thinking, Varro accepts that the beginnings of things may be lost to us. He is a realist, or, in the words of Volk, an ‘idealist realist’9 See König and Whitmarsh 2007: 3–39 for a Foucauldian consideration of the organization of knowledge which touches also on the Republic.

4

See Rawson 1978 (= 1991: 324–51) on the introduction of logical organization in accumulated material in the late Republic, with reference to Varro at 1991: 327–31; see also Rawson 1972 (= 1991: 58–79), Rawson 1985, and Mayer 2005.

5

6

See Volk (forthcoming), who sheds light on Varro’s attempts to systematize material.

7

Garcea 2008: 108.

8

Blank 2008: 249.

9

Volk (forthcoming).

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:44 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 2

14/11/2017 14:56:10

VALENTINA ARENA AND FIACHRA MAC GÓRÁIN: FOREWORD

3

and, as Piras reminds us, confident that some useful knowledge can be found: ‘if I have no knowledge of the roots (radices) of a tree, still I am not prevented from saying that a pear is from a branch, the branch is from a tree, and the tree from roots which I do not see’ (Ling. 7.4). When scientia of too obscure things cannot be reached, Varro argues, a certain degree of satisfaction must be found in opinio (Ling. 5.7–8). Opinion, however, was not sufficient for Augustine, whose primary interest was not how the Roman gods had been worshipped, but rather the essence of the divine. In trying to assess the truth about the divine, as Hadas shows, Augustine turns to the final three books of Varro’s Antiquitates rerum divinarum and, despite some elements of philosophical reflection (to which he cannot in any case give his assent as they aim to justify existing practices), all he can find is that ‘Varro himself preferred to be sceptical about everything (de omnibus dubitare) rather than to affirm anything. For when he had finished the first of the last three books, on the subject of the certain gods (di certi), in the next book, when taking up the uncertain gods (di incerti), he says: “If in this book I set down uncertain views of the gods (dubias de diis opiniones) I should not be reproved. For if anyone thinks that a definite verdict should and can be given, he will produce one for himself after hearing what I say. As for me, I can sooner be brought to withdraw and leave doubtful (in dubitationem revocem) what I have said in the first book, than to bring everything that I shall write in this book to any one conclusion.” Thus he renders uncertain not only what he says about the uncertain gods (diis incertis), but also what he says about the certain gods (illum de certis fecit incertum).’ By adopting this method of research, Augustine argues, Varro set down what he thought rather than what he firmly knew, ‘for in these matters man has opinions, but only God has knowledge’ (‘hominis est enim haec opinari’, Dei scire; De civ. D. 7.17). Recasting Varro’s Antiquitates rerum divinarum in a Christian framework, Hadas argues, Augustine shone such a harsh light on this work that he rendered it not only redundant to his contemporaries, but seemingly unattractive to posterity. Within the epistemological scheme of inductive genealogical research, it is possible to reverse the chronological order of research and to work from antiquity to the present. As Nelsestuen discusses, Varro borrows but also innovates on the scheme of the three stages of human history from Dicaearchus of Messana, the distinguished pupil of Aristotle whose thought was very influential in late Republican Rome. He accentuates the moralizing tone which seems to have been much weaker in Dicaearchus,10 and articulates the Roman present, luxurious with its pastio villatica (intense raising of profitable creatures), as a fourth stage of human existence, albeit without officially labelling it as such. The moralizing reading of the present is consistent with that found in De vita populi Romani, written less than a decade prior to Rust. and modelled on Dicaearchus’s Bios Hellados, being structured around the biological scheme of growth and decay. Its fourth book deplores the devastation of the Italian towns, the bellum horribile, the civil war, the selfish greed of politicians who would prefer that the sky should fall to not holding a magistracy, and the corrupting profligacy of the likes of Lucius Lucullus.11 Alongside his contemporaries, Varro represented his time as one of crisis, lamenting the loss of past traditions and sharpening, by contrast, the reproaches of the present for See Momigliano 1950 and Rawson 1985: 235 on the difference between Greek and Roman antiquarianism; Pittà 2015: 11 stresses the presence of moralizing themes also in Dicaearchus at fr. 49W and fr. 65W.

10

Fr. 109P (=115R; 426s); fr. 114P (=120R; 431S); fr. 115P (= 121R; 434S); fr. 118P (= 125bR; 432S); fr. 120P (=126R; 439S) which probably (if not explicitly) refers to Lucullus.

11

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:44 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 3

14/11/2017 14:56:10

BICS-60-2  2017

4

its inadequacies.12 In response to this, he proposed a recuperation, a systematization, and an organization of the Roman past. By ordering ‘the laws of the rites and of the priests’ in the Antiquitates, as MacRae shows, Varro was not only taking part in the contemporary debate over the laws of the priestly colleges, but was also contributing to the writing — and hence to the fixing — of those rules which regulated public life. As MacRae argues, Varro’s conception of theology was also, to a certain extent, normative in orientation. As well as in the sphere of the sacred iura, Varro also took an active part in the writing of civil law, which had been flourishing since the second century bc. As Todisco discusses, in 70 bc Varro composed for Pompey an instruction manual on how to convene a session of the senate, which he revised and updated forty years later in a letter to a certain Oppianus. As Claudia Moatti puts it effectively, ‘at a single stroke, a practical guide thus replaced a period of several years’ apprenticeship and imitation of existing models’.13 By making choices and selecting customs and institutions for the Romans which he enshrined in his books, Varro elaborated a system of knowledge that not only made the world more legible for his contemporaries,14 but also transformed patterns of customary behaviour into rules and regulations, which, sanctioned by the authority of the past, granted not only acceptance, but also legitimacy to chosen courses of action. Responding to the ignorance of magistrates, which, like Cicero, he sorely laments,15 Varro cast the rules concerning the convening of the senate in terms of continuity with the past, in line with the customs of the ancestors, while also enshrining new customs in his written work (for example, in relation to who could express his opinion first in the senate). In fact, of course, the mos was in constant flux; indeed, as the grammarians attest, Varro himself sketched a phenomenology of the novus mos.16 The self-styled custodians of the Roman past constantly had to redefine Roman traditions to maintain a sense of identity amid political and social turbulence, while having to validate contemporary responses to the changing needs of society on the basis of ancient precedent. This was partly a selfserving exercise to protect and legitimize the interests and political actions of the elite, and also to safeguard Varro and other writers’ special status as custodians and, to some extent, ‘inventors’ of the knowledge that kept the social and political system stable. In a passage of the Human Antiquities (reported verbatim by Gellius) concerning which magistrates may issue arrests and legal summons, Varro states that the tribunes of the commons have no power of summons, nevertheless many of them in ignorance have used that power, as if they were entitled to it; for some of them have ordered, not only private persons, but even a consul to be summoned before the rostra. I myself, when a triumvir, on being summoned by Porcius, tribune of the commons, did not appear, following the authority of our leading men (auctoribus principibus), but I held to the old law (vetus ius tenui). Similarly, when I was a tribune, I ordered no one to be summoned, and required no one who was summoned by one of my colleagues to obey, unless he wished. (Gell. 13.12.5–6) Wallace-Hadrill 2008: 231. On ancient perceptions of moral decline as a cause of the crisis of the Republic see Lintott 1972. 12

13

Moatti 2015: 110.

14

Moatti 2015: 2 refers to this phenomenon as an ‘epistemological revolution’.

15

Cic. Leg. 3.48 and Gell. 13.12.3–4.

On the fluidity of the notion of the mos maiorum and the establishment of novus mos, see Bettini 2011: 87–130 and Arena 2014. 16

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:44 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 4

14/11/2017 14:56:10

VALENTINA ARENA AND FIACHRA MAC GÓRÁIN: FOREWORD

5

Varro’s view had not been uncontested, and by asserting one version of procedure as the most authoritative account based on ancient practice,17 he establishes a precedent for future generations to reckon with. Roman construction of the past, as Wallace-Hadrill puts it, became ‘the living force that defines action in the present’.18 The ancestral past in fact provided a successful model of legitimate behaviour and considerable symbolic capital for the elite to exploit.19 As a result of ignorance or, at times, deliberate fabrications, false ancestral claims became more prominent in the first century bc.20 In response to this trend, and in the face of the constant expansion of the senatorial elite, Atticus was commissioned to compose genealogies of the Iunii, Fabii, and Claudii (Nep. Att. 18). As Marshall argues, it was in dialogue with, if not in reaction to, Atticus’s genealogical work that Varro composed his De familiis Troianis, which traced back the lineage of Roman families to the companions of Aeneas. Contrary to readings of Varro’s genealogical work as having supported Caesar’s claims to pre-eminence (and in the case of the De gente populi Romani, divinization),21 Marshall contends that Varro’s work aimed to undermine, or at least to circumscribe, the role of the gens Iulia. This erudite research had a strong engagement with contemporary politics, being both prompted by and having an impact on the contemporary discussions of the legitimacy of those in power. As Marshall shows, different versions of the mode of execution of the would-be tyrant Manlius have the potential to be politically charged: adopted in political discourse, they may support different visions of the past on which the present should model itself in order to gain wider validity. Some of these erudite formulations of past traditions, Marshall argues, were the result of scholarly conversations, which might have taken place over the dinner table or through an exchange of learned letters. If, on the one hand, scholarship has rightly underlined the importance of the development of writing in the transmission of traditions, which now, entrusted to books (as Cicero states in his praise of Varro in the Academica), could be systematized, ‘objectified’, and thereby potentially also criticized, on the other hand, the traditionally oral transmission of knowledge within a closed circle of friends was not eclipsed, but rather had entered into a dialectical relationship with the world of writing. Atticus may have recorded in his written work Varro’s opinion on Manlius’s death, which he might have heard during one of those conversations that Cicero recalls took place in Varro’s villa.22 And in turn, Gellius cited Varro’s view from Atticus’s text, creating for us the impression that this idea should necessarily be ascribed to one of Varro’s written works. Atticus and Varro, who were personally quite close to one another, conversed about Rome’s past and in so doing discussed also Rome’s present. At a time of fragmented Republican legitimacy and competing claims for the primacy of the Republic, Varro also used his erudite writings to negotiate his political position. Having fought alongside Pompey since the 70s, already in 59 bc he served on Caesar’s 17

See Cic. Leg. 3.20 and Livy Per. 48.

18

Wallace-Hadrill, 2008: 231.

19

Bourdieu 1977: 171–97.

20

On these issues see Flower 1996: 72–73 and 182–83 and Smith 2006: 15–17, 32–44, and 51–57.

21

See Todisco in this volume, 57–58.

Cic. Phil. 2.105: ‘think of what previously used to be spoken (dicebantur) and thought (cogitabantur) and committed to writing in that villa (litteris mandabantur): the laws of the Roman people, the achievements of our ancestors, a systematic treatment of all philosophy, of all learning (iura populi Romani, monumenta maiorum, omnis sapientiae ratio omnisque doctrinae)’. 22

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:44 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 5

14/11/2017 14:56:10

BICS-60-2  2017

6

agrarian commission, dedicated his Antiquitates divinae to him, and in the De vita populi Romani seems to attribute the outbreak of the first civil war to the intransigence and shortsightedness of the consuls rather than to Caesar’s belligerence. As Todisco underlines, there is no reason not to believe Caesar’s claim that Varro had spoken of his close bond to both Pompey and Caesar;23 who in 47 bc even asked him to organize a library, a project which Pollio later brought to fruition.24 Whether investigating the origins of words, families, and peoples, or establishing the sacrorum iura […] sacerdotum and the rules and the regulations of the senate, as a scholar, Varro aspired to establish connections between past and present. Varro’s authority as preserver and (re)builder of Roman tradition did not lie in his extensive military and political career (he reached the rank of praetor, although he never became consul), but rather in the seriousness of his scholarly method, the results of which also positioned him in the ever-changing political landscape. No act of constructing knowledge, whether reconstructing past traditions or an ancient text, is a neutral or objective exercise. Even what might nowadays appear to be a purely scientific exercise, such as establishing a word’s etymology, depends on choices made by scholars past or present. It is on these choices that our access to and understanding of the text rest: they may well be at the foundation of why a text has survived, why it has been partially or totally lost, or why the fragments survive in the form that they do. The techniques of composing a text and of editing it both involve an inherent element of inventiveness. Accordingly, de Melo offers here a precise object lesson in textual criticism, focusing on select passages of Varro’s De lingua Latina. Only books 5–10 of the original twenty-four books survive, all deriving ultimately from the eleventh-century Codex Laurentianus. Varro himself predicted that copyists would make mistakes when transmitting a thorny text (Ling. 8.51), and indeed the Codex Laurentianus is full of mistakes. De Melo identifies a selection of mistaken readings proposed by modern editors, which arise from attempts to correct the poorly transmitted manuscript readings. His analysis and proposed corrections demonstrate what a wide variety of skills and contexts an editor of Varro needs to bring to the task, from palaeography and linguistics to an imaginative and even creative sense of how Varro understood his own literary and linguistic culture. Crucially, de Melo argues, Varro himself seems to have been subject to similar prejudices to those that affected the judgment of later scholars. In reconstructing texts which survive mostly in fragmentary form, it is necessary to focus on the rhetoric, agenda, and biases of the transmitting source as well as the testimony itself. As Hadas shows, this applies forcefully to Augustine, such that it is virtually impossible to extricate the testimonies of Varro’s Antiquitates rerum divinarum in the City of God from the polemical use to which Augustine has put them. As Vallat argues, Servius probably derived his knowledge of Varro from intermediary sources such as Pliny the Elder and Gellius. Interestingly, Servius often cites Varro even when the citation does not seem directly illuminating from a Virgilian point of view. While these ‘inert’ or ‘digressive’ citations sometimes point to traditions which Virgil chose not to use, Vallat suggests that citation of Varro may also have another function: responding to the ideological debates of the preceding century between pagans and Christians, Servius may tacitly wish to ally himself to the pagans by bolstering Varro’s authority, and pitting him as a comparand worthy to stand alongside Virgil. 23

De vita populi Romani fr. 111P (= 116R; 427S); Caes. BCiv. 2.17.1–3.

24

On Caesar’s plans and Pollio’s library see Suet. Iul. 44 and Neudecker 2013: 316–17.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:44 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 6

14/11/2017 14:56:10

VALENTINA ARENA AND FIACHRA MAC GÓRÁIN: FOREWORD

7

Adopting different approaches, the essays in this collection exemplify some of the directions in which Varronian studies are heading. These include the investigation of the interplay between the search for the past and the engagement with the present, the modus operandi of these intellectuals, who were also active politicians and generals, and some of the important considerations that need to be taken into account when engaging with Varro’s texts and their reconstruction. While we are not aiming at comprehensive coverage, by collecting these essays together, we hope to contribute to the current regeneration of Varronian studies by highlighting that further work should concentrate on the evidence and what we really know about Varro within his intellectual and historical contexts. By doing so, we believe, we will be able to move forward towards a much-needed comprehensive treatment of Varro and his intellectual and political contribution to the transformation of the Republic. Valentina Arena, History Department, University College London Fiachra Mac Góráin, Department of Greek and Latin, University College London

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:44 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 7

14/11/2017 14:56:10

DICAM DUMTAXAT QUOD EST HISTORICON: VARRO AND/ON THE PAST GIORGIO PIRAS Abstract: Varro’s approach to his subjects is usually systematic and synchronic, but there are frequent diachronic digressions and observations on time and the past, often divided into three stages (remote past, near past, and present). I discuss Rust. 2.1, with a progressive concept of three successive stages in human history from Dicaearchus, and a fragment from Censorinus, where Varro distinguishes tria discrimina temporum. A significant affinity emerges between etymological research and the study of origins: both involve the study of antiquitas or the origo, and both use the genealogical-reconstructive method. The same image of gradus descendere indicates the sequence of logical and chronological steps in describing human history (Rust. 2.1.3–5) and etymological research (Ling. 5.7–9). Varro is fully aware of the difficulties in reconstructing the ancient past and the origins of language, because uncertainty is a characteristic of the origo of human history and of words. Keywords: Varro, De lingua Latina; Res rusticae; etymology; antiquarian; history; Dicaearchus; Censorinus; antiquitas; vetustas

In one of his famous contributions on ancient history Arnaldo Momigliano underlined Varro’s role as founder of antiquarian studies. He insisted in particular on the systematic aspect of Varro’s approach to all aspects of life, specifically of human society and Roman affairs as dealt with in the Antiquitates.1 Varro usually uses synchronic schemes based on dimensional categories (e.g., homines, loca, tempora, res), to organize his potentially unwieldy subject matter, whether farming or the Latin language or the history of the Roman people; but this arrangement does not preclude diachronic organization of the material and of specific subjects. The present contribution will mainly examine the structures of Varro’s Antiquitates, De lingua Latina, Res rusticae, and De gente populi Romani. We shall explore how Varro interweaves a mainly synchronic approach with diachronic observations and digressions. Varro is realistic about the obscurity of the earliest phases of his subject matter, but he nonetheless defends an inductive-reconstructive method whereby the antiquarian can still work backwards from the present. Varro thus reveals his interest in and attitude to the past. The general organization of the Antiquitates follows a comprehensive model that was dear to Varro, comprising the four categories: locus, corpus, tempus, and actio.2 The four categories are used both in the Antiquitates humanae and in the divinae (cf. August. De civ. D. 6.3). The Res humanae is in fact devoted to the treatment of homines, loca, tempora, res, and the AD to cults, broken down into homines, loca, tempora, res (in this case the res are the sacra). The categories locus, corpus, tempus, and actio – which evidently appeared to Varro to be particularly suited to the organization of his treatises – are also to be found in other Varronian works or divisions of works. This quadripartite grouping is also that of the words whose etymologies are provided in books 5–7 of De lingua Latina. Book 5 of the LL 1

Cf. Momigliano 1950.

2

On this quadripartite scheme see especially Piras 1998: 25–56; Blank 2008: 59–61.

BICS-60-2  2017

8

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 8

14/11/2017 14:56:10

GIORGIO PIRAS: DICAM DUMTAXAT QUOD EST HISTORICON

9

concerns the etymologies of words involving locus and corpus, book 6 tempus and actio, and book 7 the etymologies of words created and employed by the poets, still within the scheme of the four categories. It may be noted that this quadripartite division is an abstract model based not in fact on the linguistic characteristics of the words Varro examines, but rather on the objects which they represent (Ling. 5.13, quod quattuor genera prima rerum, totidem verborum). It is worth noting that, according to Cicero, this fourfold model used to inform more primitive works of history, works that were not rhetorically elaborate, and which the earliest Roman historians (Cato, Pictor, Piso) created from the annales maximi (De or. 2.52–53): ‘history began as a mere compilation of annals (annalium confectio) […] A similar style of writing has been adopted by many who, without any rhetorical ornament (sine ullis ornamentis), have left behind them bare records of dates, personalities, places, and events (monumenta solum temporum hominum (= corpora) locorum gestarumque rerum (= res) reliquerunt)’.3 Varro thus uses in his works an arrangement of material that Cicero considers specific to annalistic and early (indeed primitive) Roman historiography before there was proper historical research. He chose for his own works an approach that was felt to be nonhistorical, or at least one that was not currently in use. This scheme in Varro became primary rather than a subdivision within the typical chronological arrangement of annals. The overall arrangement of the De lingua Latina corresponds to the phenomenon of language itself. The work is divided into three parts (like language itself, Ling. 8.1): the first devoted to the impositio verborum, in other words, to etymology, the second to morphology, and the third to syntax.4 Etymology can often provide the starting point for a diachronic view of individual phenomena, even if it is limited to the history of specific individual words, for which Varro endeavours to provide a reconstruction of their original forms and subsequent changes, within a general synchronic framework. Since etymology is an enquiry into declinationes voluntariae (‘intentional derivations of words’), it concerns a precise point in time and can therefore be the object of historical scrutiny. It is noteworthy that in his synopsis of how he has arranged his contents, Varro uses the pluperfect subjunctive for the impositio verborum as opposed to the imperfect for the other two sections; note the distinction between ‘perfect’ and ‘imperfect’ status across the sentence (7.110): ‘I have made (feci) three parts of the whole work On the Latin Language, first how names were set upon things (quemadmodum vocabula imposita essent rebus), second how the words are declined in cases (quemadmodum ea in casus declinarentur), third how they are combined into sentences (quemadmodum coniungerentur).’5 He makes a clear distinction in Ling. 8.5–6, where he identifies duo verborum principia, namely impositio and declinatio (both derivation and declension), which provide the source and continued development of language: The origins of words are therefore two in number, and no more: imposition (impositio) and inflection (declinatio); the one is as it were the spring, the other the brook. Men have wished that imposed nouns (impositicia nomina) should be as few as possible, that they might be able to learn them more quickly; but derivative nouns 3 This famous passage is often used as evidence for the characteristics of the earliest Roman historiography: in addition to Leeman–Pinkster–Nelson 1985, ad loc., see Woodman 1988: 134–41, and 2008. 4

See Taylor 2015 for a new proposal on the contents of the third part.

The text of Ling. is from Goetz–Schoell 1910; translations derive with slight differences from Kent 1938; translations of Rust. are from Hooper–Ash 1934.

5

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 9

14/11/2017 14:56:10

BICS-60-2  2017

10

(declinata) they have wished to be as numerous as possible, that all might the more easily say those nouns which they needed to use. (6) In connection with the first class (ad illud genus, quod prius), a historical narrative is necessary (historia opus est), for except by outright learning such words do not reach us (nisi discend[end]o enim aliter id non pervenit ad nos); for the other class, the second (ad reliquum genus, quod posterius), a grammatical treatment (ars) is necessary. Historia here entails ‘knowledge, erudite research’6 and is equivalent to empeiria, as opposed to ars or technē, but it is conceivable that this research could also be characterized as historical-archaeological or historical-cultural study.7 Varro distinguishes between linguistic uses on the diachronic level, highlighting the difference between old and new forms. The changes due to use and to the passage of time represent a difficulty for the reconstruction of the past and therefore of the original impositiones, the first forms produced by the impositio, the prisca and antica verba. Varro returns to this subject in several significant passages of the treatise, especially in the section introducing the etymological triad, and in the proem to book 7. Vetustas and the passing of time are among the factors that render etymological research difficult and Varro’s other subjects obscuriora (cf. 5.3). It is as if the old forms are overwhelmed by the ruinae caused by the passing of time. Vetus and vetustas indicate what is ‘old’ (the age of a person or of an object), often in a negative sense: 5.4, vetustas pauca non depravat, multa tollit. At the beginning of both books 5 and 7, Varro compares the damage visited by time on the original linguistic heritage to the aging of man, a process that renders people unrecognizable. In book 5 the reference is to an ordinary life experience, observation of a boy’s aging (5.5 ‘whom you saw beautiful as a boy, him you see unsightly in his old age. The third generation (tertium seculum) does not see a person such as the first generation saw him’); in the seventh the examples are drawn from literary myth: Epimenides awakens after a fifty-year sleep and is recognized by no one, and the same thing happens after fifteen years to Teucer, in the homonymous tragedy by Livius Andronicus (7.3).8 Sometimes vetustas is identified with antiquitas, and the veteres are the antiqui tout court.9 One might say that vetustas is an obstacle to etymological research, while the reconstruction of antiquitas is its objective. Varro, however, declares himself determined to overcome the ‘dark forest’ caused by the ravages of time (5.5):

Dahlmann 1940 translates it as ‘Wissen’, Traglia 1974 as ‘ricerche storiche’. On the passage cf. Blank 2008: 66–67.

6

7 One should maintain the transmitted descendendo, against the majority of editors: it would be unusual to say that the first genus, i.e., impositio, discendo pervenit ad nos. We find a modal gerund for instance in Rust. 1.1.11, quae ipse in meis fundis colendo animadverti [see Flach, ad loc.], but there the subject of gerund and perfect is the same, while here we have an impersonal gerund with genus as subject of pervenit; discere in 8.3, recalled by Dahlmann 1940, ad loc., refers to declinatio and not impositio. Genus […] prius and genus […] posterius concern the logical and chronological priority of the impositicia nomina to the declinata and do not mean ‘first class’, ‘second class’; descendendo indicates the several stages of linguistic transformation of the original forms. See below for other examples of descendere in similar contexts. 8

Antiquarian discovery is therefore a kind of agnition; cf. Romano 2003: 106–07.

Rust. 2.1.6 veteres poetae (thus also Ling. 7.52); 2.1.9 vetere instituto; 3.1.2 vetustissimum oppidum; Ling. 5.5 verborum novorum ac veterum; 9.20 veteres leges (as opposed to the novae); the expression consuetudo vetus occurs frequently (Ling. 5.173; 6.2; 9. 13; 9.21; 10.73; cf. also usum veterem at 10.78). 9

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 10

14/11/2017 14:56:10

GIORGIO PIRAS: DICAM DUMTAXAT QUOD EST HISTORICON

11

for there is no slight darkness in the wood where these things are to be caught, and there are no trodden paths (semitae tritae)10 to the place which we wish to attain, nor do there fail to be obstacles in the paths, which could hold back the hunter on his way. The linguist may avail himself of various instruments furnished by etymological science in order to overcome these obstacles, but his research knows clear limits. The etymologist cannot hope to achieve a proper etymology of every word, but he may still be content with explaining the origin of the declinata, i.e., words which can be explained through ars, because he will thereby have explained the greater part of the lexical heritage, which consists above all in words derived from the original sources. A series of derivations may be established, and even if the origins of the earliest term remain obscure, useful work will nonetheless have been accomplished (7.4): he who shows that equitatus ‘cavalry’ is from equites ‘cavalrymen’, equites from eques ‘cavalryman’, eques from equus ‘horse’, even though he does not give the source of the word equus, still gives several lessons and satisfies an appreciative person.11 Etymology is a formidable tool in recovering the past, and its methods embrace both historia and ars,12 but the damage caused by vetustas imposes a limit on the success of etymological enquiry. The synchronic approach is evident in the only Varronian work that has been preserved intact, namely the De re rustica. The three books treat the rural world according to the model of peripatetic-Ciceronian dialogue: the first book discusses agriculture, the second livestock, and the third farmyard animals (which Varro defines as pastio villatica). His treatment of the material in the course of the work varies according to geographical area and population, but past and present are hardly ever distinguished. At the end of what we might consider the proem to the De re rustica (1.1), having announced that the work will be divided into three books, Varro announces that he will treat agri cultura in book 1, first separating that which does not pertain to it (namely the raising of animals, which will be the subject of books 2–3), and then following the ‘natural’ divisions of the material (sequens naturales divisiones). The general structure of the discussions is synchronic, but the work contains diachronic narrations as well, especially the proems, which introduce reflections on the history of culture. An important exception to the synchronic treatment is to be found in the second book, where after the proem he extols country life as opposed to life in the city.13 At the beginning of his treatment of pastoralism (2.1.2), Varro announces that he would like to deal only with the historical part of the issue (ego vero, inquam, dicam dumtaxat quod est historicon), regarding the origo and the dignitas of res pecuaria, but that Scrofa will discuss the ars and scientia pastoralis. There is no doubt here that historicon has to do with the very informed 10

Cf. Lucretius’s Callimachean lines at 1.926–27 = 4.1–2 avia Pieridum peragro loca nullius ante / trita solo.

11

Cf. also 6.37.

Taylor (2015: 25) affirms that etymology is a matter of historia and it is not an ars: the lost triad of Ling. 2–4 discussed the nature of ars and the usefulness of the disciplina verborum originis, but it seems to me more probable that Varro concluded that both methods should be used in combination.

12

On the subject cf. also Rust. 3.1.4, neque solum antiquior cultura agri, sed etiam melior, and 2. pref. 1 viri magni nostri maiores non sine causa praeponebant rusticos Romanos urbanis (the entire proem to book 2 is a moralistic exaltation of country life over city life).

13

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 11

14/11/2017 14:56:11

BICS-60-2  2017

12

description of aspects that are related to history, in a well-attested sense of historia as knowledge based on sound experience and reliable sources, and so the adverb historicos could refer to an accurate and informed treatment.14 Varro, indeed, treats the origo of pastoralism in paragraphs 3–5, and dignitas in 6–10. He offers an overview of the origo, adopted from Dicaearchus of Messana, who distinguished three successive stages in human history, a history in which human beings and animals always existed, regardless of the issue of principium that philosophers discuss (2.1.3–5 = Dicaearchus fr. 48 W. = fr. 54 M.): As it is a necessity of nature that people and flocks have always existed (whether there was an original generating principle of animals, as Thales of Miletus and Zeno of Citium thought, or, on the contrary, as was the view of Pythagoras of Samos and of Aristotle of Stagira, that there was no point of beginning for them), it is a necessity that from the remotest antiquity of human life they have come down, step by step to our age (necesse est humanae vitae ab summa memoria gradatim descendisse ad hanc aetatem), as Dicaearchus teaches, and that the most distant stage (summum gradum) was that state of nature in which man lived on those products which the virgin earth brought forth of her own accord; (4) they descended from this stage into the second, the pastoral (ex hac vita in secundam descendisse pastoriciam), in which they gathered for their use acorns, arbutus berries, mulberries, and other fruits by plucking them from wild and uncultivated trees and bushes, and likewise caught, shut up, and tamed such wild animals as they could for the like advantage […] (5) Then by a third stage man came from the pastoral life to that of the tiller of the soil (tertio denique gradu a vita pastorali ad agri culturam descenderunt); in this they retained much of the former two stages, and after reaching it they went far before reaching our stage. Even now (etiam nunc) there are several species of wild animals in various places. Varro thus gives us a natural summus gradus, during which humans lived off wild foods from nature, then a second pastoral stage, followed by a third agricultural stage. The third stage, Varro explains, develops from the previous two,15 and retains many of their characteristics, in addition to developing further until the present day. One might wonder whether he views the present as a fourth stage, but it seems to me more likely that his words must be interpreted as a contrast between the ‘earlier’ and the ‘now’, with the ‘earlier’ being tripartite (he seems to mark a break with etiam nunc).16 The binary opposition between city and country life belongs on a different level, since it concerns space rather than time, even though it has a clear temporal dimension in that country life came before city life (3.1.1–4, esp. 1):

Cf. Rust. 3.16.3 ‘Merula [‘Blackbird’, who deals with bird breeding!], as he has done in other cases, presents a learned treatment (historicos) also for beekeepers’. For the rare adverb historicos (= ἱστορικῶς) see TLL, VI 2843, 16–19 (‘fere i.q. secundum rationem eorum quae experientia inventa sunt’). For this sense of historia, cf. Varro Sat. Men. 414 (historiam necessariam, and cf. Cèbe 1994: 1737); Dicaearch. fr. 49, 30–31 Wehrli = 56A, 43–44 Mirhady ‘these things […] are not asserted by us, but by those who have spoken after thoroughly researching (ἱστορίᾳ διεξελθόντες) ancient matters’.

14

Cf. also Rust. 1.2.16 (= Dicaearch. fr. 51 W. = 55 M.) ‘You may even add this that the shepherd’s life is the treble (incentivam), and the farmer’s plays the accompaniment (succentivam), if we may trust that most learned man, Dicaearchus’, and 2. pref. 4. 15

For an analysis of Varro’s history of civilization which proposes a different conclusion on the present age as a ‘fourth stage’ in Dicaearchan terms, see Nelsestuen, section 3, in this volume. 16

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 12

14/11/2017 14:56:11

GIORGIO PIRAS: DICAM DUMTAXAT QUOD EST HISTORICON

13

Though there are traditionally two ways in which men live (cum duae vitae traditae sint hominum [note the use of vitae as in 2.1.3]) — one in the country, the other in the city […] there is no doubt that these differ not merely in the matter of place, but also in the time at which each had its beginning. While the city/country binary is not to be confused with Dicaearchus’s model of the three phases of human history, there is some overlap between the two in that the transition from rural to urban life coincides, at least in Rome, with the transition from Dicaearchus’s second to third phases. In fact, the founders of Rome learned the cultivation of the fields from their shepherd ancestors (2. pref. 4),17 even though Romulus and Remus were shepherds themselves (2.1.9). The second point that Varro had promised (2.1.2) to discuss after origins of pastoralism — that of dignitas — is also treated historically, possibly with some dependence on Dicaearchus, and with a focus on cultural matters. He begins by stating that ‘of the ancients the most illustrious were all shepherds, as appears in both Greek and Latin literature, and in the ancient poets’ (2.1.6). This lays the basis for valorizing the origins of Rome, since Romulus and Remus were shepherds (2.1.9–10). There is an undercurrent of moralistic exaltation of rural as opposed to city life, but it is also clear that Varro embraces the idea of human progress. Agriculture represents an improvement upon previous stages, not a decline; this may mark a departure from Dicaearchus, but there is in fact much debate about Dicaearchus’s moral evaluation of the development.18 In Varro, the passing from the natural to the agricultural phase is certainly seen as undeniable technical progress. On the other hand, there is more than a touch of decadence about the transition from rural to urban living. This transition probably took place during the third (agricultural) phase, or perhaps in the lead-up to the present day; accordingly, it is to be located either squarely within the third phase or beyond the entire schema. Varro’s tendency to systematic organization leads him to reflect on the category of tempus and to stratify it comprehensively into layers. Another general historical schema that reveals Varro’s view on human history and the history of culture is to be found in a fragment preserved by Censorinus (De die natali 21.1–2 = Varro De gente pop. Rom. lib. I, test. Fraccaro = Ant. hum. lib. XIX fr. I Mirsch):19 I will now speak of that period which Varro calls historic (quod historicon Varro appellat). This author divides time into three periods: the first extends from the origin of man to the first cataclysm, and he calls it uncertain (adelon), on account of the obscurity in which it is concealed (propter ignorantiam). The second extends from the first cataclysm to the first Olympiad and as it has given rise to numerous fables (quia multa in eo fabulosa referuntur) he calls it mythological (mythicon). Flach (1997: 181–82) is correct that qui condiderunt urbem is an expansion of the immediately preceding progeniem suam, rather than having the earlier pastores as its antecedent; the following progenies eorum refers to the relative sentence, qui condiderunt urbem, rather than to the preceding progeniem suam. 17

Cf. Fraccaro 1907: 115; Della Corte 1976: 129–30; Ax 2000: 345–46; Saunders 2001; Schütrumpf 2001; Feeney 2007: 113; Baier 2007: 262–64; McConnell 2014: 134–35; cf. also Nelsestuen 2011: 340–41. The key issue is how to reconstruct Dicaearchus’s opinion on the basis of frs 48 and 49 W (= 54 and 56A M.). Even in the history of language, the passage of time does not necessarily imply only damage and devastation; the consuetudo loquendi may also include improvements: cf. Ling. 9.17. According to Deschamps (1987, in particular 189), Varro would have had a cyclical concept of history. 18

On the fr. see most recently Cole 2004: 419–22; Möller 2005: 255–58 (with previous bibliography), and Feeney 2007: 77–82. 19

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 13

14/11/2017 14:56:11

BICS-60-2  2017

14

The third extends from the first Olympiad to our time. He calls this historic (historicon), because the events which transpired during this interval are related in reliable histories (quia res in eo gestae veris historiis continentur).20 (2) As to the first period, whether or not it had a commencement, it is certainly impossible to establish how many years it comprised. As to the second, we cannot say exactly, but we may believe that it covered about one thousand six hundred years. From the first cataclysm, which is called that of Ogyges, until the reign of Inachus about four hundred years are counted […] from that time until the first Olympiad, a little more than four hundred are counted. And as solely these events, although belonging to the end of the mythical period, approach the age of historians, several writers have attempted to give the number more exactly. Censorinus has just stated that if he had known the origo mundi, he would have begun there (20.12). He will treat instead the time period that Varro called historicon, reporting Varro’s distinguished tria discrimina temporum, their etymologies, and the rationale for the division. Censorinus specifies the duration of these time periods. The beginning of the first is uncertain, because it is not known whether there was a beginning at all or whether the world has always existed (cf. above Rust. 2.1.3, where Varro states decisively that life is eternal, et homines et pecudes cum semper fuisse sit necesse natura, and also Sat. Men. 84 and 268).21 Due to this uncertainty, the first period is not of an identifiable duration. The second lasted about 1600 years until the first Olympiad at 776 bc (there are doubts about the text that would indicate further subdivisions of 400 or 800 years). It is worth noting that the final 400 years of the mythical period could be better known than its earlier years, since those later years are closer in time to the era of the writers (a memoria scriptorum proximos). Censorinus afterwards tells us in reference to the third historical period that Varro studied in depth the chronology and synchrony of various civitates, eliminating ‘all confusion around the issue and caused the light of truth to emerge’ (21.5), probably alluding to a series of astronomical calculations, the curiosae computationes that Arnobius informs us were contained in the first book of the De gente populi Romani.22 The provenance of this testimonium of Varro by Censorinus is uncertain, but in fact it is very likely to have come from De gente populi Romani and not from the Antiquitates as argued by Mirsch.23 The schema is undoubtedly of Greek origin.24 The verae historiae are opposed to fabulares poetarum historiae at 4.6 (cf. also 17.3); on this opposition cf. Wiseman 2002: 331–38. Here historicon (= ἱστορικόν) means ‘“storico” nel senso originario di “contenuto in una relazione verbale o scritta (ἱστορία) risalente a un testimone oculare”’ (Rapisarda 1991: 251). Poucet (1987) points out the singularity of this threefold partition of past time in the Roman historical and literary tradition.

20

Sat. Men. 84: Varro autem in satura quae scribitur de salute ait mundum haud natum esse neque mori; 268: nec natus est nec morietur: viget, veget, utpote plurimum. The philosophical genus of theology (see below) dealt also with the question a quodam tempore an a sempiterno fuerint dii (Ant. div. fr. 8 Cardauns).

21

Arn. Adv. nat. 5.8.7 (= Varro gpR fr. 20); on this counting see Grafton–Swerdlow 1985: 456; 1986: 149. It would seem that Cicero is referring precisely to this fact in the famous laus Varronis in the Academica (1.9): tu descriptiones temporum […] aperuisti. 22

23

Cf. Rocca-Serra 1980: 66. Franceschi 1954: 411–12 thinks the testimonium derives from the Antiquitates.

It could derive from the same Dicaearchus of whom we have only a few fragments. I would point to the use of λίαν μυθικόν, as opposed to φυσικόν, in fr. 49, 11 W. (= 56A, 17–18 M.). The adjective incertus evokes the Greek sceptic tradition: cf. Varro Log. 77 (= August. De civ. D. 29.1): ob hoc distinguit ab Academicis novis, quibus incerta sunt omnia, and Cic. Luc. 54: ea dico incerta quae ἄδηλα Graeci. But the source could be also Castor of Rhodes (cf. Ax 2000: 359 and n. 64, with previous bibliography) or Eratosthenes (thus most recently Feeney 2007: 147 n. 92). 24

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 14

14/11/2017 14:56:11

GIORGIO PIRAS: DICAM DUMTAXAT QUOD EST HISTORICON

15

There is some similarity with the terminology of Varro’s ‘tripartite theology’, with the tria genera theologiae, namely mythicon or fabulare, physicon or naturale, and civile,25 especially with the mythical theology, used in particular by the poets. The three theological genera do not receive a chronological treatment but rather concern the ratio quae de diis explicatur (Ant. div. 7) and something similar could be said about the three periods of humanity, characterized by the reliability of the information we receive about them (uncertain, fictitious, true) and not by their era, even if their level of reliability is essentially a consequence of the times in which they occur; it is, after all, possible that Varro had in mind a logical and therefore chronological arrangement of the three genera theologiae also. The degree of ‘historicity’ of an era depends not only on the possibility to narrate it by reliable stories, but also to frame it with a certain chronological scheme.26 And, indeed, Varro, in describing in the De gente populi Romani the process that culminated in the birth of the Roman people, does not attempt to go back beyond the limit of the adelon time, the flood of Ogyges, ‘and does not propose to himself, as the starting point from which he may arrive at Roman affairs, anything more ancient than the flood of Ogyges, that is, which happened in the time of Ogyges’ (fr. 3 = August. De civ. D. 18.8). The same limit is used as a terminus ante quem for dating the foundation of Thebes, the beginning of urban life in Greece (Rust. 3.1.2–3), but it is impossible to calculate when rural life actually began, and so Varro merely states that the two events are separated ‘by an immense number of years’ (immani numero annorum). From what we have seen so far in our survey of Varro’s treatment of the past, it emerges he exhibits the same methodological approach in the reconstruction of the past both when he has to return to the most ancient events and when he must find traces of ancient institutions and customs by means of etymological inquiry. In both cases, Varro endeavours to reconstruct the origo of a word (Ling. 5.7 nunc singulorum verborum origines expediam), or of an institution (for instance, pastoralism), but also of the human race. The origo is shrouded in uncertainty, and uncertainty, as we have seen, is a characteristic of the most ancient era, adelon, which Varro has no hope of being able to reconstruct. We have seen that in the case of etymology the genealogical-reconstructive method cannot always access the ultimate origo, the original impositio (Ling. 7.2 tamen latent multa; 4 non omnium verborum posse dici causa). Unlike the Stoics, Varro does not believe that the origin of words is always identifiable. He also departed from Stoic thinking on human history, which posited a beginning, following instead Aristotle and the Pythagoreans (Rust. 2.1.3): if the beginning is lost in infinity, it is more understandable that it cannot be chronologically fixed. The etymological method involves reconstructing in a backward direction the successive stages of linguistic development. The proem to book 7 of Ling. is particularly instructive, where the metaphor of the tree and its roots is especially relevant for etymological reconstruction (7.4):

25 See August. De civ. D. 6.5 (= Varro Ant. div. 7 C.): tria genera theologiae dicit esse […] eorumque unum mythicon , alterum physicon, tertium civile, i.e., with Latin words, fabulare or fabulosum, naturale, civile (see also frs 8–10 C). On the three branches of theology see also MacRae, p. 43 and Hadas, p. 82–89 in this volume.

The temporal category of adelon facilitates an approach to mythical and historical periods on the basis of these two criteria (Hoces Sánchez 2012: 156).

26

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 15

14/11/2017 14:56:11

BICS-60-2  2017

16

if I have no knowledge of the roots (radices) of a tree, still I am not prevented from saying that a pear is from a branch, the branch is from a tree, and the tree from roots which I do not see. Just as one can work back from the fruit to the roots, which cannot be seen and so must be assumed, so too one can trace back from the final linguistic derivation to the first impositio; and even if one reaches the end of the line, that in itself is a considerable achievement (see also above, p. 11). There is a significant epistemological similarity here with what we saw regarding the tria tempora of human history, especially as regards the limits upon historical reconstruction as it is pursued backwards. This method allows Varro to reconstruct the chronology of astronomical events, ‘counting backwards the eclipses’ (Censorinus 21.5). In the passage of Ling. he clearly states that one who does not know the roots is not thereby prevented from deducing the subsequent derivations: the ultimate goal is always the origo, but the subsequent steps of the process under examination are also relevant. A significant affinity therefore seems to emerge between etymological research and the historical study of origins. The object of the study is the same, namely antiquitas and origo, attained through the genealogical-reconstructive method.27 The events under investigation can then also be chronologically narrated by reversing the backwards order of the research into a normal chronological order of presentation, as in the case of the De gente populi Romani28 or of the history of the res pecuaria at Rust. 2.1.3–5; but in Varro, the systematic arrangement of the results of his researches on the basis of general categories, the naturales divisiones mentioned in Rust., seems to be more frequent. Another example of the affinity between specific etymological analysis and more general study of the past emerges from several significant lexical similarities between more general passages from his treatises. I am referring to the image of the gradus descendere which indicates the successive stages of a general developing process. Gradus descendere means descending a flight of stairs.29 The gradual progression is typical of the logical reasoning in dialectics,30 but also in rhetoric, where the gradatio is the climax, or epiplokē, in which the various elements are linked as in a chain.31 Varro utilizes this image in several passages to indicate the sequence of logical and also chronological steps, which are distinct but contiguous. As we have seen at Rust. 2.1 regarding the various stages of human civilization, the humanae vitae gradually succeeded one another up until the present moment (necesse Cf. Romano 2003: 114–17. On the epistemological status of Varro’s etymology and its ultimate antiquary and historical goal see Blank 2008: esp. 68. 27

Narrative and probably organizational traces, mainly diachronic ones, are to be found in the De gente populi Romani (cf. in part frs 3 and 8) and (a little more doubtfully) in the De vita populi Romani, where Riposati 1939: 257–58 and now Pittà 2015: 12 and 61–66 argue for a general chronological arrangement.

28

Cf. Plin. HN 36.88: porticusque descenduntur nonagenis gradibus; Ovid has stairs in mind at Fast. 6.397–98: huc (i.e. Nova Via) pede matronam nudo descendere vidi (perhaps because he uses gradus in the following verse: obstipui tacitus sustinuique gradum); not literal: Varro Rust. 3.11.1: naturalis aut lacus aut stagnum aut manu facta piscina, quo gradatim descendere possint (but the otherwise related Columella, Rust. 8.15.3 may be a literal use). 29

30

See in particular Cic. Nat. D. 1.89, who clearly distinguishes descendere from praecipitare.

Rhet. Her. 4.34: ‘Climax is the figure in which the speaker passes (descenditur) to the following word only after advancing by steps to the preceding one (ad superius ascensum est)’; cf. Quint. Inst. 9.3.54–55 and 8.4.28 (on amplificatio and deminutio); Isid. Etym. 2.21.4. On gradatio see Lausberg 1990: § 623, 315–17; Wisse– Winterbottom–Fantham 2008: 324. 31

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 16

14/11/2017 14:56:11

GIORGIO PIRAS: DICAM DUMTAXAT QUOD EST HISTORICON

17

est humanae vitae ab summa memoria gradatim descendisse ad hanc aetatem), the summus gradus being natural. From this state, the following pastoralist one is derived (ex hac vita in secundam descendisse pastoriciam). In the third stage (tertio denique gradu), men moved from pastoralism to agriculture (ad agri culturam descenderunt), a phase that maintained many of the previous circumstances and from which they arrived finally at the current state (in qua ex duobus gradibus superioribus retinuerunt multa, et quo descenderant, ibi processerunt longe, dum ad nos perveniret). Still referring to Dicaearchus’s opinion on the movement between the various stages, Varro uses the term gradus to indicate that people mastered agriculture only in a later phase (1.2.16: inferiore gradu aetatis susceptam agri culturam [sc. ab hominibus]: see above n. 15). It may be instructive to compare these passages with the well-known — but difficult — passage in the De lingua Latina where he speaks of the four gradus etymologiae (Ling. 5.7–9): Now I shall set forth the origins of the individual words (singulorum verborum origines), of which there are four levels of explanation (quattuor explanandi gradus). The lowest (infimus) is that to which even the common folk has come (venit); who does not see the sources of argentofodinae ‘silver-mines’ and of viocurus ‘roadoverseer’? The second is that to which old-fashioned grammar has come (quo grammatica descendit antiqua), which shows how the poet has made (finxerit) each word, how he has composed (confinxerit) each word, and how he has derived (declinarit) each word. Here belongs Pacuvius’s rudentum sibilus ‘the whistling of the ropes’, here his incurvicervicum pecus ‘incurvate-necked flock’, here his clamide clupeat bracchium ‘with his mantle he beshields his arm’. (8) The third level is that to which philosophy ascended (ascendens pervenit), and on arrival began to reveal the nature of those words which are in common use, as, for example, from what oppidum ‘town’ was named, and vicus ‘row of houses’, and via ‘street’. The fourth is that where the sanctuary is, and the mysteries †of the king† (adytum et initia †regis†): if I shall not arrive there (quo si non perveniam) *** at full knowledge (scientiam),32 at any rate I shall cast about for a conjecture (opinionem), which even in matters of our health the physician sometimes does when we are ill. (9) But if I have not reached the highest level (summum gradum), I shall nonetheless go farther up (praeteribo) than the second, because I have studied not only by the lamp of Aristophanes, but also by that of Cleanthes. I have desired to go farther (praeterire) than those who expound only how the words of the poets are made up. For it did not seem appropriate that I seek the source in the case of the word which Ennius had made, and neglect that which long before King Latinus had made, in view of the fact that I get pleasure rather than utility from many words of the poets, and more utility than pleasure from the ancient words. And in fact are not those words mine which have come to me by inheritance from King Romulus, rather than those which were left behind by the poet Livius? Varro maintains that there are quattuor explanandi gradus of the origines singulorum verborum. The infimus, quo populus etiam venit, offers an explanation of simple compounds. The second, quo grammatica descendit antiqua, deals with words created by the poets. The third instance is that of philosophy, that is, the philosophical tradition that has concerned 32 It would seem necessary to fill the supposed lacuna before scientiam with a negation and a verb governing the noun.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 17

14/11/2017 14:56:11

18

BICS-60-2  2017

itself with the science of language, such as, above all, the Stoic tradition, which mostly treats commonly used words. The fourth instance is mysterious and the formulation referring to it is very difficult. Varro wants to go over the methods of the second instance of ancient grammar, in a sort of reconciliation between Alexandrian grammar and Stoic philosophy (quod non solum ad Aristophanis lucernam, sed etiam ad Cleantis lucubravi). He also wants to discuss commonly used words. It would not make sense to seek the origin of the words used by Ennius and to disregard those by the ancient King Latinus, since poetic words are the sources of voluptas, common words of utilitas (cum poeticis multis verbis magis delecter quam utar, antiquis magis utar quam delecter). The words inherited from Romulus do not belong to the linguistic heritage any less than those handed down by the poet Livius Andronicus. The passage has been studied assiduously, and we cannot examine here all the details and problems related to it.33 I do, however, believe that the term gradus is pertinent to our discussion, as are the verbs indicating movement (venire, pervenire, praeterire), especially descendere (secundus (sc. gradus) quo grammatica descendit antiqua) and ascendere (tertius gradus, quo philosophia ascendens pervenit). It seems clear to me that the overall scheme is essentially the same as that of the progression of the vitae humanae in the Rust., with subsequent and contiguous gradus (each following gradus presupposes the previous one and is, in a manner of speaking, its product, just as it, in turn, is the prerequisite of the next one). The order is progressive and thus chronological: ancient grammar precedes philosophy. The resulting chronology is not connected with the age of the material being studied (the common lexicon, poetic language), but with the methods, the ratio, used to study it (the grammatica antiqua assumes the existence of a ‘modern’ counterpart). If we compare this section with that in the Rust. (about ten years later), we also see how the descendit referring to the second instance of the etymology does not imply any inferiority vis-à-vis the previous stage. This verb is used also in the case of the humanae vitae (see above, p. 12) and merely indicates a progression.34 The word does therefore not require emendation, as some have suggested.35 It is, however, true that the process runs from infimus to summus gradus and therefore describes not only the succession of methods but also their progress (as in the humanae vitae). Along these lines, ascendens, which characterizes philosophy, seems to connote progress in the leap forward from Alexandrian grammar to Stoic philosophy. Here Varro uses a reversed version of the chronologicalprogressive model of Rust. 2.1, where the summus gradus is the most ancient and natural,36 with a more marked sense of progress. The summus gradus in Ling. is indeed the fourth, while the infimus is the opening and most basic one. One might conclude either that in the model of rustic life there was after all some residual basic idea of a decline vis-à-vis antiquity, or, what I think is more probable, that the four etymological instances point towards the future, and in particular towards Varro’s own treatise, allowing for the possibility of further See Piras 1998: 57–71 and 2015: 53–57 and Blank 2008: 69–71; on gradus and descendere see also Nelsestuen, p. 25, in this volume.

33

34 On the value of descendere in Rust. 2.1.3–5 cf. Cardauns 2001: 22: ‘descendisse […] ist hier nicht im Sinne einer moralischen Deszendenztheorie gebraucht, die Varro hier fernliegt, sondern neutral’; Saunders 2001: 243 n. 8 (‘purely chronological’); Schütrumpf 2001: 257–58 and n. 13. 35 Escendit Schioppius (Caspar Schoppe 1602), adopted by Kent 1938 and Collart 1954; Andreas Spengel proposed ascendit.

Schütrumpf 2001: 258 n. 13, notes the uncommon usage of summus here, usually indicating the ‘letzte Stufe’, and suspects it originates from the image of ‘Abstieg’ (descendisse).

36

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 18

14/11/2017 14:56:11

GIORGIO PIRAS: DICAM DUMTAXAT QUOD EST HISTORICON

19

progress (quo si non perveniam […] opinionem aucupabor […] quodsi summum gradum non attigero, tamen secundum praeteribo). In Rust., on the other hand, Varro seems more interested in the process which began in the very distant past (summa memoria) and has in view no stage later than the contemporary status quo. As in Rust., the stages of development can be divided into two main sections: a threefold division of the past, and a present, with the implied possibility of a future. The examples in paragraph nine on the creators of poetic and common language give rise to further reflections on Varro’s attitude to the past.37 Ennius is the poet most often cited in the Ling., which can be explained by the high regard that Varro had for the national Roman poet of the late Republic. He is compared with rex Latinus, a king before the rise of Rome, the first king of the Latins:38 ‘For it did not seem appropriate that I seek the source of a word which Ennius had made, and neglect that which long before King Latinus had made.’ After the antithesis between voluptas and utilitas (cum poeticis multis verbis magis delecter quam utar, antiquis magis utar quam delecter), which is quite frequent in Varro (cf. Baier 2007), the poet/king opposition is repeated in a chiasmus in the rhetorical question which follows. Romulus qua name-giver and Livius Andronicus stand for common and poetic language respectively (‘and in fact are not those words mine which have come to me by inheritance from King Romulus, rather than those which were left behind by the poet Livius?’). In the first pair (Ennius/Latinus), the antithesis also has a temporal dimension: ante rex Latinus […] cum […] verbis […] antiquis. Rex Latinus also represents antiquitas relative to Ennius. Certainly Varro was aware of the significant time gap between Romulus and Livius Andronicus, but the gap between Latinus and Ennius was even more significant, since Latinus dates from before Romulus and the founding of Rome. The founding coincided almost exactly with the first Olympiad on the cusp of the mythical and historical eras, as we pass from Latinus’s era in which multa fabulosa referuntur to that in which res gestae veris historiis continentur. It is notable that Latinus belongs to the last part of the mythical era, which it is not completely impossible to reconstruct (cf. p. 14 above, and Censorinus 21.2). This contrast also falls within another trend that has been observed in Varro, namely that of using the concept of antiquitas in a relative way, without an exact chronological specification.39 Latinus, and perhaps also Romulus, are antiqui when compared with Ennius and Livius Andronicus, as well as with Varro himself. The contrast between past and present is indeed ubiquitous in Varro, but in several passages one may also note a threefold temporal gradation, namely antiquissimum/antiquum/nunc, a distinction between the remote past, the near past, and the present. Thus, e.g., in Ling. 7.3, where there is a discussion of the antiquity of poetic words, it is said that the Carmina Saliorum, the first attestation of verba poetica latina, date back seven hundred years to the era of Numa, unless they are from an even earlier age (neque ea ab superioribus accepta). So here too there is the distinction between past and remote past, and an attempt at precise dating to Numa’s reign is only made in the historical era, in the case of the origin of the Carmina Saliorum. Immediately thereafter Varro adds that in the reconstruction of the past one might reproach someone who cannot identify the atavus (‘great-grandfather’s grandfather’) or 37

Cf. Piras 2015: 56–57.

Cf. gpR, fr. 29 ‘Troy was overthrown […] And this was done in the reign of Latinus the son of Faunus, from whom the kingdom began to be called Latium instead of Laurentum.’

38

Romano 2003: 112–13. Cf., e.g. Ant. hum. 17.1 ‘Varro thought that the Romans borrowed the names of their months from the Latins. He demonstrated in quite a plausible manner that the persons who created (auctores) these names are older (antiquiores) than the city of Rome.’

39

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 19

14/11/2017 14:56:11

BICS-60-2  2017

20

tritavus (‘great-great-grandfather’s grandfather’) of a hero. But such a reproach would be unjust, since even in regular life it is often the case that one cannot determine the mother of one’s tritavus, and ‘this interval is much closer to us than the stretch from the present time to the beginning of the Salians, when, they say, the first poetic words of the Romans were composed in Latin’. In this case as well we have comparison between present (the writer, the interlocutor) and past (the hero on the one hand and the tritavus on the other) and remote past (the hero’s distant ancestors or the female ancestor of the tritavus, who is even more difficult to trace because of the complexities of the Roman genealogical system). We have already seen (p. 10) in the proem to book 5 the difficulties in recognition caused by a gap of two generations, tertium seculum non videt eum hominem quem vidit primum. In the same passage it is said that even maiores, the Romans of several generations previously, had in their turn a remote past which was hard to access and reconstruct, ‘those things that oblivion has snatched away even from our ancestors, the painstaking labours of Mucius and Brutus, although they have pursued the runaways, cannot bring back’ (5.5).40 The Romans who were already ancient from Varro’s standpoint in fact lived closer to the boundary between myth and history. On this basis, the examples in Ling. of a tripartite division of time across present, past, and remote past, seem coherent in their own right, and comparable (while not coinciding precisely) with models of time in the other treatises which we have discussed, from the tria discrimina temporum as reported by Censorinus to the three stages of human history borrowed from Dicaearchus. For Varro, uncertainty about the ultimate origo and progression by successive stages are features of the past which need to be explored through a genealogical-reconstructive method. This method is oriented backwards and substantially inductive, based on specific traces and clues, often the testimonies of the veteres, in particular the poets, even though these testimonies are not always completely reliable. The research and the reconstruction proceed step by step, determining events and data (synchronism, eras, forms or significances of words) with the goal of extracting as far as possible what is ‘buried by the lapse of time’ (vetustas: Ling. 6.2), but always with awareness of the difficulties of getting beyond what veriis historiis continentur. Sapienza, University of Rome

Note that Mucius is Mucius Scaevola I, mentioned in De oratore as promoter of the collection of annales maximi, and Brutus is the jurist Iunius Brutus, also mentioned in De oratore (and see above, p. 9). 40

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:49:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 20

14/11/2017 14:56:11

VARRO, DICAEARCHUS, AND THE HISTORY OF ROMAN RES RUSTICAE 1 GRANT A. NELSESTUEN Abstract: This article reconstructs the historical narrative underlying the account of Roman farming in De re rustica, thereby shedding new light on Varro’s intellectual engagements in composing his agricultural treatise and methodology as a scholar of Rome. Central to this account is Dicaearchus of Messana’s three-stage theory of human development, which not only provides an anthropological pedigree for agriculture (in book 1) and animal husbandry (in book 2), but also establishes the theoretical framework for Varro’s innovative treatment of so-called ‘animal husbandry of the villa’ (in book 3). By presenting this last sphere of rustic production as a newly viable mode of farming contingent on the vast wealth and decline of morality in contemporary Rome, Varro uses it to emblematize ‘urban life’ as a fourth stage of human existence. The history of Roman farming thus offers a profoundly ambivalent narrative, entailing technological and material progress at the cost of moral decline. Keywords: Varro, Dicaearchus, farming, morality, Roman history, technology, anthropology, Porphyry, city

Published around 37 bc, Varro’s De re rustica (Rust.) presents an account of contemporary Roman farming in three dialogues and as construed via three spheres of production: field cultivation (agri cultura) in Rust. 1, animal husbandry (pastio agrestis) in Rust. 2, and socalled ‘animal husbandry of the villa’ (pastio villatica) in Rust. 3. At various points, Varro and his interlocutors comment on the quality of these practices, often using ancestral ones as benchmarks for their sometimes disparate estimations.2 In itself, this recourse to the past to judge the present is unsurprising given Roman dedication to the mos maiorum and Varro’s paramount interest therein.3 Yet, what may be surprising is that Varro, despite his keen interest in origins and history elsewhere,4 does not provide a dedicated account of the beginnings, development, and, as we shall see, apparent decline of Roman farming, instead leaving this history to be gleaned from a handful of scattered passages. By analysing them and exploring their intellectual debts and innovations, this article reconstructs the historical narrative underlying De re rustica’s account of Roman farming practices (i.e. res rusticae).5 In so doing, it sheds new light on one of Varro’s specific intellectual engagements in composing a treatise de rebus rusticis and, to an extent, his methodology as a scholar of I would like to thank Valentina Arena, Fiachra Mac Góráin, and the anonymous readers for their astute comments and help. I have used Flach’s 2006 edition of De re rustica. All translations are mine.

1

2 See, e.g., 1.7.2, 2.8.3, and 3.12.1. On the composition of Rust., see Martin 1971: 213–35; White 1973: 482–88; and Flach 1996: 7–15. On the author and interlocutors, see Kronenberg 2009: 33–35, 76–93; Nelsestuen 2011: 315–51; and 2015: 9–30.

On the mos maiorum, see especially Wallace-Hadrill 1997: 3–22; Blösel 2000: 25–97; Bettini 2011: 87–130; and Arena 2014: 217–38. Cf. Cic. Acad. 1.9, and Todisco in this volume.

3

4

See, especially, Della Corte 1976: 111–36 and Rawson 1985: 233–49. See also Piras in this volume.

For Varro’s conceptualization of res rusticae as an ars and scientia, see Nelsestuen 2015: 31–72. Building on the observation that a Dicaearchus-inspired ‘temporal dimension’ (49) underlies Rust.’s analytical definitions for the three subfields of res rusticae, this article newly argues for an understanding of the vita urbana as a fourth stage of human existence. 5

BICS-60-2  2017

21 © 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms

BICS 60-2 Nov14-2 FINAL-FINAL.indd 21

14/11/2017 14:56:12

BICS-60-2  2017

22

res Romanae. Through close readings of the second and third prefaces, section 1 unravels the evolution of rustic production over the course of Rome’s history. Section 2 then examines Varro’s engagement with the three-stage theory of human development attributed to the fourth-century bc Peripatetic philosopher Dicaearchus of Messana. Providing an anthropological pedigree for agri cultura and pastio agrestis, this account also establishes the theoretical framework for Varro’s signature contribution to the field of res rusticae: the systematic explication of pastio villatica as a discrete form of production (3.1.8). Section 3 turns to this innovative treatment in Rust. 3. By presenting the husbandry of birds, small game, and fish as a new mode of farming contingent on both the contemporary influx of wealth into the city of Rome and its attendant decline in morality, Varro uses pastio villatica to emblematize the vita urbana as a fourth stage of human existence. For Varro, then, the history of Roman res rusticae offers a profoundly ambivalent narrative, which entails technological and material progress at the cost of moral decline. 1. The history of res rusticae at Rome De re rustica’s underlying history of Roman farming emerges from a synthetic reading of the authorial prefaces to the second and third books. For the aid of the reader, I offer the following preliminary conspectus: 1.) Before the city’s founding, proto-Romans pursued a pastoral lifestyle. 2.) Upon the city’s founding, Romans practised an undifferentiated form of agri cultura, which entailed the indiscriminate use of land for grazing and cultivation. 3.) Once wealth had reached a sufficient level, the proper distinction between pastor and agricola and, hence, between pastio agrestis and agri cultura emerged. 4.) Eventually, this distinction became muddled due to the illicit conversion of fields into pastures. 5.) Thereafter — and seemingly as a further consequence of 4 — Romans began to eschew country farming for urban living. 6.) A third form of rustic production, pastio villatica, was also practised at Rome, though its point of origin remains unspecified and its proper identification has, so Varro claims, gone unnoticed heretofore. After lamenting the declined state of Italian agriculture (5), the abandonment of fields and vineyards for villas and the city, and the resulting reliance on merchants for imported foodstuffs (2. pref. 1–3), Varro provides a historical rationale for this state of affairs which doubles as a précis of the chronology of res rusticae at Rome (2. pref. 4). Distinguishing between three generations of Roman practitioners of res rusticae — shepherds (pastores), their descendants (sua progenies), and the descendants of the latter group (progenies eorum) — Varro maintains that this first group herded animals for their livelihood (1), while the second received instruction in the art of agri cultura from their pastoral forefathers (2). Though he does not identify either group’s members, Varro is referring to the tradition in which the shepherd Faustulus played the part of foster father to Romulus and Remus, who were likewise herdsmen before they founded the city of Rome.6 While Varro does not specify that Faustulus taught his foster sons the art of field cultivation, he does credit Romulus with the allocation of farmland to fledgling citizens in Rust. 1 (1.10.2),7 which 6

Cf. Rust. 2.1.9.

On Romulus’s assignations, see Gabba 1979: 55–63; Flach 1996: 260; Wiseman 2009: 81–98; and Roselaar 2010: 20–22. Cf. Dion. Hal. Ant. Rom. 2.7.4, 2.28.3.

7

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 22

14/11/2017 14:56:12

GRANT A. NELSESTUEN: VARRO, DICAEARCHUS, AND ROMAN RES RUSTICAE

23

presumably enables them to put the agricultural precepts of the pastores into practice.8 Finally, the third group converts grain fields into pasturelands out of greed (avaritia), against laws, and due to ignorance of the proper distinction between agri cultura and pastio (4). Usually identified as those Romans from the second century onwards, whose misuse of public land (ager publicus) included the pasturing of animals on a scale and in places otherwise not legally permissible,9 this group bears partial responsibility for the failure of Italian agriculture to satiate contemporary Romans. Thus does the preface to Rust. 2 establish the basic chronology for the history of Rome’s res rusticae. Returning to this history in the preface to Rust. 3, Varro expands on his reasoning and integrates the third sphere of pastio villatica (3.1.7–8). He explains that the shepherds’ descendants ‘initially’ (primo) treated agri cultura as an ‘undifferentiated’ (indiscreta) practice and concept due to their ‘poverty’ (paupertas) and indiscriminate use of the ‘same field’ (idem ager) for both arable cultivation and animal husbandry (2).10 At some point ‘thereafter’ (postea), they ‘made a distinction’ (diviserunt) between agricola and pastor (3). The reason for this linguistic development was, so Varro elliptically claims, greater ‘wealth’ (pecunia): whereas paupertas forced early Romans to use the same land for both cultivation and grazing, an increase in pecunia presumably enabled the farmer to allocate specific parts of his holdings to specific enterprises or to focus solely on one or the other practice, which in turn prompted the practice of agriculture on fields and animal husbandry on pastures and, consequently, the distinction between the agriculturalist and the shepherd.11 In this way, Varro provides material and linguistic rationales for two of the three genera of res rusticae. Thereafter, Varro claims that pastio is a ‘twofold’ (duplex) matter: ‘animal husbandry on pastureland’ (pastio agrestis) and ‘animal husbandry at the villa’ (pastio villatica) (6). Unlike the former type, which is ‘well known and noble’ (nota et nobilis) and has frequently been practised by wealthy men, the latter ‘is seen as lowly’ (humilis videtur), has traditionally been included under the heading of agri cultura, and ‘as a whole has not been distinguished separately’ (neque explicata tota separatim). This final claim sets the stage for Rust. 3’s intellectual programme, which is a dedicated and systematic account of the raising of birds, small game, and fish at the villa for sale as luxury foodstuffs.12 That pastio villatica is viewed as ‘lowly’ (humilis), yet trades in the sumptuous, initially appears paradoxical, but its humble reputation, as we shall see in section 3, turns out to be a matter of antiquated thinking and muddled perception (cf. videtur), and the dialogue itself ultimately rationalizes the pursuit of luxury produce as a viable farming practice on material and linguistic grounds. Thus, by identifying pastio villatica as a discrete category of rustic production, Varro not only begins to articulate the third sphere of Roman res rusticae (6), but even insinuates himself into the tradition of differentiation begun by the shepherds’ descendants. Marked by the advent of new practices, evolving concepts, and changing material circumstances, Varro’s account of Roman res rusticae is thus one of initial development (1– 3) and more recent decline (4–5). Ignoring for now this latter turn (4–5) and the seemingly 8 Presumably instability or a lack of resources (e.g., land) precluded the pastores from arable cultivation. Cf. Flach 1997: 181–82. 9

See Noe 1977: 292–96; Gabba 1988: 134–42; Flach 1997: 182; and Roselaar 2010: 173–79.

10

For ‘mixed intensive farming’, see White 1970: 50–52.

I.e. once the farmer has increased ‘wealth’ (e.g., two parcels of land or a greater number of animals), he divided up his estate or specialized: hence, the impetus for linguistic distinction. Cf. 3.3.1–10.

11

For pastio villatica, see Rinkewetz 1981 and Purcell 1995: 151–79. On Varro’s systematic structures of arrangement, see Piras in this volume. 12

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 23

14/11/2017 14:56:12

24

BICS-60-2  2017

indeterminate historical position of pastio villatica (6), it is worth observing that the gradual assignment of particular types of res rusticae to distinct agents (i.e. 1–3) resembles ‘specialization’, or a ‘division of labour’, in modern economic terms. In espousing a rudimentary form of this concept, Varro is not alone in antiquity, for historians of economic thought have long recognized its inchoate presence in the Greek intellectual tradition: particularly in Plato’s Republic (369b5ff.), but also in Xenophon’s Cyropaedia (8.2.5–6).13 But unlike his Greek precursors, who emphasize the economic (and political) benefits that are a consequence of specialization, Varro focuses on economic viability as its precondition. On its own, this difference in focus may not matter so much, for one might imagine that such economically contingent specialization could procure even more economic benefits, though Varro does not explicitly make this claim here. Instead, his interest in the priority of material wealth for specialization stems from a categorically different concern; rather than examining the allocation of resources and the division of labour in a ‘political economy’, Varro seeks to situate the origins and development of Roman res rusticae within the more universal scope of human history. For this larger, what we might call ‘anthropological’ dimension, we must turn from the prefaces to the second dialogue. As we shall see, this account, which explicitly derives from Dicaearchus of Messana, in turn provides further context for understanding the various claims that Varro makes in the second and third prefaces as well as for clarifying his view of res rusticae in contemporary Rome. 2. Varro, Dicaearchus, and the three stages of human development In the first dialogue, Varro cites Dicaearchus’s (non-extant) Bios Hellados for the idea that pastoralism preceded agriculture (Rust. 1.2.16 =F 55 Mirhady =F 51 Wehrli).14 But it is not until the second dialogue that Varro provides a more robust explanation of that claim. After agreeing to speak on the ‘origin’ (origo) and ‘dignity’ (dignitas) of pastio agrestis (2.1.2), Varro turns to Dicaearchus (2.1.3–5 =F 54 Mirhady =F 48 Wehrli) to recount how human life ‘came down from earliest memory by stages into this present age’ (a summa memoria gradatim descendisse ad hanc aetatem). The ‘earliest stage’ (summus gradus) was the ‘natural’ (naturalis) one, in which humans procured their sustenance by gathering those fruits that ‘the untouched earth bore spontaneously’ (inviolata ultro ferret terra). In the ‘second’ (secundus) stage, humans continued to collect uncultivated produce, but now also kept animals for their dietary and bodily needs. In this respect, the ‘pastoral’ (pastoricia) stage offers a new mode of livelihood, which introduces the new foodstuffs of milk and cheese and the use of animal skins for clothing. In the ‘third’ (tertius) stage, humans ‘at last’ (denique) practised agri cultura, but Varro is also quite clear that ‘many’ (multa) practices On Plato, see Lowry 1987: 16–18 and 1998: 21–26; and Silvermintz 2010: 747–72. On Xenophon, see Finley 1973: 135 and Lowry 2003: 18. 13

For surveys of Bios Hellados, see Wehrli 1967: 526–34; Bodei Giglioni 1986: 629–52; and Schütrumpf 2001: 255–77. On its three-stage scheme, see Reischl 1976: 90–96 and Saunders 2001: 237–54. For Bios Hellados’s influence on Varro’s De gente populi Romani and De vita populi Romani, see Della Corte 1976: 126–36; Gabba 1991: 98–101; Ax 2001: 279–310 [=2000: 337–69]; and Purcell 2003: 329–58. For Dicaearchus’s influence on Roman political thought, see Arena 2012: 83–85, 109; and McConnell 2014: 115–60. Note that Dicaearchus is not included in Varro’s list of fifty-two authors (1.1.7–11) — presumably because he does not treat farming per se. Note also that Varro’s famous tripartite division of time — ‘obscure’ (adelon), ‘mythical’ (mythikon), and ‘historical’ (historikon) — does not appear in Rust. For this scheme, see Censorinus DN 21.1–5; Fraccaro 1907: 255–57; Dahlmann 1935: 1172–77; Morgan 1974: 117–28; Grafton–Swerdlow 1985: 454–65; Ax 2001: 300–02; Feeney 2007: 77–82; and Piras in this volume. 14

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 24

14/11/2017 14:56:12

GRANT A. NELSESTUEN: VARRO, DICAEARCHUS, AND ROMAN RES RUSTICAE

25

of the prior two stages were retained; humans thus acquired a third way to procure their livelihood to go along with the previous two. The gathering of wild produce, the tending of animals, and field cultivation — so does Varro present an account of human development in terms of the gradual (cf. gradatim) acquisition of new modes of securing sustenance. Insofar as it is presented in Rust. 2, Varro’s rendition of Dicaearchus’s account achieves its immediate purpose, which is to provide a historical ‘origin’ (origo) of animal husbandry motivated by the ‘utility’ (utilitas) it offers to humans — a motivation that accords with practising animal husbandry in the present as well (2.1.11; cf. 1.4.1, 3.3.1). Still, numerous questions about Varro’s account abound. For one thing, we might wish to consider its overall teleological thrust. Is Varro providing a so-called progressivist15 narrative of human development, which traces the technological advances in modes of procuring food and their increasingly ameliorative effect(s) on human life? Or is there a primitivist element to this account, which freights those innovations with an attendant decline in the material, physical, and/or moral conditions of human existence? Further complicating these questions is the relationship of Varro’s account to Dicaearchus’s original sentiments in Bios Hellados. Has he faithfully recounted what Dicaearchus originally wrote? Or does he manipulate or even distort his predecessor’s thoughts for his own purposes? Disparate are the answers that scholars have put forward and the debate largely depends on the weight placed on each of the two primary pieces of evidence for Dicaearchus’s lost work: Rust. 2.1.3–5 and a fuller, but possibly tendentious, account in Porphyry’s De Abstinentia ab Esu Animalium 4.2.1–9 (=F 56A Mirhady =F 49 Wehrli). Lastly, the force of the passage’s final sentence (2.1.5) remains somewhat opaque: what does Varro mean when he says that, after humans ‘had come into [the agricultural] stage, there they went forth for a long time until it came to our age (ad nos)’? The valence of Varro’s account — i.e. whether it is progressivist or primitivist — is surprisingly difficult to ascertain, for a number of reasons. For one thing, Varro is completely reticent about the quality of human life in all three stages: he simply does not provide any details or evaluative statements about the effect(s) of these various livelihoods on humankind’s existence. For another, the vocabulary provides little in the way of help. The primary terms for the succession of the stages, particularly descendisse and summus, do have the potential to impute evaluative or moralistic connotations, but they are also standard terms for describing the passage of time and should thus be understood in a ‘neutral’ sense in the present context.16 While the earth’s description in the natural stage as ‘untouched’ (inviolata) and bearing fruits ‘spontaneously’ (ultro) could imply a degree of force or violence in the agricultural stage’s acts of ploughing and sowing, this description would seem to be prompted by Dicaearchus’s rationalization of the Hesiodic Myth of Ages, in which fields are said to have borne fruit ‘ungrudgingly’ (ἄφθονον) and ‘of their own accord’ (αὐτομάτη).17 Moreover, farmers are later said to have been upheld by the maiores as ‘leading a pious and useful life’ (piam et utilem agere vitam, 3.1.5) and ‘the sole remnants of the offspring of King Saturn’ (solos reliquos […] ex stirpe Saturni regis). Or in the case of keeping animals for milk, cheese, and clothing, there is nothing particularly moralistic or pejorative in Varro’s description. In fact, what may be the pastoral stage’s most 15

For progressivism and primitivism, see Lovejoy and Boas 1935; Edelstein 1967: 35–51; and Dodds 1973: 1–25.

So Reischl 1976: 93–95; Saunders 2001: 243 n. 8; and Schütrumpf 2001: 257–58. Cf. OLD s.v. summus 2.d. See also Piras, p. 18, in this volume. 16

See Boys-Stones 2001: 14–16; Saunders 2001: 246–49; and Schütrumpf 2001: 260–61. For the spontaneous generation of vegetarian foodstuffs, see Campbell 2003: 341–43.

17

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 25

14/11/2017 14:56:12

BICS-60-2  2017

26

objectionable aspect from a primitivist perspective — namely, the slaughter of animals for meat-eating18 — receives no mention and is largely glossed over in the remainder of the dialogue as well.19 In short, there is little in 2.1.3–5 to suggest that Varro associates a decline in morality or some adverse change in human existence with the advent of pastio or agri cultura, i.e., that he is espousing a primitivist view of human history. Conversely, the gradual assumption of new livelihoods, a consequently greater variety of foodstuffs, and the preservation of aspects of the previous stage(s) in each successive one would all seem to imply that Varro is taking, if anything, a moderately progressivist view of human development,20 which focuses on the technological advances in procuring sustenance from res rusticae. Consideration of Dicaearchus’s original sentiments in Bios Hellados may shed further light on the nature of Varro’s account. To this end, it is necessary to turn to Porphyry, the third-century ad Neoplatonic philosopher, whose evidence (De abstinentia 4.2.1–9 =F 56A Mirhady =F 49 Wehrli) corroborates a number of Varro’s aspects, calls others into question, and has prompted strenuous debate over the value of each source for reconstructing Bios Hellados.21 Structurally, Porphyry’s version accords with Varro’s. There are three stages of human development: the earliest one, in which the ‘Golden Race’ (χρυσοῦν γένος, 4.2.2) lived off what grew ‘of its own accord’ (αὐτόματα, 4.2.3) and ‘did not yet possess any technē, agricultural or otherwise’ (μήτε τὴν γεωργικὴν ἔχειν πω τέχνην μήθ’ ἑτέραν μηδεμίαν ἁπλῶς). Then came the ‘pastoral life’ (νομαδικὸς βίος, 4.2.7), in which humans captured ‘harmless animals’ (τὰ ἀσινῆ) and attacked ‘destructive and dangerous ones’ (τὰ δὲ κακοῦργα καὶ χαλεπά).22 Finally there appeared the ‘agricultural form’ (γεωργικὸν εἶδος, 4.2.8) of life, which, in keeping with Porphyry’s focus on humans’ treatment of animals and like the one in Varro’s account, receives little attention except insofar as it constitutes a ‘third’ (τρίτον) stage of life. Still, a brief aside about the motivation for the shift from the second to third stages reveals another similarity between Varro’s and Porphyry’s accounts: just as the pursuit of utilitas motivates humankind’s assumption of pastoralism in Varro’s recounting, so τὰ χρησίμα — or, at least, human perception of what seemed useful (cf. κατανοοῦντες ἀεὶ τῶν χρησίμων εἶναι δοκούντων, 4.2.8) — prompted the transition from the pastoral to the agricultural stage and, by implication (cf. ἀεὶ), from the natural to the pastoral one as well.23 In sum, both Varro and Porphyry recount a three-stage development of human life in which the pursuit of ever greater utility motivates each transition.

For meat-eating as a marker of generational decline, see Empedocles DK31 B128 [=Porph. De Abst. 2.20] and Arat. Phaen. 129–36. On ancient vegetarianism, see Newmyer 2011: 97–111.

18

Mention of animal skins could imply the pastoralists’ destruction of animal life, but the omission of meat-eating within a history of humankind’s food systems remains striking. References to animal slaughter in Rust. 2 are sparing: pigs (2.4.3–22) and bovines (2.5.3–4, 2.5.11). 19

20

Cf. Reischl 1976: 94–96.

Those scholars more trusting in Porphyry’s account include Cole 1967: 54–55; Bodei Giglioni 1986: 629–52; and Schütrumpf 2001: 255–77. Conversely, Wehrli 1967: 56; Della Corte 1976: 128–30; Reischl 1976: 90–96; Ax 2001: 287–89; and Saunders 2001: 237–54 all take critical views. 21

For Porphyry’s pro-vegetarian distortion of Dicaearchus, see Saunders 2001: 241–54. For early humans’ conflicts with animals, see Plat. Prot. 322a–c; Polyb. 6.5.4–6.12; Cic. Rep. 1.40; Diod. Sic. 1.8.2; and Sen. Ep. 90 [=Posidonius F 284 E-K]; and for discussion, see Campbell 2003: 227–30 and Newmyer 2011: 105–08. 22

23

Pace Baier 2007: 263.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 26

14/11/2017 14:56:12

GRANT A. NELSESTUEN: VARRO, DICAEARCHUS, AND ROMAN RES RUSTICAE

27

But the two versions diverge markedly in their tenors regarding the effects of each new mode on human existence. Whereas Varro offers a seemingly progressivist account devoid of evaluative statements on each period’s physical and moral conditions, Porphyry provides numerous details about the quality of human life and his narrative is decidedly one of decline. For the most part, he frames his discussion in terms of antitheses between the earliest humans and contemporary ones. Owing to the ‘scarcity’ (σπάνις, 4.2.4) of their food supply, the former did not produce ‘excesses’ (περιττώματα), remained pure and without disease, and spent their lives in leisure, friendship, and peace with one another (4.2.5).24 In contrast to them stand present humans (4.2.1), whose suffering from disease began with the dietary introduction of meat. Moreover, wars and social discord arose because possessions — presumably in the form of herds — came to be regarded as worthwhile (4.2.8; cf. 4.2.5), which prompted the rise of individual ‘ambition’ (φιλοτιμία) through defending or stealing property.25 For Porphyry, Dicaearchus’s Bios Hellados powerfully demonstrates that ‘abstinence from animals contributed no less than other things’ (οὐχ ἧττον τῶν ἄλλων καὶ ἡ ἀποχὴ τῶν ἐμψύχων συνεπλήρου, 4.2.9) to the earliest people’s ‘blessed life’ (μακαρίος βίος) — one without the war, ambition, ‘greed’ (πλεονεξία), ‘luxury’ (τρυφή), and ‘injustice’ (ἀδικία) that define contemporary existence. Though it acknowledges advances in human technē, Porphyry’s version presents Dicaearchus’s account as a primitivist one in no uncertain terms. The points of convergence between Varro’s and Porphyry’s testimonies confirm that the core of Dicaearchus’s account consists of three stages motivated by human pursuit of greater utility. But how are we to negotiate between the different trajectories of each version? Appeals to each work’s context offer some help. In the case of Porphyry, the polemic of De abstinentia has fundamentally coloured his presentation of Dicaearchus’s account.26 In particular, Porphyry marshals his discussion in book 4 against two commonly held propositions (4.1): 1.) that eating meat provides ‘benefits’ (τὰ συμφέροντα) to humans; and 2.) that neither a.) an entire people nor b.) any wise individuals have ever abstained from it. Thus, the existence of the earliest humans who lived without slaughtering animals supports Porphyry’s arguments against 2.a, while the contrast between their apparently beneficent state of life and the more calamitous and disadvantageous one of the latter two stages attacks 1. The result is that Porphyry’s account distorts in two directions: romanticizing what was actually a ‘minimally autarkic’27 state of existence for the earliest humans and minimizing any potential benefits (e.g. more food and of a better quality) that later stages introduced. Thus, if we scale back Porphyry’s hyperbole regarding the first stage and allow for a greater degree of progress in the latter two, which is detectable in Varro and even still manages to shine through in Porphyry’s account, it is not unreasonable to concur with those scholars who detect in Dicaearchus’s thought an ‘ambivalent theory of progress’28 — one that accommodates both progressivist developments in technē and material resources and primitivist anxieties over the concomitant rise of social evils and physical travails. 24

On the medical content, see Saunders 2001: 249–52 and Schütrumpf 2001: 275–77.

For the correlation of meat-eating with war, see Empedocles DK31 B128 [=Porph. De Abst. 2.20]; Arat. Phaen. 129–36; and Ov. Met. 15.75–142. For the advent of property entailing warfare, see e.g., Hes. Op. 156–201 and Plat. Crit. 120e–121c.

25

26

So Saunders 2001: 237–54. For an opposing view, see Schütrumpf 2001: 255–77.

27

Saunders 2001: 248.

28

Such is the formulation of Reischl 1976: 93. Cf. Ax 2001: 289–90.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 27

14/11/2017 14:56:12

28

BICS-60-2  2017

But Porphyry’s tendentiousness does not mean that Varro himself is a disinterested reporter. For, in the course of serving as a historical prelude to Rust. 2’s technical treatment of pastio agrestis, Varro’s account strips away the social and physical maladies that befell humans with the advent of the pastoral stage in order to focus on the technological advances in res rusticae. Moreover, if we understand Varro’s separate treatment of the dignitas of pastio (2.1.6–10) to have some degree of Dicaearchan inspiration,29 the social worth that he observes is accorded to famous pastores and their wealth would seem to have transformed the philotimia and pleonexia at the root of social ills in Dicaearchus into decidedly less pejorative phenomena. In other words, Varro’s interest in historical origins and his penchant for analysis — manifested herein by the division between origo and dignitas — obscure important counterweights to the progressivist thread in the Dicaearchan three-stage theory. But rather than effacing any traces of decline for uncritical veneration of res rusticae,30 Varro applies, as I shall argue in section 3, this aspect of Dicaearchus’s thought to pastio villatica in Rust. 3. Finally, Varro’s concluding statement — that humans ‘went forth there [i.e. in the agricultural stage] for a long time until it came to our age’ (ibi processerunt longe, dum ad nos perveniret)31 — merits comment. Crucial to understanding this statement is the phrase ad nos: could it mean ‘to us’ in the specific sense of ‘to us Romans’, or does Varro mean it in the sense of ‘to our age’ on the model of ad hanc aetatem (‘to this day’ 2.1.3)? Previous commentators are divided as to how to take this phrase,32 but it is best to understand ad nos as ‘to our age’ on contextual grounds.33 On this view, Varro relates the universalizing perspective of Dicaearchus back to contemporary existence while acknowledging the lengthy span of time (cf. processerunt longe) for the agricultural stage for humankind and, perhaps most crucially, hinting at a distinction between that stage and his own present day. By availing himself of Dicaearchus’s three-stage account of human development, Varro provides pastio and agri cultura with an anthropological pedigree. In so doing, Rust. 2.1.3–5 has several implications for understanding Varro’s historical account of Roman res rusticae. First, the introduction of agri cultura alongside pastio agrestis with the foundation of the city mirrors the progressive addition of field cultivation to pastoralism for Dicaearchus’s humans; in this sense, Rome would seem to offer a microcosmic exemplum of sorts for the universal experience of humankind. Second, Varro would seem to take the Dicaearchan association of technological development with increases in material resources one step further in his claim that early Romans practised an originally undifferentiated form of agri cultura that included pastio agrestis due to their poverty, but eventually made a proper conceptual distinction once their wealth increased. Finally, Varro’s extension of the Dicaearchan three-stage scheme to his own present day also suggests that he is actively bringing the theory of his Greek philosophical precursor to bear on Rome’s res rusticae. As we shall see, he continues to exploit elements of Dicaearchus’s account in Rust. 3 to present pastio villatica as a new mode of livelihood and a new stage of human existence, which is 29

For possible connections, see Schütrumpf 2001: 259–60.

30

So Reischl 1976: 90–96 and Schütrumpf 2001: 258–60.

31

For impersonal perveniret, see Keil 1889: 135 and Guiraud 1985: 86. Cf. 1.23.3.

For ‘to us Romans’, see Guiraud 1985: 14 and Flach 1997: 190. For ‘to this stage’, see Hooper–Ash 1934: 315 and Schütrumpf 2001: 257. 32

In addition to the earlier ad hanc aetatem (2.1.3), Varro subsequently glosses ad nos with etiam nunc (‘even now’) to demonstrate that contemporary life retains aspects of prior stages. Cf. Ling. 8.6.

33

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 28

14/11/2017 14:56:12

GRANT A. NELSESTUEN: VARRO, DICAEARCHUS, AND ROMAN RES RUSTICAE

29

contingent on the urbanization of Roman life, its increased material resources, and its moral decline. 3. Pastio villatica and the vita urbana of Rome Though it may not be immediately evident, the preface to Rust. 3 lays the foundation for Varro’s presentation of animal husbandry of the villa as a new mode of livelihood that is emblematic of a new stage of human existence. Observing that ‘country life’ (vita rustica) and ‘city life’ (vita urbana) have a ‘different origin in time’ (tempore diversam originem, 3.1.1), Varro explains that the former is far older given that ‘there was a time when humans inhabited the countryside and had no city’ (fuit tempus, cum rura colerent homines neque urbem haberent). To illustrate this claim, he adduces Thebes and Rome as the oldest cities in Greek and Roman territories respectively and founded some 2,100 and 700 years prior (3.1.2–3). Varro then compares these times to that ‘starting point, when fields began to be tended and humans dwelt in cottages and huts and did not know what a city wall and gate were’ (principium, quo agri coli sunt coepti atque in casis et tuguriis habitabant nec murus et porta quid esset sciebant, 3.1.3), concluding that ‘farmers predate city dwellers by an immense number of years’ (immani numero annorum urbanos agricolae praestant). Drawing on a variety of sources — including Ennius, his own work, and possibly Dicaearchus again34 — Varro thus ‘proves’ the antiquity of the vita rustica over the vita urbana. More importantly, by fixing the ‘starting point’ (principium) for the vita rustica at the time ‘when fields began to be tended’ (quo agri coli sunt coepti) and before cities were founded, Varro makes a hard and fast temporal distinction between it and the vita urbana. In other words, the beginning of agri cultura — i.e., the third stage of human existence in Dicaearchus’s account — initiates what Varro now calls the vita rustica, while the later founding of cities inaugurates the vita urbana. Given that the vita urbana has no parallel in Dicaearchus’s extant thought,35 the question thus arises: is Varro appending it as some sort of fourth stage in human existence to his predecessor’s threefold scheme? On the basis of this chronology’s framing via ‘modes of life’ (vitae), such an inference does not seem unreasonable. Still, that Varro does not explicitly call the vita urbana a ‘fourth’ stage is perhaps reason enough to remain cautious about formally denoting it as such. Nonetheless, it is worth considering whether the vita urbana merits de facto identification as a fourth stage according to Dicaearchan criteria. As we saw in section 2, the progression from one stage to another was marked by three features: a.) the acquisition of a new mode of procuring sustenance; b.) human pursuit of greater utility as the motivation for this technological innovation; and c.) the preservation of practices from the previous stage(s). Thus, the advent of animal husbandry in the second/pastoral stage (a) due to its utilitas (b) complemented the gathering of fruits already found in the first/natural stage (c), while the introduction of field cultivation in the third/agricultural stage (a) for the same purpose The idea that rustic life preceded urban life is pervasive in ancient anthropologies (e.g. Pl. Leg. 3.676b–c; Isoc. Pan. 4.39–40; Cic. De or. 1.36, Sest. 91; Vitruv. De arch. 2.1.1–3), but Varro explicitly cites Ennius (F 4.154–55 Skutsch) and twice uses the formula traditum est, which suggests engagement with particular sources, including the Chronica of Castor of Rhodes (FGrH 250 F 1.79 Jacoby), his own De gente populi Romani (F 1.4–5 Fraccaro), and Dicaearchus; see Della Corte 1976: 129–32, esp. n. 32; and Ax 2001: 290–302. Cf. Cole 2004: 355–422. Pace Thibodeau 2011: 89–92, pseudo-Plutarch’s Pro Nobilitate 20 is a sixteenth-century forgery and thus does not support his view that Varro used ‘Greek agronomists’ for 3.1.1–4; see Boscherini 1985: 651–60 and Blank 2011: 40–41.

34

35

So Schütrumpf 2001: 257. Cf. Feeney 2007: 113.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 29

14/11/2017 14:56:12

30

BICS-60-2  2017

of utility (b) saw the continuation of the prior stages’ practices (c). In the case of the vita urbana, the ongoing practice of pastio agrestis and agri cultura is consonant with the Dicaearchan notion of continuing previous practices (i.e. c) and would thus not present any challenge to its conceptual existence. And, as we saw in the case of the Romans (2.pref.4), it is only with the city’s foundation that they began to practise agri cultura alongside pastio agrestis. Yet, insofar as it is delineated in the third preface, the vita urbana would seem to offer neither a new mode of production (i.e. a) nor any promise of increased utilitas (i.e. b). Indeed, if the second preface is anything to go by, it would actually seem that these two criteria had become profoundly inapplicable to the vita urbana of contemporary Romans: not only has the failing state of Italian agriculture forced imports of provincial produce, but any concern for utilitas has given way to the pleasures of ostentatiously adorned villas and the urbane delights of the theatre and circus (2. pref. 1–3). So far, the vita urbana fails to fulfil the Dicaearchan criteria for technological development for humans along progressivist lines and would thus seem to constitute an alternative and debased mode of existence within the agricultural stage of human history. But it is here that the subject of Rust. 3 — pastio villatica — becomes relevant. After elaborating on the moral superiority of the vita rustica over the vita urbana (3.1.4–5) and supplying further evidence for the former’s greater antiquity (3.1.5–6), Varro offers the material and linguistic analyses of agriculture and animal husbandry (3.1.7–8) examined in section 1. As we saw there, Varro claims that the shepherds’ descendants practised an undifferentiated form of agri cultura, which entailed using the same parcel of land for both cultivation and pasturage, but that an increase in ‘wealth’ (pecunia) led to the linguistic distinction between agricola and pastor and the conceptual differentiation between agri cultura and pastio agrestis. When he finally introduces pastio villatica, Varro observes that it is at once distinct from pastio agrestis and has at the same time traditionally been considered part of agri cultura because of its ‘lowly’ (humilis) reputation. The intellectual project of Rust. 3, then, is to render pastio villatica ‘sufficiently distinguished’ (satis discreta) and ‘explained as a whole [and] in separate’ (explicata tota separatim), which the interlocutors will do in the dialogue. Yet, the ‘lowly’ (humilis) reputation of pastio villatica merits further consideration — especially given that this putative (cf. videtur) attribute stands in tension with the dialogue’s representation of pastio villatica as an elite practice associated with immense wealth and luxury.36 Varro has not articulated the rationale behind this claim, but a reconstruction on the model offered by agri cultura and pastio agrestis yields the following reasoning: just as pastio agrestis had been included under the ‘undifferentiated’ (indiscreta) concept of agri cultura, so pastio villatica has been ‘attached’ (adiecta) to agri cultura; the reason for the lack of distinction between agri cultura and pastio agrestis was the early practitioners’ poverty (paupertas), but an increase in their wealth (pecunia) motivated the differentiation; unlike pastio agrestis, however, pastio villatica has remained ‘insufficiently distinguished’ ([non] satis discreta) from agri cultura precisely ‘because it seems lowly’ (quod humilis videtur), though in point of fact it is not lowly and is a separate sphere of res rusticae. Varro does not specify the italicized portion, but it is a reasonable inference given the contextual logic. Moreover, the emphasis on ‘poverty’ and ‘wealth’ as the causal factors for the differentiation of agri cultura and pastio agrestis imparts a material sense to humilis and implies that an increase in wealth is prompting Varro’s project of articulating The elite nature of pastio villatica is best exemplified in the ostentatious estates of Hortensius (3.3.10, 3.6.6, 3.13.2, 3.17.5–8), Lucullus (3.3.10, 3.4.3, 3.17.9), Q. Fulvius Lippinus (3.12.1), and Varro himself (3.5.9–17, 3.13.1). Cf. the relevant entries of Shatzman 1975. 36

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 30

14/11/2017 14:56:13

GRANT A. NELSESTUEN: VARRO, DICAEARCHUS, AND ROMAN RES RUSTICAE

31

pastio villatica. As we shall see, its representation in the dialogue as a sphere of res rusticae that is contingent on increased wealth and urban luxury in contemporary Rome not only substantiates this explication of 3.1.7–8, but also establishes pastio villatica as a new form of rustic production for the vita urbana. Set on the day of the final pre-civil wars election of aediles in 50 bc, the third dialogue involves Varro and his friend Axius (‘Mr. Worth’), happening upon the augur Appius Claudius Pulcher (cos. 54) and his aviary of bird-named Romans at the Villa Publica (3.2.2).37 Included amongst this last group is a certain Cornelius Merula (‘Blackbird’), who provides the foundational definition of pastio villatica as a scientia of ‘how animals can be nourished or reared in or around the villa in such a way that they be a source of profit and delight for the master’ (qui in villa circumve eam animalia ali ac pasci possint, ita ut domino sint fructui ac delectationi, 3.3.1). Thereafter, he divides pastio villatica into three categories: aviaries (ornithones), animal pens (leporaria), and fishponds (piscinae). He also provides each with a historical pedigree in ‘two stages’ (bini gradus): ‘an earlier stage, which ancient frugality [offered], and a later stage, which later luxury added’ (superiores, quos frugalitas antiqua, inferiores, quos luxuria posterior adiecit, 3.3.6). Whereas the maiores only kept chickens in the farmyard and doves in makeshift cotes, contemporary owners have entire buildings larger than their ancestors’ villas and wholly dedicated to raising all kinds of fowl (3.3.6–7). The same trend applies to the other two genera: present leporaria are vastly larger and contain a greater variety of animals of a higher quality than past ones (3.3.8), while today’s saltwater piscinae with their exotic types of fish make the freshwater ones of yesteryear look no better than puddles of frogs (3.3.9–10). In assigning the cause for these changes to increased wealth and luxuria, Merula has provided a material explanation for the historical development of pastio villatica. Moreover, in the cases of aviaries and animal pens, there is also a concomitant linguistic shift: what were once called aviaria in Latin are now called ornithones in Greek (3.3.7),38 while leporaria, which once only housed ‘little rabbits’ (lepuscula), now include boars and roe deer (3.3.8). In this way, Merula’s historical analysis of pastio villatica replicates the material and linguistic rationales behind the third preface’s explanation for the differentiation between agri cultura and pastio agrestis. In so doing, it also substantiates the reconstructed logic of the related claims regarding pastio villatica: no longer the ‘lowly’ (humilis) practice of the maiores, animal husbandry of the villa has emerged as a distinct sphere of res rusticae due to increased wealth. As a newly emergent form of rustic production alongside agri cultura and pastio agrestis, pastio villatica needs to hold out the promise of greater utilitas to satisfy the third Dicaearchan requirement for marking a new stage of human existence. Merula’s definition (3.3.1) already hints that this practice at least aims for utilitas just as the other spheres of res rusticae do. But, as the dialogue conveys in a variety of ways, the profits from pastio villatica can actually outstrip those of the other spheres. In addition to the large sums of money attached to produce within the technical discussion,39 the anecdotal evidence is replete with such illustrations. For example, the case of Lucius Abuccius, whose returns from his villa are said to have been twice as great as those from his fields (3.2.17), explicitly demonstrates pastio villatica’s potential for greater profitability. Indeed, the astonishment of 37 For the setting and characters of Rust. 3, see Nicolet 1970: 113–37; Linderski 1985a: 248–54 and 1989: 105–28; Green 1997: 431–32; and Nelsestuen 2015: 175–82. 38

Cf. 3.7.1, 3.7.2–3, 3.9.2, 3.10.1, 3.11.1, 3.13.2.

39

See esp. 3.6.1, 3.6.6, 3.7.10, 3.17.3.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 31

14/11/2017 14:56:13

BICS-60-2  2017

32

Axius at the 60,000 sesterces procured from the aviary of Varro’s aunt (3.2.15) — more than double the annual yield of his own 200 iugera of farmland at Reate — further highlights the lucrative sums acquired through pastio villatica and underscores its status as the most productive form of res rusticae. Insofar as Varro conceives it, pastio villatica is a new form of rustic production that arises from increased wealth, makes use of recent technological innovations, and holds out the possibility of greater utility than prior forms of res rusticae. Fulfilling all three Dicaearchan requirements, pastio villatica is poised to emblematize a new stage of human existence. On its own, the preface’s historical discussion would reasonably imply that this stage is the vita urbana, yet a crucial exchange in the dialogue patently establishes the close connection between pastio villatica and the urban life of Rome. In response to Axius’s aforementioned incredulity, Varro explains: sed ad hunc bolum ut pervenias, opus erit tibi aut epulum aut triumphus alicuius […] aut collegiorum cenae, quae nunc innumerabiles excandefaciunt annonam macelli. reliquis annis omnibus etiam si hanc expectabis summam, spero, non tibi decoquet non ornithon; neque hoc accidet his moribus nisi raro, ut decipiaris. quotus quisque enim est annus, quo non videas epulum aut triumphum aut collegia non epulari? But to haul in this score, you will need to have either a banquet or someone’s triumph […] or the collegial dinners, which these days in great numbers light up the price of butcher’s meat. Even if you await this amount in all the other years, your aviary will not, I suspect, go bankrupt; nor will it happen that you are cheated — except rarely — in these present mores. For how few are the years in which you do not see a banquet or a triumph or collegia feasting? (3.2.16) Because pastio villatica deals in the sale of prestige foodstuffs and requires tremendous outlays of capital, its economic viability hinges on the occurrence of public banquets, triumphal feasts, and collegial dinners — the sorts of inherently urban phenomena that drive up (cf. excandefaciunt) the demand and, consequently, the market price for its luxury products. In this respect, the successful practice of pastio villatica is fundamentally beholden to the potentially volatile market conditions of the city.40 Yet, as Varro also declares, the present state of mores means that the chances of such public occasions happening in the city are actually quite good. Indeed, as Merula plainly affirms in response, ‘it is on account of luxury […] that there is, in a certain way, a daily banquet within the gates of Rome’ (propter luxuriam […] quodam modo epulum cotidianum est intra ianuas Romae, 3.2.16). Not simply contingent on increased wealth, pastio villatica depends on the city itself, its festal calendar, and the contemporary state of urban mores. By offering a new mode of production that is intrinsically tied to and, in fact, can even take place in the city,41 pastio villatica emblematizes the vita urbana as a fourth stage of human existence.42 Viewed solely from the perspectives of technē and utilitas, pastio villatica and the vita urbana it stands for would seem to represent an unabashedly positive development in the history of Roman res rusticae. Yet, the dialogue also reveals that its progressivist qualities remain freighted with considerable baggage. In addition to the aforementioned financial risks, 40

For an exception, see 3.16.10–12. Cf. Nelsestuen 2015: 204–07.

Varronian pastio villatica takes place in or around villas (3.3.1) located in rural, suburban, or urban areas (3.4.2, 3.7.11; cf. 3.2.6). See also Purcell 1995: 151–97. 41

42

Contrast Piras, p. 12 in this volume.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 32

14/11/2017 14:56:13

GRANT A. NELSESTUEN: VARRO, DICAEARCHUS, AND ROMAN RES RUSTICAE

33

it prompts many practitioners to engage in ostentatious and untenable building programmes — sometimes even in direct competition with one another and against the common good.43 But the dubious morality of pastio villatica is what makes it most suspect: it slaughters animals kept in lamentable living conditions (e.g. 3.5.1–7, 3.9.19), meticulously fattened up through ‘cramming’ (sagina) and other questionable practices (e.g. 3.7.10), and expressly raised for the prestige eating of meat delicacies;44 as a form of production that explicitly depends on luxuria, it trades in the increasingly fastidious and gluttonous tastes of producers and consumers;45 and the thought of its lucrative profits wrought from the moral exploitation of the public whets the avaricious appetites of men like Axius.46 Even the dialogue’s literary frame associates pastio villatica with the predation of fellow citizens and the trope of cannibalism; after all, the conversation takes place on the eve of civil war and in the Villa Publica, where an eructating augur fondly reminisces over a fine plate of cooked birds in the presence of fowl-citizens (3.2.3; cf. 3.4.3). In a way that both evokes and transcends Dicaearchus’s association of the pastoral stage with the start of humankind’s physical maladies and social ills, the emergence of pastio villatica as a mode of production for the vita urbana is fundamentally symptomatic of Rome’s precipitous decline in morality and civil disorder. Two further implications follow for understanding Varro’s view of the history of Roman res rusticae and the terms of his engagement with Dicaearchus. First, the new and productive, but morally suspect, practice of pastio villatica is of a piece with the sort of avaritia that Varro observed was the cause of the illicit conversion of arable fields into pasturelands (2. pref. 4). In this respect, pastio villatica may hold out greater profits for the practitioner, yet does nothing to reverse — and, if Axius’s example is anything to go by,47 possibly even exacerbates — the decline of Italian agriculture (2. pref. 1–4). From this vantage point, the ascendancy of pastio villatica as a new and profitable form of rustic production intrinsically tied to the moral depravity of urban Rome perhaps adds further significance to Varro’s comment about the maiores’ recurring efforts to return their fellow citizens to the countryside and their estimation of agricolae as ‘the sole remnants of the stock of Saturn’ (3.1.5).48 Second, in the course of engaging with Dicaearchus’s three-stage theory, Varro would also seem to have transferred the social ills — particularly, greed, ambition, and luxury — that his predecessor associated with the second stage of pastoralism and third stage of agriculture to his fourth stage of animal husbandry of the villa. Not simply a case of intellectual appropriation, then, Varro’s engagement with Dicaearchus’s thought also entails the innovative application of both its progressivist elements and primitivist anxieties to his view of the past and present states of Roman res rusticae. In so doing, Varro has likewise provided an ambivalent theory of progress, but keyed it specifically to his subjective experience of contemporary Rome. University of Wisconsin, Madison 43

See esp. 3.7.11, 3.17.6–8, 3.17.9. Cf. 1.13.6–7.

Note that M. Seius, an exemplary practitioner of pastio villatica (3.2.7–14, 3.6.3), may also have been the inventor of foie gras (Plin. HN 10.52).

44

45

See also 3.3.6–7, 3.3.9, 3.6.3.

46

For Varro on moral corruption in the late Republic see also Todisco in this volume.

Axius owns a farm (3.2.5–9) and a cattle ranch (3.2.3–4; cf. Shatzman 1975: 308), but his fervour for pastio villatica’s profits (3.2.16, 3.2.18, 3.4.1, 3.7.11) implies that he would readily abandon the other two spheres of res rusticae should the opportunity arise.

47

48

For the religious dimensions of such claims, see North 1995: 135–50.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:54:56 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 33

14/11/2017 14:56:13

‘THE LAWS OF THE RITES AND OF THE PRIESTS’: VARRO AND LATE REPUBLICAN ROMAN SACRAL JURISPRUDENCE1 DUNCAN MACRAE Abstract: Starting from Cicero’s famous panegyric on Varro’s Antiquitates and attempting to look past the image of the book provided by Augustine, this article proposes a new reading of that work and its place in late Republican intellectual culture. Cicero’s specific claim that Varro opened up ‘the laws of the rites and of the priests’ for his readers allows us to contextualize the Antiquitates within a contemporary jurisprudence. The rise of Roman legal studies in general in the first century bc extended to the laws of the priestly colleges: there are signs of lively debate over their nature and the production of texts on the details of these iura. By re-reading the fragments from the Antiquitates alongside the evidence for this sacral-legal turn, we can gain both a new appreciation for the place of law (ius) in Varro’s textualization of Roman religion and a fuller understanding of Republican legal thinking. Keywords: Roman law; pontifical law; pontifical college; Varro; Antiquitates Rerum Divinarum; Cicero; Q. Mucius Scaevola

‘That is the case, Varro,’ I said, ‘for we were wandering and roaming in our own city like outsiders, it was as if your books (tui libri) led us home so that we could finally know who and where we were. You revealed (aperuisti) the age of the homeland, the divisions of periods, you revealed the laws of the rites, the laws of the priests (tu sacrorum iura tu sacerdotum), you revealed the method (disciplinam) of domestic affairs and of campaign, you revealed the site of the regions, of the places, you revealed the names, types, functions, and reasons for all things, divine and human.’ (Cic. Acad. 1.9: author’s translation) At the beginning of the sixth book of his De civitate Dei, Augustine of Hippo opens what will be a two-book-long polemic against the late Republican intellectual Marcus Terentius Varro with a long passage of praise for his long-dead opponent, presumably with the goal of making clear that Varro was a worthy target of the rhetorical assault to come. At the heart of his cynical panegyric, Augustine quotes the famous passage from Cicero’s revised version of his Academica with which I opened.2 For Augustine, as for many readers since, this passage stands as the statement of Varro’s intellectual achievement. The Christian bishop, of course, could read parts of Varro’s own work — notably the Antiquitates, the very likely referent of Cicero’s tui libri — that we now cannot.3 I am grateful to Valentina Arena and Fiachra Mac Góráin for their excellent organization of the Varro workshop and edited volume. I thank Paul du Plessis, Caroline Humfress, and Adam Gitner for assistance. I also thank John Bodel and his Fall 2016 Varro seminar at Brown for timely feedback on this material.

1

August. De civ. D. 6.2: denique et ipse huic tale testimonium perhibet. See Hadas (this volume) for the cynicism.

2

3 The idea that the Antiquitates are the libri mentioned here has been the scholarly consensus since Schneider 1794: 1.230, who thought it manifestum. For more recent views see, e.g., Dahlmann 1935: 1229 and Cardauns 1976: 12–13. Note too that the character Varro in Cic. Acad. 1.8 speaks of the antiquitatum prooemia and Augustine quotes Cicero in the introduction to his polemic against this work.

BICS-60-2  2017

34

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 34

14/11/2017 14:56:13

DUNCAN MACRAE: ‘THE LAWS OF THE RITES AND OF THE PRIESTS’

35

Despite this loss, the Ciceronian statement stands as a potential map of Varro’s (lost) learning. Although we cannot know if Varro would have framed his own work in the same way, it allows us to see how it ‘made sense’ in late Republican Rome. Like the famous Chinese encyclopedia invented by Borges, which divided animals into groups ‘(a) those that belong to the emperor, (b) embalmed ones, (c) those that are tamed, (d) sucking pigs, (e) mermaids, (f) fabulous ones, (g) stray dogs, (h) those that are included in this classification’ and provoked Foucault to laughter and the composition of his The Order of Things, Cicero outlines a culturally specific ‘Roman encyclopedia’ that Varro apparently provided to his contemporaries, and led them ‘as if to home’.4 Cicero, in a single rapid sentence, charts out the salient taxa of Roman knowledge that Varro offered in the Antiquitates: the specific revelation of Roman time, sacral iura, political disciplina and space, as well as the terms, categories, functions, and origins of all of Roman culture.5 Cicero’s order of Roman things, however, has often been simply read as a blanket summary of Varro’s encyclopedic antiquarianism.6 In this reading, Cicero outlines a project that feels familiar: a handbook of Kult-, Privat-, and Staatsaltertümer. Varro, however, is not a colleague; neither was he a colleague of the early modern antiquarians who appropriated him as a scholarly predecessor.7 Instead, I suggest, we should read Cicero’s list of Varro’s accomplishments without assuming that the items straightforwardly correlate with more modern disciplinary projects. For example, the recent debate on the intellectual origin of the medieval disciplines — Varronian or late antique? — is a fine reminder about the difficulty of articulating the relationship between ancient and later fields of knowledge.8 Instead, a historicist approach to Roman modes of knowing the world can provide new insights into Republican intellectual life and the work of Varro.9 Important recent studies by Andrew Wallace-Hadrill and Denis Feeney have already set Varro’s contribution to reframing Roman time and space in a first century bc context.10 We can extend this kind of investigation to another element of the cultural periodic table laid out by Cicero: ‘you revealed the laws of the rites, the laws of the priests’ (tu sacrorum iura tu sacerdotum […] aperuisti). What did it mean for Cicero to speak of ‘laws’ related to the worship of the gods? Modern scholars have often assumed that there is an underlying category of sacred law in Roman thought, which existed in parallel to public and private law.11 In fact, the Romans of the late Republic did not have a single unified concept of

4

Borges 1999: 231 and Foucault 1970: xv–xx.

Although Cicero writes of ‘all things, divine and human’, a comment by Augustine reveals that Varro’s Rerum humanarum were ‘not on global affairs but about Rome alone’, (non quantum ad orbem terrarum, sed quantum ad solam Romam pertinent: August. De civ. D. 6.4 = Ant. div. fr. 5 [I cite from Cardauns 1976 throughout]).

5

6

See, e.g., Dahlmann 1935: 1229; Rawson 1985: 263; Fuhrmann 1987: 144.

Varro has often been mistaken for a colleague: see, e.g., Taylor 2015: 21, for the idea that Varro is a modern ‘language scientist’. On antiquarianism, see MacRae 2018, which argues that the discipline is an early modern invention. 7

8

See Hadot 2005, Shanzer 2005.

Moatti 1997 is a model for this kind of study, but more remains to be done. On Varro, see the contextualist approach of Gitner 2015.

9

10

Feeney 2007: 198–201; Wallace-Hadrill 2008: 231–43, 260–64.

See, e.g., Cenderelli 1973: 163 (‘diritto sacro’). For criticism of the idea, see New Pauly s.v. ‘Sakralrecht’ (Rüpke). 11

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 35

14/11/2017 14:56:13

36

BICS-60-2  2017

religious law.12 There were no ‘religious’ courts or religious police; rather, the connection of ius with ritual practice seems to have been made with reference to the priestly colleges, especially the pontifices and the augurs.13 In Cicero’s own De domo sua, for example, he articulates — for an audience of priests — the idea that there are separate iura appropriate to either college and contrasts these iura with public law.14 I will suggest in this essay that the iura of the priestly colleges had become a significant object of study in the first century bc and will place Varro’s work in relation to this contemporary jurisprudential culture; following Cicero, I argue that sacral jurisprudential material had a substantial place in Varro’s Antiquitates rerum divinarum. Now is the time to turn to the relationship between Varro and Roman sacral jurisprudence. In recent scholarship on Roman religion, learned late Republican texts on this topic, including Varro’s own, have been assimilated to the broad category of ‘antiquarianism’ and, thus, separated from the ‘real’ domain of religion.15 John Scheid, for example, marginalizes these works as ‘the province of private scholars, writing books that were learned but not official’.16 On the other hand, work on Varro’s writing on the law has often been conducted according to the frameworks of the ‘Romanist’ tradition of study of Roman law. Romanists have been preoccupied by whether Varro deserves the title of ‘jurisprudent’ and with his ideas about private law.17 In line with the palingenetic tendencies of the field, the focus has been on the otherwise unparalleled notice of a work in fifteen books by Varro, De iure civili, given by Jerome (Ep. 33) and the status of various notices of legal procedure and doctrine given by Varro in preserved works (most notably the sections in the De re rustica on the law of sale).18 The most comprehensive and recent work in this vein, by Aldo Cenderelli, argues that Varro does not deserve the title of jurisprudent, but that students of Roman law should still use Varro in a limited way as an indirect source.19 This kind of approach, however, is defined See North 2009 for this point and an overview of the topic. A useful survey of Latin terms for ‘sacred law’ and ‘pontifical law’ can be found in the Rutgers dissertation of Michael Johnson 2007. I set aside here the term ius divinum, which is generally not used to refer to a body of Roman norms, but rather to either a dharma-like concept of divine order (see, for example, Cic. Part. or. 129 and Livy 1.20.6) or as an expression of the subjective property interest of the gods (Gai. Inst. 2.2–9, cf. Dig. 1.8.1). cf. RE s.v. ‘Ius divinum’ (Berger). 12

These colleges did not have broad jurisdictional remit and most often seem to be learned advisors to the senate: see Beard 1990, Santangelo 2013, and Rüpke 2011a. 13

The idea is raised early in the speech: Cic. Dom. 32: ‘these matters are divided between ritual law and public law; I’ll leave aside the part on ritual law, which is much more lengthy to explain, and will speak about public law’ (quae cum sit in ius religionis et in ius rei publicae distributa, religionis partem, quae multo est verbosior, praetermittam, de iure rei publicae dicam). It becomes clear that ius religionis is a blanket term for separate iura of augural and pontifical colleges, about which he repeats his refusal to speak several times in the speech (Dom. 39; 121; 128; 138). This was not in good faith: a substantial part of the speech deals with pontifical law: see Linderski 1985 and MacRae 2016: 64–68.

14

See, for example, the discussion of these texts in Prescendi 2007: 16 and Beard–North–Price 1998: 1.152–53 for the idea that they are separate from religion proper (ritual); by contrast, MacRae 2016 presents a case for their significance in shaping ideas of ‘Roman religion’. 15

Scheid 2006: 33. The same thing, however, could be said of the civil jurists (I owe this point to Caroline Humfress). 16

Sanio 1867 and Stella Maranca 1934 argue in favour of Varro as jurisprudent; Cenderelli 1973 argues against (see below). 17

18

Rust. 2.2–7. On these passages, see Harries 2006: 29–32.

19

Cenderelli 1973; see also Cenderelli 1976a and 1976b.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 36

14/11/2017 14:56:13

DUNCAN MACRAE: ‘THE LAWS OF THE RITES AND OF THE PRIESTS’

37

by a vision of what ‘really’ counts as Roman law that is strongly shaped by the example and words of the classical jurists who were canonized by Justinian’s Digest. Cenderelli, for example, argues that Varro is not a jurist because, despite his evident knowledge of the technicalities of law, he did not think like a jurist; he lacked the ‘drive, which we could call instinctive, perhaps unconscious’ to isolate the law from other social facts.20 This view of the jurists as guardians of the autonomy of the law is manifestly derived from the classical Roman lawyers and the civilian tradition that follows them.21 Recent thinking about Roman law has begun to turn away from this position. As Roman law has lost its place in legal curricula, it has begun a ‘descent into history’ and has started to seek out ‘new frontiers’.22 This new wave of Roman legal history has challenged the image of the social and intellectual autonomy of Roman law and jurisprudence. In a provocative recent essay, Georgy Kantor points out that the juristic stereotype that made the Roman legal sphere coterminous with private, secular law is not supported by the evidence and suggests that we take a more inclusive view of ‘Roman law’, encompassing religious law.23 Looking closer, then, at the work of Varro and other late Republican writing on sacral and priestly law can contribute to this new historiography of Roman law.24 This essay aims to substantiate Cicero’s characterization of Varro’s Antiquitates as a work, in part, on ius and, more broadly and in support of that argument, to investigate the place of sacral jurisprudence in Roman elite intellectual culture.25 I start from the wider context by investigating how pontifical and augural law became the object of written works and intellectual debate in the late Republic, and then turn to the fragments of Varro to show how the Antiquitates was engaged with that legal discourse on the sacred. 1. Writing the iura In a passing comment in Cicero’s Brutus, designed to authenticate the reputation of the great jurist and statesman Ser. Sulpicius Rufus, the eponymous character mentions that he had encountered the jurist on Samos: ‘Just recently, I listened to him carefully and often on Samos, when I wished to know how our pontifical law was connected with civil law.’26 This is a tantalizing and allusive passage; Cicero gives us no more details about this encounter and seems to expect the audience of the Brutus to understand why Brutus would have been so keen to hear Sulpicius on this topic. Cenderelli 1973: 166–67: ‘il suo impulso, diremmo istintivo e forse inconsapevole, ad isolare in tale fatto l’elemento giuridico’. For arguments in favour of the ‘isolation’ of Roman law, see Watson 1995 and 2007. See, against this view (all espousing forms of legal realism), Frier 1985, Meyer 2004: 1–7, Harries 2006, Cairns and du Plessis 2007, du Plessis 2013. 20

21

See Tuori 2007 for the role, in particular, of the civilian legal tradition in Roman legal history.

See the titles of Frier 2000 and the volume introduced by du Plessis 2013. This new direction itself has a long history: Momigliano 1966. 22

23

Kantor 2012.

Sacral jurisprudence, on these terms, can become one of the ‘pieces’ of Roman legal culture that Bryen 2014 urges historians to pick up in search of their distinctive culture of legality. 24

25

On Varro’s interest in ius publicum see Todisco in this volume.

Cic. Brut. 156: audivi enim nuper eum studiose et frequenter Sami, cum ex eo ius nostrum pontificium, qua ex parte cum iure civili coniunctum esset, vellem cognoscere. This is all the more jarring in light of Cicero’s claim at De or. 3.136 that ‘nobody studies pontifical law’ (pontificium [ius] […] nemo discit), but that idea belongs to a loaded discussion of the ‘decline’ of the Roman political elite. 26

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 37

14/11/2017 14:56:13

38

BICS-60-2  2017

Another Ciceronian work, the De legibus, however, allows us to see the likely issue. There, Cicero discusses precisely the question of the relationship of pontifical and civil law in two separate passages of polemic against the father and son Publius and Quintus Mucii Scaevolae, who were both pontiffs and jurists. In the first, he decries the position taken by the elder Scaevola that ‘no one is a good pontiff except the one who knows the civil law’; in the second, he complains that the Scaevolae ruined the pontifical law by importing a dodge from the civil law that could allow heirs to avoid having to keep up familial rites (sacra).27 If we read these passages against the grain, we can see that the Scaevolae and Cicero were on two sides of a debate over the problem posed in the Brutus: how were pontifical and civil law related? The Scaevolae seem to have emphasized that the two were closely linked and to have extended pontifical law by analogy with civil law; Cicero maintains that the connection between the two was tenuous (quantulum).28 Historians have tended to side with one or the other party in this dispute — the popular option has been to seek substantive connections between the ius civile and the ius pontificium; less often scholars have argued for the relative autonomy of the pontifical law.29 Neither argument is quite satisfying: we simply lack the evidence to know what the ‘proper’ relationship between the two iura might have been and the very fact of Brutus’s question to Sulpicius perhaps suggests that it was unknown even in the late Republic.30 Instead, we are better off focusing on the dispute itself as evidence that the boundaries of these two bodies of law were being worked out in intellectual terms during the late Republic. I contend that this debate was so pressing because in the same period members of the Roman elite had started articulating the substance of both civil law and priestly law in written works. This new ‘internal’ discourse of the law, as Frier calls it, following Lawrence Friedman, can hardly have failed to raise questions about the relationship between the different forms of ius.31 Certainly, we can trace, through the second and first centuries bc, an accelerating written discourse on pontifical and augural law. The earliest certain writings on sacral jurisprudence in Rome belong to the second century: we know something of a work on the ius pontificium

Cic. Leg. 2.47; 2.52–53. On the technicalities of the dispute over familial sacra, see Harries 2006: 149–53 for a concise guide; for more detail, see Bruck 1945, Bona 2003, and Johnson 2015.

27

28 Cic. Leg. 2.47: ‘What has a pontifex to do with the law of party walls or water channels or anything at all, except that bit of law connected with ritual? And how little that is! I believe it’s the law concerning rites, vows, holidays, tombs, and anything else of that kind.’ (quid enim ad pontificem de iure parietum aut aquarum aut ullo omnino, si eo quod cum religione coniunctum est? id autem quantulum est! de sacris credo, de votis, de feriis et de sepulcris, et si quid eius modi est.) See Fontanella 2012: 71–78 for possible philosophical background to Cicero’s position.

There is a very long Romanist bibliography on this topic, often without clarity about what connection or separation might entail, though see Schulz 1946: 12 for the stakes. Earlier literature: Pernice 1885, 1886; Mitteis 1908: 26–28; and Watson 1992. See Wieacker 1988: 310–40 for further bibliography. Of recent and non-Romanist works, see in favour of a connection: Scheid 2006; against a connection: Tellegen-Couperus 2012, and Johnson 2015. 29

Our sources suggest that an original pontifical monopoly over the civil law was removed in the late Republican period: Pomponius’s juristic history of Roman ius gives the full narrative (Dig. 1.2.2; see also Livy 9.46). The question of the historicity of this narrative goes beyond the scope of the present paper; it suffices to note that even in the late Republic, the relation of the pontifices to the civil law was up for debate. 30

31 Frier 1985: 141. In addition, the new ‘external’ discourse on codification of law (see Suet. Iul. 44 and Cicero’s lost de iure civili in artem redigendo) could also have provoked reflection on the limits of the iura.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 38

14/11/2017 14:56:13

DUNCAN MACRAE: ‘THE LAWS OF THE RITES AND OF THE PRIESTS’

39

by a certain Fabius Pictor.32 Another Fabius, a Fabius Maximus Servilianus, if this is not the same person, also wrote on this topic in the middle of the second century.33 It is in the first century, however, that we find much more evidence for writing on the ius pontificium and the ius augurale.34 We have several references to these works (it is unclear whether these are references to titles of works or just summaries of their content):35 Cicero himself was responsible for a treatise de auspiciis; a Veranius wrote a work on pontificales quaestiones; Lucius Caesar composed augurales libri; like Cicero, Valerius Messalla Rufus wrote de auspiciis; Appius Claudius Pulcher dedicated a book de iure augurali to Cicero himself; Granius Flaccus is associated with a work de iure Papiriano, formally a commentary on old regal law, but apparently focused on ritual norms; and Trebatius Testa wrote a work de religionibus. In the next generation, Antistius Labeo wrote de iure pontificio and Ateius Capito, the exegete of the Augustan Ludi Saeculares, also produced texts de iure pontificio, de iure sacrificiorum, de iure augurali.36 We also know that Quintus Mucius Scaevola the pontifex and the augur Claudius Marcellus had opinions on these topics, though it is uncertain whether these were contained in texts or circulated orally.37 All these works seem to have taken their cue from the slightly earlier development of a literature on the civil law: the Tripertita of Sex. Aelius Paetus Catus is traditionally credited as the breakthrough.38 That work appears to have been an extended commentary on the Twelve Tables and, therefore, concerned with the ius civile. Aelius had some secondcentury followers, including Junius Brutus and Manilius, but again, this literature seems to have flowered in the first century. Quintus Mucius Scaevola has a reputation as a key figure in this movement and his systematic eighteen books de iure civili apparently set a pattern followed by Roman jurists for centuries.39 Many others followed the example of Scaevola: the most famous, thanks to Cicero, are Sulpicius Rufus, Trebatius Testa, and Aelius Tubero.40 In the Augustan period, Ateius Capito and Antistius Labeo were particularly prominent and, eventually, fictive ancestors of the two leading schools of the classical jurists.41 This person is most likely to be the Ser. Fabius mentioned in Cic. Brut. 81 and should therefore be distinguished from the historians Q. Fabius Pictor and Q. Fabius Maximus Servilianus. For discussion, see Rüpke 2008: 677 n. 5.

32

Macrob. Sat. 1.13.28. This may be a case of garbled transmission through Macrobius, but it is better to suspend judgment. 33

The evidence for all of the following except Cicero can be found in the collection of jurisprudential fragments by Bremer 1896 and 1898, though some book/work attributions and the conclusions in his commentary are incautious. For Cicero, see Müller 1898–1908: 4.3:312. The evidence for the juristic style of these books rests either on their titles (or, at least, how they are referenced by later authors), which echo the form of other juristic works (de iure, quaestiones), or on reports of their contents. 34

Citation of titles for Latin books, especially in the Republican period, tended to be approximate; confusion between authorial titles and indications of subject matter (in the form of de + topic) is particularly common in our evidence. On these issues, see Daly 1943 and Horsfall 1981. 35

36

For Capito, see also Strzelecki 1967 in addition to Bremer 1898.

37

For Q. Mucius Scaevola, see below. For Claudius Marcellus, see Cic. Leg. 2.32 and Div. 2.75.

38

Dig. 1.2.2.38: ‘cradle of the law’ (cunabula iuris). See Bauman 1983: 129–32.

Dig. 1.2.2.41. See Bremer 1896: 48–104 for the fragments. See Schiavone 2012 and Frier 1985: 159–71, esp. at 171: ‘Q. Mucius is the father of Roman legal science and of the Western legal tradition.’ We should perhaps be a little more cautious (see Tuori 2007: 21–69). Cf. Watson 1987, for the idea that Q. Mucius was not innovative, and Harries 2006 and Zetzel 2013, for the role of Cicero in promoting Scaevola as a symbol of legality. 39

40

For biographies, see Kunkel 1952: #40, 44, and 46, and Bauman 1985.

41

See Bauman 1989: 25–55.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 39

14/11/2017 14:56:14

BICS-60-2  2017

40

The repetition of names over the last two paragraphs is no coincidence: it was broadly the same group of men who wrote books on the different iura, civil, augural, and pontifical. If we look to the surviving fragments of the Republican works on jurisprudence — and frankly this is not a large set of texts — it is clear that this was a single ‘internal’ or expert discourse. There were, of course, points of difference between the texts on different forms of ius, derived from the subject material: the genre of collections of juristic responsa, whether presented in specific or generalized form, were particular to the individualized cautelary jurisprudence of civil law; the texts on priestly law instead collected the decreta and responsa of the colleges, which, despite the shared name, had a different valence as expressions of the corporate body and were often produced in response to magisterial or senatorial referrals.42 But beyond these differences, we find elements of a common intellectual style: concern to collect norms and record verbal formulae, interest in commentary on archaic laws, and the definition and distinction of terms. Taking these habits in turn, we can find examples in the various preserved fragments of these works. We can start with the second-century Fabius, whose fragments manifest the collection and edition of rules around the pontifices and efficacious verbal formulae. The well-known passage on the taboos around the flamen Dialis, preserved by Aulus Gellius and probably transmitted to him in a work by the early imperial jurist Masurius Sabinus, is a concatenation of rules that constrained that priest: It is forbidden for the flamen Dialis to be carried on horseback […] it is forbidden for the flamen Dialis ever to swear an oath; it is forbidden for him to use a ring unless it is perforated and broken. It is not lawful (ius non est) for fire to be taken from the flaminia, the house of the flamen, unless it is for rites.43 This bare report of pontifical norms hardly appears to our eyes to constitute anything deserving of the word ‘systematic’, but Fabius Pictor’s work does appear to have initiated learned writing on the ius pontificium. In this respect, he has much in common with the other juristic writers of the second century, like Aelius Paetus and his successors, who seem to have been primarily collectors of rules — often in the form of cautelary responsa — and actiones. In the first century, we find in several fragments on pontifical and augural law a continuing concern with formulae. Trebatius Testa, for example, gave the formula for a libation and specified that the word vinum had to be qualified by the adjective inferius in order to avoid the accidental consecration of all the wine in the household storage.44 Etymology was used as a way to understand the terminology. We find a good example of this in Testa’s de On priestly decreta and responsa and their recording, see Linderski 1985b, Linderski 1986: 2154–62 and 2241– 56, Scheid 1994, and North 1998. 42

Fabius Pictor fr. 3 (Bremer) = Gell. 10.15.3–7: equo Dialem flaminem vehi religio est […] item iurare Dialem fas numquam est; item anulo uti nisi pervio cassoque fas non est. ignem e flaminia, id est flaminis Dialis domo, nisi sacrum efferi ius non est. Schulz 1946: 98 identifies this passage as representative of the style of Roman legal discourse in the ‘Hellenistic period’. See also Fabius Pictor fr. 1 (Bremer) = Gell. 1.12.14 for a religious formula.

43

Trebatius Testa fr. II. 9 (Bremer) = Arn. Adv. nat. 7.31: ‘“May the god be honored by this sacrificial wine.” The word sacrificial, says Trebatius, is added for the sake of this and mentioned for this reason: so that not all the wine that is stored in the cellars and magazines from where that wine which is being offered was taken becomes consecrated and removed from human consumption.’ (‘mactus hoc vino inferio esto.’ inferio, inquit Trebatius, verbum ea causa est additum eaque ratione profertur, ne vinum omne omnino quod in cellis atque apothecis est conditum, ex quibus illud quod effunditur promptum est, esse sacrum incipiat et ex usibus eripiatur humanis.)

44

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 40

14/11/2017 14:56:14

DUNCAN MACRAE: ‘THE LAWS OF THE RITES AND OF THE PRIESTS’

41

religionibus. Trebatius defined the word sacellum as ‘a small place dedicated to a god, with an altar’ and etymologized this word as a compound of sacra and cella.45 Another form of writing on law in the Republican period, starting with Aelius Paetus’s interpretation of the Twelve Tables, was commentary on archaic statutes.46 We find traces of a work of this kind by Granius Flaccus among the fragments of sacral jurisprudence. His name was attached to a work de iure Papiriano, which indicates it was a commentary on a compilation of royal statutes under the name of a legendary early Republican pontifex maximus named Sextus Papirius.47 As far as we can see, Granius explained various cult regulations, mostly attributed to Numa. For example, he explained that the word paelex in a regulation relating to the cult of Juno meant ‘concubine’.48 Quintus Mucius Scaevola apparently opened up new approaches to jurisprudential thinking in the early first century. The most central of these was a concern with the definition and distinction of terms and the making of analogies. His dialectic definition of the term gentilis — found in Cicero’s Topica — exemplifies the habit of working through the meaning of legal terms by use of increasingly particular distinctions.49 We can also see this sort of intellectual work in Appius Claudius’s definition of a sollistimum tripudium as whatever fell from the mouth of a bird, because it belonged to a wider class of portentous falling objects.50 The definition alludes to the auspicia pullaria, the Roman practice of feeding chickens before battle in order to generate positive omens (the seed would naturally fall from the beaks, getting around the idea that the objects should fall without human interference). We know more about this Claudian definition than is given in the terse report in Festus’s lexicon. As Jerzy Linderski pointed out, Cicero complains in the De divinatione that the augural college had issued an old decretum that defined the tripudium in exactly this way, as something that fell from the mouth of any bird; he almost certainly learnt this from Appius’s book on augury, which was dedicated to him and he read while governor of Cilicia.51 In other words, Appius Claudius was not simply transmitting a traditional understanding of the form of omen; rather he recorded the decree of the college that had authorized it. In this, he was not alone: the preservation of collegial decreta and responsa appears in several fragments of these jurisprudential texts.52 For example, Ateius Capito explained that the Trebatius fr. II. 5 (Bremer) = Gell. 7.12.5–6: Trebatius in libro de religionibus secundo: […] sacellum ex duobus verbis arbitror compositum sacri et cellae, quasi sacra cella. For Varro’s discussion of different categories of temple and precinct, see de Melo in this volume, section 2.1.

45

46

On the role of the Twelve Tables in juristic textual ‘topography’ see Wibier 2014: 59–65.

See Schulz 1946: 89, with n. 2 for the debate over whether the ius Papirianum was an authentically old document or, more likely, was the work of Granius Flaccus himself (perhaps a new collection of old laws). 47

48

Granius Flaccus fr. I. 1 (Bremer) = Macrob. Sat. 3.11.5. For the law, see Paulus Diaconus 248 L.

49

Cic. Top. 29.

Appius Claudius Pulcher fr. 1 (Bremer) = Festus 386 L: ‘Appius Pulcher in book 1 of Augural Practice says that a sollistimum tripudium is when something falls from a bird, out of the mouth, that she picked up herself; or a solid rock or a rooted tree falls, which are not previously cut down or thrown down or pushed over by rot or human force.’ (sollistimum, Ap. Pulcher in auguralis disciplinae liber I ait, esse tripudium quom avi excidit ex ore, quod illa fert; saxumve solidum aut arbos viviradix ruit, quae nec prae vitio humanave vi caedunturve, iacianturve, pellanturve.) This liber auguralis disciplinae is likely the same that was dedicated to Cicero and discussed the ius augurale (Cic. Fam. 3.9.3). 50

Cic. Div. 2.73: decretum collegii vetus habemus omnem avem tripudium facere posse. See Linderski 1985b: 227 and Fam. 3.4.2; 3.9.3; 3.11.4 for the connection with Appius Claudius’s book. 51

Aside from the example below, Gell. 5.17.1–2, Aelius Gallus = Festus 424 L. See also Macrob. Sat. 1.16.28 (Messalla Rufus seeking a responsum), and Veranius fr. I.1 (Bremer) = Festus 366 L (absence of augural decretum). See Cohee 1994 for priestly decreta and responsa, especially 39–41 on their preservation in jurisprudential literature. 52

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 41

14/11/2017 14:56:14

42

BICS-60-2  2017

third-century pontifex maximus Tiberius Coruncanius had obtained a decretum of a college that permitted the performance of feriae praecidianeae on a day of ill omen (dies ater).53 In other words, juristic texts on pontifical and augural law were not simply ‘external’ texts, but also transmitted individual rulings of the colleges, articulating them as sources of ius. The poor preservation of the texts limits how much more can be said about sacral jurisprudence in the late Republic. In subsequent periods, this literature apparently failed to find a broad readership: as will be apparent, Gellius, Festus, and Macrobius dominate as tradents of this literature. This subsequent (non-)reception, however, should not prevent us from seeing that these iura were a significant locus of intellectual activity in the late Republic, conducted on the same terms, by some of the same people, as civil jurisprudence. Members of the colleges, jurists, and members of the elite all wrote on sacrorum iura […] sacerdotum: they collected and discussed rules and formulae, commented on old legal texts and defined terms. When they did so, they used the tools of etymology, dialectic, and historical research that were the common patrimony of intellectual culture of the period. 2. tu sacrorum iura tu sacerdotum: Varro’s achievement What was Varro’s place in this intellectual context? Can we say why Cicero gave him credit for ‘opening up’ the laws of the rites and of the priests? As should now be clear, the inclusion of the iura in the list from the Academica does not indicate that Varro was first to discuss them. On the contrary, Varro was far from alone in writing on this topic in the mid-first century bc and he, like others, was following predecessors like Fabius Pictor. In highlighting the juristic aspect of Varro’s work, Cicero was responding, I suggest, to two broad tendencies apparent in the fragments of the Antiquitates: concern with ritual and priestly norms and engagement with the wider late Republican discourse on pontifical and augural law. If we have not been accustomed to think about Varro’s Antiquitates in this way, we must blame (or thank) Augustine.54 His De civitate Dei is by far the most important text for understanding the Antiquitates, particularly the sixteen books on res divinae. This leaves us dependent on a highly tendentious account of the late Republican text. In light of this, we should note the difference between the Ciceronian summary that included sacrorum iura […] sacerdotum and the ‘table of contents’ provided by Augustine for the res divinae. Among the books listed, he explains that three books were dedicated to sacerdotes (the pontifices, the augurs, and the quindecimviri) and three others were dedicated to sacra (consecrationes, sacra privata, sacra publica), but does not frame these in terms of law. However, as has frequently been pointed out, Augustine’s interest in the book by Varro is theological; the part of the work that he discusses at greatest length is the three-book section on the gods.55 In fact, none of Cardauns’s thirteen fragments from the books on sacerdotes or sacra are taken from Augustine’s text. Ateius Capito fr. 10 (Strzelecki) = Gell. 4.6.10: ‘Therefore, I wrote out the words of Ateius Capito, from book 5 of his De pontificio iure: Feriae praecidaneae were celebrated on a day of ill omen when Tiberius Coruncanius was pontifex maximus. The college decreed that this was not a ritual offence that there were feriae praecidaneae on that day.’ (propterea verba Atei Capitonis, ex quinto librorum quos de pontificio iure composuit, scripsi: Tib. Coruncanio pontifici maximo feriae praecidaneae in atrum diem inauguratae sunt. collegium decrevit non habendum religioni, quin eo die feriae praecidaneae essent.) The identity of the college here is debated: see Linderski 1986: 2190 n. 159. 53

54

See Hadas (this volume) for Augustine’s influence on ‘our’ Antiquitates.

55

See, e.g., O’Daly 1996 and Hadas in this volume.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 42

14/11/2017 14:56:14

DUNCAN MACRAE: ‘THE LAWS OF THE RITES AND OF THE PRIESTS’

43

Nevertheless, Augustine preserves enough of Varro’s programme that we can discern the late Republican author’s interest in advancing a normative religious system. For example, one of his justifications for the book was that he could inform the reader which god should be worshipped — effectively, a declaration that he would offer theological and ritual norms.56 Going further, Augustine writes that Varro’s work was framed in the terms of the so-called theologia tripertita — making a distinction between poetic, philosophical, and civic conceptions of the gods — and leaves us enough to see that Varro’s work aimed at reconciling these three theologies, in order to rationalize and defend the tradition of the city.57 In practice, this seems to have involved the harmonization of myth and philosophy to civic ritual; in other words, civil theology was at the centre of the book, with the other two theologies deployed to support it.58 In this light, then, Varro’s apparent definition of the civil theology is particularly significant: tertium genus est, inquit, quod in urbibus cives, maxime sacerdotes, nosse et administrare debent. in quo est, quos deos publice † sacra et sacrificia colere et facere quemque par sit.59 The third form [civil theology] is that which citizens in cities, especially priests, should know and carry out. It includes which gods each person should worship publicly, which rites and sacrifices each should perform. The passage is unfortunately corrupt; my English translation gives the clear sense, but editors have not agreed on how to reconstruct Augustine’s (or Varro’s) words in this passage from the manuscript tradition of the De civitate Dei.60 At any rate, what is clear is that Varro set up his civil-theological project in terms of sacerdotes, deos, and sacra and used a vocabulary of norms, if not quite explicitly the language of ius.61 Civil theology, he claimed, provided ‘canons’ of both knowledge and practice for the civic priests: the gods they should worship and the rites they should perform. This characterization of civil theology may have been even more pointed if the originator of Roman tripartite theology was a previous pontifex maximus, Quintus Mucius Scaevola. In a passage that is difficult to interpret, Augustine writes that Scaevola — unlike Varro himself — did not harmonize the three theologies, but championed the civic one as the only form useful to the city. The gods of the poets and of the philosophers were dangerous or superfluous for the state.62 In other words, the great jurist appears to have pushed a Ant. div. fr. 3 (= August. De civ. D. 4.22): ‘From this we will be able, he says, to know which god we should invoke and pray to and for what reason’ (ex eo poterimus, inquit, scire quem cuiusque causa deum invocare atque advocare debeamus). 56

On the theologia tripertita see Pépin 1956, Lieberg 1973, Lehmann 1997: 193–225, Ando 2010, and Hadas in this volume, pp. 82–89. 57

58

See van Nuffelen 2010 and North 2014: 236–45 for how Varro used the tripartite theology.

59

Ant. div. fr. 9 = August. De civ. D. 6.5.

See the apparatus in Cardauns 1976: 20. My translation follows Merkel and Agahd in taking colere with deos and inserting quae before sacra, but this can only be a tentative solution.

60

61

Rüpke 2005: 109–10 unduly downplays this normative element.

August. De civ. D. 4.27: ‘It is recorded in books that the pontifex Scaevola claimed that three kinds of gods had been passed down: the first by poets, the second by philosophers, the third by the leading men of the city. He says that the first kind is worthless, because many things about the gods are made up; the second does not fit with states, because it has some things that are superfluous and some things that are harmful for populations to know.’ (relatum est in litteras doctissimum pontificem Scaevolam disputasse tria genera tradita deorum: unum a poetis, alterum

62

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 43

14/11/2017 14:56:14

BICS-60-2  2017

44

civil theology as a counterpart to the civil law with which he is so closely associated.63 Augustine’s phrasing implies that Varro discussed the Scaevolan idea in the Antiquitates rerum divinarum, though, as we have seen, he differed on the particulars of the relationship between the three theologies.64 Nevertheless, his decision to take up the system associated with Scaevola and to emphasize the normative content of civil theology is a useful signpost towards the possible jurisprudential orientation of his book. We can also place the dedication of the work to Julius Caesar in the context of this programme. Both Augustine and Lactantius mention the dedication and use parallel phrasing when they refer to the Antiquitates rerum divinarum as ad C. Caesarem pontificem (maximum).65 The similarity suggests that Varro himself placed the emphasis on Caesar’s pontificate in the dedication — the Christian authors are unlikely to have independently described Caesar as a pontifex. Considering that his addressee was the current head of the pontifical college, which had a strong proprietorial interest in these sacral norms, Varro’s bold choice to take up Scaevola’s distinction and to emphasize civil theology as a domain of rules for the priests in particular (maxime sacerdotes) suggests his confidence in the canons that he claimed to provide in the book and their potential concordance with the ius pontificium. The way this normative programme played out in the Antiquitates can be reconstructed from Varronian fragments on sacra and sacerdotes — not all expressly quoted from the Antiquitates, but all congruent with the apparent coverage of that work. In these texts we can find signs of an engagement with the pontifical and augural jurisprudence that I sketched in the last section. As we saw, recording and explanation of decreta and responsa of the priestly colleges was a salient part of late Republican writing on ius. Fragments from the Antiquitates demonstrate that Varro also incorporated these pontifical texts into his work. Gellius records that the Romans declared a holy day if an earthquake was felt or announced. Anyone who violated the holy day had to make a propitiatory sacrificial offering.66 However, a philosophis, tertium a principibus civitatis. primum genus nugatorium dicit esse, quod multa de diis fingantur indigna; secundum non congruere civitatibus, quod habeat aliqua supervacua, aliqua etiam quae obsit populis nosse.) I concur with North 2014: 236: ‘There is in fact a very strong case for thinking that the two positions of Scaevola and Varro, whatever the order in which they were presented in public form, were completely distinct in their arguments.’ 63

See Schiavone 2012: 232–33 and 239–41.

August. De civ. D. 4.27: ‘Varro himself does not mind saying that in his books on divine matters’ (quod dicere etiam in libris Rerum divinarum Varro ipse non dubitat). I take Augustine’s Varro ipse to imply that Varro was the tradent of the Scaevolan passage. Cardauns 1960: 33–37 cleverly hypothesized that Augustine’s source was not the Antiquitates, but the Curio de cultu deorum, another work certainly known to Augustine. He also suggested that the Scaevola was not, in fact, the real person here, but a character in the dialogic Curio. Against this proposal, note the difference between the introductory formula for this notice (relatum est in litteras) and the way that Augustine mentions the character Balbus in Cicero’s De natura deorum at De civ. D. 4.30: ‘Quintus Lucilius Balbus makes a case in his (Cicero’s) second book De natura deorum’ (disputat apud eum Quintus Lucilius Balbus in secundo De deorum natura libro). See also Schiavone 2012: 227 and North 2014: 234 for reservations. 64

Lactant. Div. inst. 1.6.7: ‘Marcus Varro […] in the books on divine matters, which he addressed to Caesar the pontifex maximus (M. Varro [...] in libris rerum divinarum, quos ad Caesarem pontificem maximum scripsit); August. De civ. D. 7.35: ‘those written and published books of Varro dedicated to Caesar pontifex’ (istos Varronis ad Caesarem pontificem scriptos atque editos). On the relationship between Varro and Caesar, see Todisco in this volume. 65

66 Ant. div. fr. 78 = Gell. 2.28.3: ‘They sacrificed a victim “whether to god or goddess”. Varro says that was done by a decree of the pontifices; since it was unknown both what power and which of the gods or goddesses was responsible for the earthquake.’ (hostiam ‘si deo si deae’ immolabant, idque ita ex decreto pontificum observatum esse Varro dicit; quoniam et qua vi et per quem deorum dearumve terra tremeret incertum esset.)

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 44

14/11/2017 14:56:14

DUNCAN MACRAE: ‘THE LAWS OF THE RITES AND OF THE PRIESTS’

45

since the deity responsible for the earthquake was unknowable, the offering was given to the deity ‘whether god or goddess’. Varro, Gellius tells us, explained that this was the result of the pontifical decree. In other words, Varro grounded the details of a specific ritual in the decree. Gellius does not preserve either Varro’s own wording or the text of the decree in this case, but another fragment does seem to preserve the legislative language of this kind of decree. Macrobius gives us a verbatim text from a work of Varro: ‘[A magistrate] should not summon men on a holy day; if he calls them, let there be a propitiatory sacrifice.’67 Varro’s interest in pontifical law relating to this topic also shows up in his De lingua Latina, where he documents Quintus Mucius Scaevola’s responsum that distinguished between a negligent and deliberate violation of the holy day and limited expiation to the negligent violation.68 Tertullian preserves another decretum on the validity of an imperatorial vow — the dedication was only permissible with the assent of the senate — and gave its context: an attempt by Marcus Aemilius to dedicate a shrine to an otherwise unknown god called Alburnus.69 As editors of the Antiquitates have noticed, the late Republican author was very likely responsible for the transmission of this decretum. A fragment preserved in Festus’s lexicon on the spolia opima, the special dedication of the armour of an enemy general killed by a Roman soldier in combat, attests to Varro’s concern to preserve sacral norms.70 The manuscript is lacunose, but enough survives for us to see that Varro recorded the rules for sacrifice and donations that were required for the dedication of the stripped armour. There were three levels of spolia, with the distinctions perhaps depending on the rank of the soldier involved. Varro gave two sets of rules for these three levels: the first, which he gave on the testimony of pontifical books (libri pontificum), fixed the sacrificial animals to accompany this dedication: ‘The state should sacrifice (publice fieri debere) a bull for the first spolia, a suovetaurilia for the second spolia, and a Macrob. Sat. 1.16.18 = Ant. div. VIII App. (e): ‘They avoided [summons] even on holy days, as Varro writes in these words in his books on the augurs: [a magistrate] should not summon men on a holy day; if he calls them, let there be a propitiatory sacrifice.’ (vitabant etiam ferias sicut Varro in augurum libris scribit in haec verba: viros vocare feriis non oportet; si vocavit, piaculum esto.) This citation of libri augurum creates uncertainty: there was a liber de auguribus (book 3) in the Ant. div., but Cardauns 1976: 176 argues for this fragment’s placement in the eighth book of the work. Macrobius’s practices of citation in this part of the Saturnalia are otherwise generally precise: see Rüpke 2011b: 94. 67

Ling. 6.30: ‘The praetor who makes a decision on that day, if he did it accidentally, is purified by an expiatory sacrifice; if he did it on purpose, Quintus Mucius used to say that he was impious and could not be purified.’ (praetor qui tum fatus est, si imprudens fecit, piaculari hostia facta piatur, si prudens dixit, Q. Mucius aiebat eum expiari ut impium non posse.) A parallel passage in Macrob. Sat. 1.16.11 makes clear that this was a responsum by Quintus Mucius Scaevola. See Tellegen-Couperus 2012: 158–63. 68

Ant. div. fr. 44 = Tert. Ad nat. 1.10.14; and Tert. Apol. 5.1: ‘there was an ancient decree that a god should not be consecrated without senate approval. Marcus Aemilius is proof of this, with regard to his god Alburnus’ (vetus erat decretum, ne qui deus ab imperatore consecraretur, nisi a senatu probatus. Scit M. Aemilius de deo suo Alburno). This decretum should be pontifical decretum: see the parallels with the pontifical responsa at issue in Cicero’s Dom. 136.

69

Festus 204 L: M. Varro ait opima spolia esse, etiam si manipularis miles detraxerit, dummodo duci hostium … non sint ad aedem Iovis Feretri poni, testimonio esse libros pontificum; in quibus sit: pro primis spoliis bove, pro secundis solitaurilibus, pro tertiis agno publice fieri debere; esse etiam Pompili regis legem opimorum spoliorum talem: ‘cuius auspicio classe procincta opima spolia capiuntur, Iovi Feretrio darier oporteat, et bovem caedito, qui cepit aeris CC … secunda spolia, in Martis ara in campo solitaurilia utra voluerit caedito … tertia spolia, Ianui Quirino agnum marem caedito, C qui ceperit ex aere dato. cuius auspicio capta, dis piaculum dato.’ I take the accusative with infinitive constructions to depend on ait and thus report Varro, but the lacunae mean that this must remain simply a likely suggestion. See Rüpke 1990: 220–23 on this text. This text is not included in Cardauns’s edition of the Ant. div., but, as Harrison 1989: 410 points out, its subject matter fits with that work. 70

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 45

14/11/2017 14:56:14

46

BICS-60-2  2017

lamb for the third spolia.’ He then recorded a law of Numa (lex Pompili regis) that supports this distinction in sacrificial animals and explains the different requirements of payment in bronze and to which deities the spolia should be dedicated. In both cases, the language used conforms to the language of Roman legislation. Festus’s text — despite its unfortunate state — shows how Varro collected these normative texts. But where did he find them? Scholars have debated the meaning of libri pontificum since the nineteenth century, but a strong consensus now holds that these were not primary records of the college, but learned works on pontifical law.71 Similarly, we have seen that Granius Flaccus collected Numa’s legislation in the de iure Papiriano. In other words, Varro’s interest in these normative texts was most likely mediated through other works on sacral law. This passage, then, allows us to see Varro as himself a writer on ritual law and as a reader of other works of the same type. Varro’s reading on the spolia opima was not an isolated case: we have other fragments that indicate that Varro read and talked with jurists on topics of pontifical law. Nonius quotes him saying that he read the details of the preparation of the salty muries, a sacrificial dish prepared by the Vestal Virgins, in an old commentary (vetus commentarius) by Fabius Pictor.72 Elsewhere, Macrobius reports that Varro said that he used to hear a man very learned in pontifical law say that it was permitted to repair ditches on the festival days, but not to dig new ones (ius non esset). By analogy, widows were allowed to marry on these days, but virgins were not.73 In Macrobius, this jurist’s name is given as Verrius Flaccus, but this Augustan name is an obvious error on the grounds of chronology. In this case, the jurist’s actual identity must remain uncertain — the two most likely candidates are Granius Flaccus and Veranius — but we can at least get a sense here of how Varro presented his interactions with these experts. Finally, in the De lingua Latina, admittedly a different work, but one that had explicit overlaps with material in the Antiquitates, Varro gave the etymology of pontifices propounded by Quintus Mucius Scaevola himself, who claimed that the word derived from the verbs posse (‘can’) and facere (‘act’).74 In this case, again, it seems likely that transmission was oral rather than written — this is the best way to understand Varro’s use of the imperfect verb dicebat — and he again offered his own better suggestion for the etymology of the word, from pontem facere (‘to build a bridge’).75 To sum up, fragments from Varro’s Antiquitates rerum divinarum and related works suggest that Cicero’s characterization of the work as, in part, an exposition of sacrorum iura et sacerdotum was founded on a perceptible orientation of the text towards ritual and priestly law. Varro’s own conception of civil theology, possibly under the influence of Scaevola, was normative in orientation. In this vein, he collected the decreta and archaic legislation that governed ritual action and read and used works of sacral jurisprudence. Beyond this, there are signs that his work shared the interests and material of the wider late Republican learned discourse around pontifical and augural ius. See Rohde 1936: 19–21, Linderski 1985b: 220, and Scheid 1994. Note the citation to Varro’s libri augurum in Macrob. Sat 1.16.18 (see above). 71

Non. 223 M = Ant. div. XIII App. (a): ‘Varro: “I read this in an old commentary by Fabius Pictor: muries is made out of salt, because humble salt is crushed and then thrown into a rough clay jar.” (Varro: in conmentario veteri Fabi Pictoris legi: muries fit ex sale, quod sale sordidum pistum est et in ollam rudem fictilem adiectum est.) 72

73 Macrob. Sat. 1.15.21 = Ant. div. VIII App. (f): sed Verrium Flaccum iuris pontificii peritissimum solitum dicere refert Varro, quia feriis tergere veteres fossas liceret, novas facere ius non esset. ideo magis viduis quam virginibus idoneas esse ferias ad nubendum. 74

Ling. 5.83: pontufices, ut [a] Scevola Quintus pontufex maximus dicebat, a posse et facere.

75

On oral transmission and intellectual activity see Marshall, esp. section 3, in this volume.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 46

14/11/2017 14:56:14

DUNCAN MACRAE: ‘THE LAWS OF THE RITES AND OF THE PRIESTS’

47

What would Varro’s incorporation of law into his book have looked like? In the absence of a continuous portion of the text, it is difficult to be sure, but one possibility is offered by part of the extant De re rustica. In the second book, on animal husbandry, Varro includes several actiones for the purchase of herd animals, apparently taken from the second-century jurist Manilius.76 These formulae, and comments on them, are included in the text alongside discussions, organized by species, of how to feed, breed, and care for the animals. It is tempting, then, to speculate that Varro used an analogous technique in the Antiquitates to insert priestly law into his discursive treatment of Roman cult. On this model, the Antiquitates rerum divinarum was not a juristic text per se — the later books preserved by Augustine show the distinctly philosophical and theological outlook of that part of the work — but I suggest that we take Cicero’s praise as reflective of a particularly prominent thread in what must have been a massive and varied text.77 3. Conclusion: Sacral Jurisprudence in the intellectual life of the late Roman Republic For thirty years, Elizabeth Rawson’s Intellectual Life in the Late Roman Republic has served — and will continue to serve — as an essential map of the significant and varied intellectual activities of the late Republican literati, both members of the Roman elite and their Greek teachers and assistants. However, like all maps — especially maps of knowledge — the categories of the cartographer have shaped the view of the territory: her disciplinary chapters are palpably selective and shaped by modern preconceptions.78 The subject of this essay sits in one of the blank spots on Rawson’s atlas — since it is absent or only partly discussed in her different chapters on ‘Law’, ‘Antiquarianism’, or ‘Theology and the Arts of Divination’. If we are to advance Rawson’s work on the intellectual history of Rome, we must also consider native categories for the organization of knowledge. In this case, I have argued that pontifical and augural law were salient objects of writing and debate in the late Republic and that, as Cicero advertised, this discourse was incorporated into Varro’s Antiquitates. Forms of ius were articulated from various old documents and traditions — decreta and responsa of the colleges, laws attributed to Numa — and current ritual practices. Even if it was somewhat modest in volume compared to the significant literature on civil law that started in the same period, this discourse on pontifical and augural law shows the potential capaciousness of ius as an ordering principle for the Roman tradition. In fact, the debate preserved by Cicero about the borderlines of the iura of the colleges and the ius of the populus at large was a dispute about nothing less than the spheres of the respective iura and, therefore, of the empire of ‘law’ itself. At the same time, the late Republican interest in the ius of sacra and sacerdotes can provide a new angle of vision on Varro’s contemporary ‘big book’: the Antiquitates. Instead of seeing this work as an ancestor of early modern antiquarian compendia, I have followed Cicero’s praise in the Academica by tracking the sacral-jurisprudential element in the text. Rust. 2.2.5–6 (antiqua formula); 2.3.5; 2.4.5; 2.5.10–11; 2.7.6. Ius is rarer in other parts of the work; to justify Cicero’s choice to highlight the revelation of iura, the Antiquitates would presumably have needed to be more like this first section of book 2. For an analysis of Rust. see Nelsestuen in this volume. 76

77

See Hadas (this volume) for the philosophy of Augustine’s Varro.

Rawson 1985 was admirably self-conscious about this: ‘the latter part [Rawson’s chapters on fields of knowledge] has proved hard to organize, for a period of incomplete specialization, in which the influence of certain disciplines is felt in a number of fields’ (vii). 78

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 47

14/11/2017 14:56:14

BICS-60-2  2017

48

Still, as a branch of knowledge, jurisprudence was not absolutely divided from philosophy or grammatikē or historiography, but borrowed methods and preoccupations from these other elements in Roman intellectual culture. Cicero’s laudation of the work already suggests that Varro’s Antiquitates may stand as an example of exactly this compound, as it combined an exposition of sacrorum iura et sacerdotum alongside other elements: theological, lexical, and historical accounts of Roman religious life.79 University of California, Berkeley

79

See Rüpke 2014 for the historical vision and van Nuffelen 2010 for the theological argument of the work.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:08 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 48

14/11/2017 14:56:14

VARRO’S WRITINGS ON THE SENATE: A RECONSTRUCTIVE HYPOTHESIS ELISABETTA TODISCO Abstract: On first becoming consul in 70 bc, Pompey asked his friend Varro to provide him with a manual on how to conduct a session of the senate. The manual was later lost. Varro returned to the subject decades later in one or more letters to Oppianus. Aulus Gellius reports on both stages of Varro’s composition, political assistance and literary composition, summarizing the contents of the letter to Oppianus. Here I will attempt to reconstruct the work, putting it in its context and discussing its implications. Keywords: Varro, Pompey, senate, Oppianus, Gellius

1. Introduction Pompey was consul for the first time in 70 bc. He was skilled in military affairs and latterly in ius belli ac pacis,1 but not in senatorial procedure, and so he asked his young familiaris to supply him with a manual on conducting a session of the senate. Aulus Gellius in his Attic Nights (14.7) reports as follows: Gnaeus Pompeius was elected consul for the first time with Marcus Crassus. When he was on the point of entering upon the office, because of his long military service he was unacquainted with the method of convening and consulting the senate, and of city affairs in general. He therefore asked his friend Marcus Varro to make him a book of instructions εἰσαγωγικός, as Varro himself termed it, from which he might learn (disceret) what he ought to say and do when he brought a measure before the House.2 Pompey had a trusted friend in Varro, closely allied to him since the war in Spain in 77 bc. Varro had already dedicated a book of Ephemerides to him at the time of this expedition,3 but most of all he was an expert on antiquitates. Since the Romans had no fixed written constitution, experts in antiquarian matters were indispensable. Their knowledge was to assume even greater importance during the political power games of the first century bc, when procedures hallowed by custom were set aside.4 Now, the manual in question was lost, but Varro rewrote it by treating the same material in a letter5 to a certain Oppianus, which ended up in the fourth book of the Epistolicae Quaestiones, as Gellius attests (14.7): 1

See Cic. Balb. 15.

Trans. Rolfe 1961, occasionally adapted. On Gellius’s citation of Varro see also Marshall, section 2, in this volume.

2

Della Corte 1970: 52, n. 8 and 250 relates the commentarius for Pompey to Varro’s Ephemerides, the first of which was composed in 77 bc. Moatti 1997: 337, n. 44 [= 2015: 65, n. 102] disagrees.

3

4

Rawson 1985: 232–49; Wallace-Hadrill 1997: 14–5; Pani 2006: 727–40.

5

The term epistula in Gellius’s text is sometimes singular, sometimes plural; I use the singular for convenience.

BICS-60-2  2017

49 © 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:23 +00:00 All use subject to https://about.jstor.org/terms

BICS 60-2 Nov14-2 FINAL-FINAL.indd 49

14/11/2017 14:56:14

BICS-60-2  2017

50

Varro in letters which he wrote to Oppianus, contained in the fourth book of his Investigations in Epistolary Form, says that this notebook which he made for Pompey on that subject was lost; and since what he had previously written was no longer in existence, he teaches6 again (docet rursum) in those letters a good deal bearing upon the same subject. The present article aims to reconstruct these two works: their form, structure, date, intended readership, purpose, and the implications of their composition. 2. Title As Gellius underlines, Varro himself called the earlier work Commentarius εἰσαγωγικός. The noun commentarius relates Varro’s work to those writings which assisted magistrates in carrying out their duties, such as the commentarii consulares.7 Nevertheless, Varro needs to define his commentarius for Pompey as εἰσαγωγικός — ‘introductory’ — which suggests that it imparted basic information that allowed access to more complex levels of understanding. Before Varro’s time the word was used thus in dialectics (Chrysippus, περὶ συλλογισμῶν εἰσαγωγικῶν πρὸς Ζήνωνα, and tellingly Gell. 16.8.1: ‘When I wished to be introduced to the science of logic and instructed in it, it was necessary to take up and learn what the dialecticians call εἰσαγωγαί or “introductory exercises”’), in ethics (Chrysippus, εἰσαγωγὴ τῆς περὶ ἀγαθῶν καὶ κακῶν πραγματείας), and in the science of siege warfare (Ph. Bel. 56.12, in the third century bc), sometimes even in the title of the work.8 Aside from the introductory character of the work, Varro also hints at its didactic function: in Gellius’s summary, ex quo disceret… suggests the work’s aim.9 This said, the possibility cannot be excluded that Varro intends also the term’s more literal meaning beyond the didactic: Pompey was also being introduced into the senate (Pompeius cum initurus foret). Nevertheless, on rewriting the work, Varro casts it in a different genre: as a letter to a certain Oppianus. The Epistolicae Quaestiones may have comprised eight books, and are believed to have occupied Varro on and off for his entire life.10 Gellius in his preface compares his own Attic Nights to a work entitled Epistolicae Quaestiones, and may thus suggest something about the character of Varro’s work of that title: it would be a miscellany of various writings (coniectanea), without a common thread (uariam et miscellam et quasi confusaneam doctrinam conquisiuerant).11 It has been conjectured that, unlike the Epistulae Latinae, the Epistolicae Quaestiones were literary letters which discussed

6

Rolfe translates ‘repeats’, but in my opinion this does not reflect Varro’s perspective.

Aricò Anselmi 2012: 19, nn. 9–10 evaluates the prescriptive or descriptive status of the commentarii with further bibliography; Mommsen 1887: 5–7, n. 4 asserted their prescriptive value and considered Varro’s Isagogic commentary a possible model.

7

8

Many of Chrysippus’s titles contain the noun or the adjective: see Goulet 1994, s.v. Chrysippe de Soles 337–64.

Dahlmann 1935: 1249 thinks that Varro is the first to use this genre in the Roman world, under the influence of the Stoa. Comparable is the Commentariolum petitionis, written to instruct Cicero on organizing his election campaign for the consulship in 63 bc. On the authorship of the Commentariolum, whether Quintus or an expert in elections, see Traversa 2009–10: 116, n. 4. 9

They are absent from Jerome’s list of Varro’s works. Like the Epistulae, the EQ might have collected a continuous stream of writings from Varro’s youth until after the triumviral proscriptions. See Della Corte 1970: 251, and n. 24, for the eight-book hypothesis.

10

11

On the EQ see further NA 14.7.3 and 14.8.2; and possibly 6.10.2, about which see Todisco 2016a: 481, n. 9.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:23 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 50

14/11/2017 14:56:15

ELISABETTA TODISCO: VARRO’S WRITINGS ON THE SENATE

51

different subjects from law to grammar to antiquitates.12 It is noteworthy that Gellius attests that the instructions contained in the epistula about the workings of the senate are in the subjunctive, as was the case in the Commentarii,13 which would suggest that Varro’s work had a prescriptive status.14 3. Structure It seems likely from Gellius’s account that the letter to Oppianus explained its status as a rewrite of the manual for Pompey, with some information on the occasion of the earlier version. The short history of the text is followed in Gellius’s summary by the list of the themes discussed, which could be arranged in four groups: relationship between magistrates and senate; times and places of the sessions; management of a session (topics, discussion, vote); disciplinary sanctions. Here is a more detailed synopsis: • first (primum): the right to convoke the senate (ius consulendi senatus) is due by ancestral custom (more maiorum) to dictator, consuls, praetors, tribunes of the plebs, interrex, urban prefect — according to their rank (gradus), if they are present in Rome simultaneously; then, by an extraordinary legal measure (iure extraordinario), even to military tribunes with consular power, decemviri with consular imperium, and triumviri rei publicae constituendae; • thereafter (postea): the right of intercession (ius intercessionis) over the decree of the senate (senatus consultum) is accorded only to those who have power (potestas) equal to or greater than that of the proposer; • then (tum): the senate can issue a valid senatus consultum only in spaces designated (loco… constituto) by the augurs, that is the templa, and the curia Hostilia, Pompeia, and, finally, Iulia; • immediately after that (post haec deinceps): this procedure is only valid during the hours from sunrise to sunset, and if this rule is broken, those responsible for breaking the rule are to be sanctioned; • then (deinde): days when it is not lawful (ius) to hold senatorial sessions; the convener has to immolate a victim and take auspices before the session; among the topics to be discussed, res divinae precede res humanae; one may discuss general questions about the res publica, or specific cases; votes are expressed by stepping to one side or other of the House (discessio) if everyone is in agreement; otherwise, everyone is asked his opinion in hierarchical order, beginning from senators with the rank of consul; on that point, Varro says that in former times the first to express his opinion was the Leader of the House (princeps senatus), thereafter, according to an innovation in custom (novum morem), this privilege is granted to a person chosen by the princeps senatus on the basis of personal or political relationship, provided that that person is of consular rank; • apart from this (praeter haec): penalty to be imposed on absentee senators. Della Corte 1970: 251 follows Cugusi 1967: 83 in assimilating the Epistolicae Quaestiones to Valgius Rufus’s De rebus per epistulam quaesitis; see also Rawson 1985: 231; Traglia 1993: 878–9; and Dahlmann 1935: 1226. 12

13

Cf. Varro’s quotations from the Commentarius anquisitionis M. Sergi (Ling. 6.90), to which I return below.

14

See MacRae in this volume for Varro’s normative precepts on augural law and for Varro as jurisprudent.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:23 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 51

14/11/2017 14:56:15

BICS-60-2  2017

52

Gellius’s summary omitted other contents (haec et alia quaedam id genus in libro) probably related to the same themes. Moreover, the manual must have contained a detailed description of procedures and formulas to be declaimed by the magistrate. Varro had full access to this kind of information: we know this from a passage in De lingua Latina, where, talking about the inlicium,15 a convocation of an assembly, he reports what a censor (6.86– 87), a consul (6.88), and a quaestor (6.89) have to do in a capital case, and he quotes the sources he is drawing on, the tabulae censoriae, the consular commentarii, and the Commentarius anquisitionis M. Sergi respectively.16 Besides, the aforementioned use of the subjunctive is a vestige of such prescriptive sources or of the intention to evoke sources of this kind. Gellius’s list (primum, postea) suggests he was following the scheme of the epistula, reproducing its divisions. One final but extremely notable feature of Varro’s treatise, paralleled in his other works, is the assiduous attention to the historical development of procedure: in this case in relation to the right to convene the senate (more maiorum […] deinde extraordinario iure), or where to hold a session (propterea et […] et post), or in which order to consult the senators (antea […] tum autem, cum haec scriberet, novum morem).17 4. Dating In the letter to Oppianus, Varro says that, having lost the manual for Pompey, he returned to the same subject matter, but he does not say when or why. The Commentary for Pompey may have gone missing after Pharsalus18 and we may imagine that Varro’s private copy went missing along with other precious books amid proscriptions after 39 bc.19 It is virtually impossible that Varro would have reproduced years later the same content as what he had written for Pompey, especially in light of all the political and institutional changes that had occurred, and particularly given Varro’s interest in changes over time (cf. Ling. 6.95, quoted below, n. 17). Despite this, scholars have generally taken the view that Varro’s second work on the senate corresponded to the version of 70 bc.20 Such a view has sparked debate, especially in relation to public law or institutional procedures, because as we shall see, it has been presumed sometimes that Varro misunderstood or omitted something about senatorial procedures. In fact Gellius’s text does provide considerable evidence that Varro’s rewrite involved significant updating. Among the officials permitted to convene the senate he Giovannini 1993: 85–6 relates these procedures to augural law, saying that they combine ‘precetti rituali sugli auspicia e regole costituzionali’. He also thinks (83) that Varro’s volume for Pompey comes from an augural book about the senate. 15

Santalucia 1994: 50–56 sees in the first extract a handbook of good practice for quaestors. On this passage by Varro, see the recent commentary by Aricò Anselmo 2012: 1–94. 16

On Varro’s interest in historical research, see Moatti 1997: 114–15, 140 [= 2015: 111–12, 148]. See also Piras and Nelsestuen in this volume; cf., e.g., Ling. 6.95: et dicis causa fieba[n]t quaedam neque item facta neque item dicta semper. 17

Dahlmann 1935: 1249 thinks that the work was written just for Pompey and that it was not intended to be published; that could explain its loss. 18

19

Della Corte 1970: 243; Osgood 2006: 209.

E.g. Bonnefond-Coudry 1993: 122–23; Ryan 1998: 254 (‘the contents of his letters to Oppianus accurately reflect the substance of the Commentarius he wrote for Pompeius’); Rafferty 2011: 11, n. 35 (‘However, Varro does mention that some revisions have been made; my argument rests on the assumption that the relevant material was in the original manual’). 20

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:23 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 52

14/11/2017 14:56:15

ELISABETTA TODISCO: VARRO’S WRITINGS ON THE SENATE

53

mentions the triumviri rei publicae constituendae causa, an office which dates from 43 bc; among the locations where senatus consulta could be issued he mentions the curia Iulia, which Octavian did not inaugurate until 29 bc.21 It has been presumed that these elements were updated by Gellius, but nothing in the text supports that hypothesis:22 Gellius does not intervene in any of the much-disputed questions about Varro’s representation of the senate, although he himself could have provided all the missing information; on the contrary, he discusses these questions elsewhere, or he refers the reader to his own particular discussions elsewhere (e.g. 14.7.12–13; 14.8). Moreover, he consistently ascribes to Varro’s own text the content that he is reporting (sic enim ipse Varro appellat; Varro ait; ponit; scripsit; adscripsit; docet; disserit), and several phrases suggest his fidelity to Varro’s text (tum autem, cum haec scriberet). It emerges, then, that Varro’s Epistula ad Oppianum cannot be earlier than the second half of 29–early 28 bc.23 This chronology helps to explain some other passages of the work that hardly make sense without distorting their content. We consider here two examples: who has the right to speak first in the senate (ius primae sententiae), and on what days it is not ius to convene the senate. Gellius reports that Varro had the following to say about the ius primae sententiae (14.7.9): furthermore the senators ought to have been asked their opinions in order, beginning with the rank of consul. And in that grade in former times the one to be called upon first was always the one who had first been enrolled in the senate; but at the time when he was writing he said that a new custom had become current, through partiality and a desire to curry favour (per ambitionem gratiamque), of asking first for the opinion of the one whom the presiding officer (qui haberet senatum) wished to call upon, provided however that he was of consular rank. Varro says that in former times this ius was accorded to the princeps senatus, but that at the time when he was writing the presiding officer chose whom to invite to speak first per ambitionem gratiamque.24 Gellius had previously dwelt upon this subject with reference to the time before the passing of the lex Iulia de senatu habendo in 9 bc (4.10.1–4):25 Before the passage of the law which is now observed in the proceedings of the senate, the order in calling for opinions varied (ordo rogandi sententias varius fuit). Sometimes the man was first called upon whom the censors had first enrolled in the senate (princeps a censoribus in senatum lectus), sometimes the consuls elect (qui designati consules erant); some of the consuls, influenced by friendship (studio) or some personal relationship (necessitudine aliqua), used to call first upon anyone they pleased, as a compliment, contrary to the regular order (extra ordinem). However, The inauguratio is reported for 28 August 29 bc (D.C. 44.5.1–2). About the curia Iulia (the phases of its construction and its relationship with politics) see Tortorici 1993: 332-33. The senate used the curia Pompeia for its sessions from 55 bc to 44 bc, when Caesar was killed there; after that it was closed and turned into a latrine (D.C. 47.19.1). On places where the senate used to convene in general, see Bonnefond-Coudry 1989: 25–197. 21

On Gellius’s respect for Varro, see also Baldwin 1975: 76–77; on Varro in Gellius, Astarita 1993; see also Marshall, section 2, in this volume. 22

23

On the dating see also Kumaniecki 1974–5: 41–42.

24

See also Tansey 2000: 18, n. 74.

25

See Ryan 1998: 254–56 for a commentary on Gellius’s account of the workings of the second-century senate.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:23 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 53

14/11/2017 14:56:15

BICS-60-2  2017

54

when the usual order was not followed, the rule was observed of not calling first upon any but a man of consular rank. He goes on to mention Julius Caesar, who as consul departed from the order on just four occasions, first for Crassus, then for Pompey who had become his son-in-law. Gellius mentions as his sources Tiro’s Commentarii, which claimed to rest on the authority of Cicero, and Ateius Capito’s De officio senatorio. One might surmise that the procedural changes mentioned came about at around the same time. By contrast, two particular elements in Varro’s version have attracted the attention of scholars from Mommsen to Willems and beyond: the assertion that the right of the princeps senatus to speak first belonged to the past (antea),26 and the fact that no mention is made of the consules designati.27 Gellius was well familiar with the issues and with Varro’s work, and emphasizes that the new mos was contemporary with the time of Varro’s writing to Oppianus. As I have discussed elsewhere, some political and institutional changes may help to explain Varro’s version of events.28 The previous lectio senatus before that held by Octavian in 29 bc had been in 50 bc, and this implies that there would not have been a princeps senatus until 28 bc, when the title of princeps senatus was bestowed on Octavian. From 43 bc onwards, procedures for electing magistrates were repeatedly subverted, which led to a complete delegitimization or significant disempowerment both of election procedures and of elected magistrates.29 Moreover, from 31 bc, Octavian would continuously be consul.30 As for the custom that the consul-elect should speak first, it is noteworthy that in 43 bc Cicero (Phil. 5. 35) speaks of this procedure as confirmed by mos maiorum; he seems to polemically recall a practice fallen into disuse, though theoretically still in force: ‘But I shall maintain the accustomed order in soliciting opinions (qui ordo in sententiis rogandis servari solet) … and so in accordance with ancestral custom let us begin with Brutus, the consul-elect (a Bruto igitur, consule designato, more maiorum, capiamus exordium)’.31 Given Octavian’s assumption of the title of princeps senatus in 28, we may assume this moment as a t.a.q. for the rewriting of the Commentarius. Secondly, Varro’s rewriting referred to days on which it was not lawful (ius) to convene the senate. These prescriptions have generally been understood as a reference to mos rather

Mommsen 1888: 967 (followed by Badian 1996: 1247) ascribed the eclipsing of the princeps senatus to Sulla. Bonnefond-Coudry 1993: 122–23 believes that the change happened in the 70s spontaneously and for political reasons, reflecting the high level of competitiveness among the Roman ruling class, and that the crisis in the ancient system ran parallel to the censors’ loss of status in the senate. Tansey 2000: 15–22 also disagrees with Mommsen 1888: 969–70. Rafferty 2011: 1–22 discusses the entire bibliography and concludes that the sidelining of the princeps senatus was gradual. 26

Ryan 1998: 259–92 investigates the development of this ius, engaging with Willems 1883: 181–84 and Mommsen 1888: 973, n. 2, especially in relation to the relative priority of princeps senatus and consul designatus, which Varro omits but Gellius and Cicero discuss. He concludes (1998: 256) that the princeps senatus probably had priority if the consul designatus was absent; according to him, before 71 bc a senatus consultum had given the consulares permission to speak; according to Tansey 2000: 25, C. Cotta proposed the measure (Sall. Hist. 2.42). 27

28

Todisco 2016a: 479–88.

See De Martino 1974: 141–42; for example, the arbitrary choosing of the consuls by the triumvirs (D.C. 48.43.1– 2; 53.2), or the eight consuls nominated in 33 after five changes (see Broughton 1952: 413). 29

30

See De Martino 1974: 145–53.

Mommsen 1888: 973, n. 3, regarded this as a rhetorical expression by Cicero corresponding to Varro’s novus mos; Ryan 1998: 254 thinks that Mommsen misunderstood, because Varro’s novus mos concerns consulares, while Cicero is speaking of a consul designatus. 31

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:23 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 54

14/11/2017 14:56:15

ELISABETTA TODISCO: VARRO’S WRITINGS ON THE SENATE

55

than legal measures.32 No laws governing the days on which the senate should meet are known to have been passed as early as 70 bc, and so if we dated the essence of Varro’s rewrite to this date, then it would be natural to assume that Varro was referring to mos. But what if we were to date the rewrite to late 29 or early 28 bc? The lex Iulia de senatu habendo (9 bc) decreed that the senate should convene twice a month, on the Ides and on the Calends (Suet. Aug. 35; D.C. 55.3.1–3). Prior to this there had been two laws on the scheduling of senatorial activity: the lex Pupia and the lex Gabinia, both later than 70 bc. The first one, whose text is uncertain but was surely intended to avoid overlaps between sessions of the senate and of the assemblies, probably dates to 61 bc; the second one, which decrees that the senate ‘should be reserved for senatorial embassies’ in the month of February, dates very probably from 61 or 58 bc at the latest.33 Theodor Mommsen dated the lex Pupia to the second half of the second century bc, so he thought that Varro was referring here to this law.34 However, research on the dates of senatorial sessions shows that even before the lex Pupia there was a preference for not convening the senate on dies comitiales.35 Equally, prior to the lex Gabinia it was the norm to hold senatorial embassies in February.36 Therefore, presumably these procedures were customary in Rome, in the first case because comitia and senate could not overlap, in the second case probably because in that period both consuls were in Rome. Nevertheless, during the first century bc there was a need to reassert those customs by means of law. A remark by Dio on the intervention of the tribune Cornelius in 67 bc about solutio legibus (‘exemption from statutes’) exemplifies the principle (36.39): ‘This indeed had been the law from very early times, but it was not being observed in practice.’37 The lex Pupia and the lex Gabinia were probably intended as a contribution to the same end. Gruen has rightly noted that friends of Pompey led both initiatives.38 To sum up, it is highly improbable that Varro referred only to customary law when addressing the subject of days on which it was not ius to convene the senate in his rewrite of the Commentarius; on the contrary, it is highly probable that the part summarized by Gellius’s statement (‘on what days it was not lawful to hold a meeting of the senate’) either contained or was followed by detailed specification of those days and related laws.39 Varro will not have returned to a subject to repeat slavishly what he said forty years earlier, but rather will have wanted to update it.

32

Michels 1967: 45; Bonnefond-Coudry 1989: 229.

The lex Gabinia is still much debated; Carcopino 1932: 117–32 and Monaco 1996: 130–56 think that the lex should be assigned to Gabinius, tribune in 139 bc; Mommsen 1888: 1156 and Bonnefond-Coudry 1989: 333–47 (with ample details) opt for 67 bc; Willems 1883: 154 for 61 bc; Gruen 1974: 251–52 for 58 bc. Todisco 2018a discusses the different views and proposes 61 bc for both laws. 33

34

Mommsen 1888: 922–3.

35

Bonnefond-Coudry 1989: 220–21.

The problem of the reception of the embassies by the senate is related with the problem of the corruption of the magistrates who scheduled the audiences, cf. Bonnefond-Coudry 1989: 336–40.

36

The same idea about solutio legibus can be found in Asconius’s commentary to pro Cornelio 58C; see Lewis 2006: 260 (with bibliography) on the differences between Dio’s and Asconius’s versions.

37

38

Gruen 1974: 252.

39

See Todisco 2018a on the text, context, and dating of the lex Pupia and the lex Gabinia.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:23 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 55

14/11/2017 14:56:15

BICS-60-2  2017

56

5. Author and purpose Varro was eighty-five years old in 31 bc, when he was working on the ninth book of the Disciplinae.40 Why, at this phase of his life in the second half of 29–early28 bc, would he return to this particular old subject?41 The intervening years between the two versions had been ravaged by political turbulence, and the second certainly mirrored the changed circumstances. Some of Varro’s political positions are known, while others can only be conjectured.42 A reader of Gellius’s summary could hardly fail to notice that in the letter to Oppianus Varro was expressing a certain distance from new practices which had become customary, either by condemning them or advising remedies, explicitly or implicitly, such as the matter of who has the right to speak first, and the notice of penalties for absentee senators, both of which imply the degeneration of the curia. Varro must have been affected by contemporary political tensions when writing his manual for Pompey in 70 bc. There had been conflict between senate and populus since the mid-second century bc, and Sulla as dictator had reinforced the senate’s power against the institutional populus. After Sulla’s retirement there were many efforts to undo the mortification of popular action. Thus, the lex Aurelia de tribunicia potestate (75 bc) intended to restore the authority and privilege of the tribune of the people, followed by the lex Pompeia Licinia de tribunicia potestate, proposed by Pompey and Crassus in 70 bc. In preparation for this decisive act, and probably also because of his unusual path to magistracy (App. B. Civ. 1.131; Val. Max. 8. 15. 8; Livy Per. 97), Pompey asked his familiaris Varro to write the Commentarius for him. Varro had been by his side in the war against the pirates, when a corona navalis was bestowed on him, and maybe also when he fought against Mithridates.43 I have already mentioned the laws passed after 70 bc against the senate’s disrespect for the customary privileges of the populus: the lex Cornelia de legibus solvendis, the lex Pupia de senatu diebus comitialibus non habendo, the lex Gabinia de senatu legatis dando. One may surmise that in a similar spirit Varro recorded in his Commentarius canonical procedures for managing the senate confirmed by mos, while he stressed or condemned new practices as illegitimate.44 The booklet which helped Pompey in 70 bc may also have served as a model for later revisions of senatorial procedure. Varro remained close to Pompey until the end of the general’s life.45 Letters from Cicero to Atticus confirm their strong link: Atticus informs Cicero about Pompey’s political and strategic intentions, even on the eve of civil war, quoting Varro as his source.46 Besides, it is known that he dedicated an entire work to Pompey, perhaps a eulogy.47 40

Dahlmann 1935: 1178.

Moatti 1997: 160 [= 2015: 165] relates Varro’s work for Pompey to Cicero’s and Sallust’s works for Caesar, with reference to Cicero’s letter to Varro in 46 bc (Fam. 9.2.5); see McConnell 2014: 191. Bretone 1971: 17–18 relates Varro’s Isagogic commentary to the same genre as that of Cincius Alimentus’s Liber de consulum potestate and M. Valerius Messalla Rufus’s Libri de auspiciis. 41

42

For Varro’s biography, see Della Corte 1970; on his political positions, see now Wiseman 2009: 109–29.

43

Wiseman 2009: 113, with reference to ancient sources.

Wiseman 2009: 116 argues that Varro’s manual portrayed a person subordinate to the authority of the populus, and sought to bring the senate back under popular authority. 44

45

Wiseman 2009: 117–25.

E.g. in a letter to Atticus dated 17 August 58 (3.15), Varro informs Atticus about Pompey’s intentions and possibly also about an intervention by Caesar; in a letter dated 5 July 51 (5.11), Varro again informs him about Pompey’s intention to go to Spain. On Varro as a Pompeian see also Marshall, section 4, in this volume. 46

47

Drummond 2013b: 419; Taylor 1934: 227 thinks it may have been a defence of Pompey’s actions against Caesar.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:23 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 56

14/11/2017 14:56:15

ELISABETTA TODISCO: VARRO’S WRITINGS ON THE SENATE

57

All the data we have about Varro’s political position make clear his liking for civitas popularis (with reference to Cicero’s formulation in his De republica), a res publica where the populus is assigned total centrality. Wiseman in his Remembering the Roman People (2009) rightly underlines decisive passages on that subject from Varro’s works: for instance, in book 21 of the Antiquitates humanae he recalls a citizen’s right to arraign a magistrate (vocare in ius, Gell. 13.13.4); in De lingua Latina 9.6 he assigns sovereignty (potestas) to the populus as far as linguistic correctness is concerned; he uses uncommon and therefore significant expressions such as senatus populi Romani (‘the senate of the Roman people’, Gell. 17.21.48, quoting Varro’s De poetis), augures publici (‘augurs of the people’, Ling. 5.33). I should like to add several brief comments relevant to the subject of this paper to Wiseman’s persuasive argument, with particular reference to Varro’s political outlook especially from the Ides of March to the early 20s, when he rewrote his Commentarius. The years between the defeat of Pompey and Caesar’s death witnessed Varro’s natural transition from Pompey to Caesar.48 Caesar was presumably not lying when he wrote in his De bello civili (2.17.1–3) that Varro had spoken of his close bond (necessitudo) to both Pompey and Caesar.49 Varro’s dedication of his Antiquitates divinae to Caesar as pontifex maximus50 and the office of public librarian assigned to him by Caesar in 46 bc also reveal the mutual respect between the two, given Varro’s long adherence to partes populares and to Pompey.51 Caesar’s death and the political earthquake that ensued opened a new phase in Varro’s life. Wiseman is right to think that a profound interest in history came about during the eighteen months between Caesar’s death and the war against his assassins.52 Most likely around 43 bc, Varro wrote his De vita populi Romani, a work dedicated to Atticus and closely related to his De gente populi Romani, undoubtedly written after 43 bc.53 Varro chooses to put the populus Romanus at the centre of these two works for ideological and political reasons of great contemporary resonance.54 The fragments of these two works clearly show their deep link to contemporary history: in them Varro attempted to suggest a solution for the crisis of his age.55

Wiseman 2009: 124–125 points out that after Pompey’s death, the optimates did not consider Varro as one of them; see also Drummond 2013b: 419. 48

Wiseman 2009: 116 notes that in 59 bc Caesar was, of necessity, consul of the people. For Varro’s closeness to Caesar, see Cicero’s letter to Atticus of 17 August 58 bc (3.15.3).

49

50 There is still intense debate on the dating of the work, between the late 50s and 47 bc (see Drummond 2013b: 415); Wiseman 2016: CXXVI puts the work in the 50s.

On the vicissitudes of Varro’s political allegiance, see Della Corte 19702: 117–34; Wiseman 2009: 123–25; Drummond 2013b: 414–15. I shall return to this subject in a forthcoming work.

51

52

Wiseman 2016: CXI.

Two factors are essential to the dating of De vita populi Romani: first, it mentions that the civil war between Caesar and Pompey is over; second, it is closely related to De gente populi Romani, which dates to the consulship of Hirtius and Pansa in 43 bc (fr. 9 Peter). Della Corte 1970: 238 opts for 43 bc; Rawson 1985: 242 for the late 40s; Purcell 2005: 17 for the years from 47 bc onwards, although he specifies that some date it after Varro’s De gente populi Romani; Drummond 2013: 417 for the period between 49 and 32 bc (the year of Atticus’s death); Pittà 2015: 7 for 43–42 bc; according to Wiseman 2016: CXI, n. 6, Varro’s De vita populi Romani follows his De gente populi Romani. On Varro’s friendship with Atticus see Marshall, esp. section 3, in this volume. 53

54

Purcell 2005: 17, consistent with Wiseman 2009: 107–29.

On these works’ historical and not just antiquarian value, see Drummond 2013: 417; see also Todisco 2016b: 489–97; Wiseman 2016: 128. See also Piras in this volume. 55

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:23 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 57

14/11/2017 14:56:15

BICS-60-2  2017

58

In his De gente populi Romani, Varro inserts Rome into universal history, which is traced back to the flood at the time of Ogyges. It seems likely that this work was influenced by contemporary political issues and that it contributed to Octavian’s controversial efforts to have Caesar deified post mortem by recording deified mortals from Greek and Roman history, including Aeneas, Aventinus, and Romulus.56 The populus is here central: the apotheosis of kings would surely have benefited the people, and so while the precedents could be read as general, they could also have alluded specifically to Caesar, not least because of his democratic policies.57 Nonetheless, this will not have been Varro’s main aim: in both De gente populi Romani and De vita populi Romani he was much more interested in reinstating the Roman people as heir to ancient history and culture.58 In his four-book De vita populi Romani Varro portrays the society, institutions, and customs of the Roman people from a physiological point of view from the origins of Rome to the present decay. In the final book he denounces the decadence of the present time, a time scarred by the horrible civil war between Caesar and Pompey, by political and institutional squalor, by the corruption of magistracies at the hands of power-hungry individuals, when the body politic has bloody gangrene (gangraenam sanguinulentam) spreading through all its limbs (per omnes articulos populi, fr. 123R=437S=117P).59 Against this present decadence Varro opposes Rome’s venerable past with its political, institutional, and social probity, hallowed by mos maiorum. Varro envisages an ethical, political, and social recuperation based on a return to the moral probity of the past.60 The state (civitas), fractured since the time of Gaius Gracchus (114R=425S=108P),61 is to be reconstituted (124R=438S=106P) in civic concord (concordia civilis). In this project of healing, Varro draws on his encyclopedic knowledge of antiquarian matters, his juridical skills, and his experience as a man and as a politician. He addresses himself to a wider audience than he had done in the Antiquitates62 to transmit his ethical and political solution. Etymology could play a role in his project, and to take one example relevant to my chapter, he gives an etymology of curia that differs from that given in De lingua Latina, which characterizes the senator as dedicated to the res publica both inside and outside the curia.63 After the death of Caesar, as the state was being torn apart, Varro denounced in the pages of this work every form of conflict which contributed to the dismemberment of the state or which compromised the Roman people, and propounded a return to a unified state (civitas una) in which civic concord (concordia civilis) would reign.

Taylor 1934: 221–29 includes (227) this work in the ‘propaganda literature’ produced by Octavian’s and Antony’s supporters during the civil-war years. On the treatment of Trojan elements in the history of Rome see Marshall in this volume. 56

57

Taylor 1934: 225.

Gabba 1996: 92–93; Drummond 2013: 417. Taylor 1934: 229 argues that it was De gente populi Romani which saved him from proscriptions; see also Della Corte 19702: 209 58

59

At least since the time of the Human antiquities, magistrates had been immune from citizen arrest (Gell. 13.13.4).

60

On this moral probity see Pittà 2015: 11; for a summary see Todisco 2016b: 489-97.

I highlight here Wiseman 2009: 90–91 and 2016: CXXIII on C. Gracchus’s choice of installing on the tribunals men of equestrian rank who had not been vetted by popular legitimization. 61

62

See Drummond 2013: 417–18.

63

Contrast fr. 70R = 385S = 69P with Ling. 6.46; see Todisco 2016b: 489-97.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:23 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 58

14/11/2017 14:56:15

ELISABETTA TODISCO: VARRO’S WRITINGS ON THE SENATE

59

Appian tells us that Varro was proscribed perhaps for his opposition to monarchy, but saved by the protection of Fufius Calenus, an adherent of Antony (BC 4.47).64 Remarkably, significant evidence of Varro’s importance after his survival was the fact that his statue, the only one of a living person, stood in Asinius Pollio’s public library (Plin. HN 7.115), which Varro may have helped Pollio organize.65 It was amid this atmosphere that Varro returned in his letter to Oppianus to the subject that he had previously treated in 70 bc. It would be interesting to know more about his attitude to Octavian, about whom he had initially been sceptical (Cic. Att. 16.9, November 44 bc). A living monument to erudition and survivor of two civil wars and proscriptions, Varro may have seen in Octavian an opportunity to further his own project of restoring the Republic of the Roman populus (restitutio rei publicae populi Romani).66 Along the way he contributed his own authority. 6. Addressees The Commentarius isagogicus was addressed to Pompey in 70 bc, but who was Oppianus, the addressee of the rewritten version from 29–28 bc? It would seem probable that Varro would address such a letter to a politician or a man with political and institutional interests who needed exhaustive information about senatorial procedure, including recently introduced practices.67 But there is no record of a contemporary Oppianus. One might postulate a corruption in textual transmission — Varro complained of imprecise copyists (Ling. 8.51) — and think of Oppius, who might have been an acquaintance of Varro. Contemporary with Varro there was C. Oppius, an active politician whose date of death is unknown, Caesar’s familiaris, and a correspondent of the dictator, a recipient of decisive commands, and later close to Octavian. C. Oppius was still active in 31 bc, when he presented a report which proved that Caesarion was not Caesar’s son, in response to Antony’s question in the senate (Suet. Iul. 52. 2).68 Besides, he himself wrote significant biographies, and Aelius Tubero dedicated an otherwise unknown work to him (Gell. 6.9.11).69 One should not be too hasty to use these details to push for an identification with Oppius, not least since Gellius’s twofold mention of Oppianus reduces the likelihood of textual corruption.70

On the reasons for Varro’s proscription see Della Corte 1970: 201–03; Drummond 2013: 415; Wiseman 2016: CXXVII.

64

65

Osgood 2006: 252.

Todisco 2007: 341–58; Mantovani 2008: 5–54; Wiseman 2009: 131–32. An aureus dated 28 bc reads Leges et iura P. R. restituit: Lacey 1996: 85 and n. 36; Rich and Williams 1999: 169–21 = Bringmann 2002: 187=Zehnacker 2003: 1–3; besides, the Fasti Praenestini read Corona querc[ea a senatu, uti super ianuam Imp. Caesaris]/ Augusti poner[etur, decreta quod cives servavit, re publica]/p(opuli) R(omani) rest[itu]t[a]: CIL I2 231=Inscr.It. XIII,2,17, Jan. (conjectural integration by Todisco 2007: 341–58); on the recurrent populus Romanus in Augustan propaganda, see Wiseman 2009: 131 and n. 1. 66

According to Della Corte 1970: 251, the addressees of Varro’s Epistulae Latinae were politicians: Caesar, Fabius, Fufius Calenus, and Marullus.

67

68

‘C. Oppius’, in RE XVIII 1, 9, 729–36.

Rawson 1985: 97 and n. 75 points out that Tubero, father or son, was also the protagonist of a logistoricus by Varro.

69

There is a slim possibility of textual corruption in Gellius’s time, when an Oppianus Periegetes was very well known. Alternatively, Oppianus could be an adopted son of Oppius. One more suggestion (highly improbable in light of the letter’s subject matter): Oppius Chares (RE XVIII 1, 26, 743), vir doctus (Suet. Gram. 3), author of a treatise De silvestris arboribus (Festus 184L. 274L; Macrob. 3.18.7).

70

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:23 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 59

14/11/2017 14:56:15

BICS-60-2  2017

60

I should like to recall another hypothesis which Dahlmann considered probable based on the presumed literary nature of the Epistolicae Quaestiones, namely that Oppianus could have been a fictional character: Varro could have concocted an allusive name based on Oppius and the suffix -anus.71 Thus, Varro would be referring indirectly to Oppius’s political circle, which included the young Octavian. However, without further support, this suggestion must remain a tentative hypothesis, not least since all other addressees of Varro’s letters are real men. 7. Conclusion To sum up, the identity of the addressee of Varro’s letter remains obscure. Whoever Oppianus might have been, the letter ended up serving Octavian’s political purposes. In 29 bc Octavian began his first of several lectiones senatus in front a thronged house comprising more than a thousand senators, which concluded in 28 bc.72 This was followed by many measures which involved the senate and which resulted in the lex Iulia de senatu habendo.73 When Octavian and his entourage set about reforming the senate, Varro’s work for Pompey was the first and only summary of procedure, of what Pliny the Younger would later call ius senatorium (8.14, letter to Aelius Ariston).74 It seems likely that someone in Octavian’s circle who was aware of Varro’s manual for Pompey called upon him to reproduce it, and that the illustrious investigator antiquitatis, perhaps already involved in the project of restoring the Republic, not only rehashed the old material (docuit rursum) but also gave an account of what had happened in the intervening forty years, combining his antiquarian and institutional knowledge with all his political nous, since he had once participated actively in the life of the institutions.75 Varro probably also contributed to the Augustan institutional project after his own death: his work was read by jurisprudents from Augustus’s group and contributed to the reorganization of the senate.76 Varro’s writings became essential for the project of restitutio rei publicae, as far as revival of mos and normalization of institutional life were concerned. He would also contribute to the ideological — and even physical — construction of the city.77 Dipartimento di Studi umanistici, Università degli Studi di Bari ‘Aldo Moro’, Bari (Italy)

It is difficult to establish whether this exchange of letters was real or not: Talbert 1984: 23 quotes Dahlmann 1935: 1225–26, who thinks that the hypothesis of a fictional addressee is wahrscheinlicher. 71

72

Chastagnol 1992: 15–29; 49–56; Talbert 1984b: 55–63; Bonnefond-Coudry 1989: 411–13.

According to Talbert 1984a: 222–24 and 488, this law was intended to normalize the situation, so that the princeps could exercise total control over institutions; on this law see Bonnefond-Coudry 2014; Scott Perry 2014: 51. 73

74

Bonnefond-Coudry 1989: 18 thinks that no treatise on how the senate worked existed in the Republican era.

75

Dingmann 2007: 40–41; Wiseman 2009: 113; Drummond 2013b: 413–415.

76

Todisco 2018b, forthcoming.

77

Wiseman 2016: 131.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:23 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 60

14/11/2017 14:56:16

VARRO, ATTICUS, AND ANNALES Author(s): R.M.A. MARSHALL Source: Bulletin of the Institute of Classical Studies , 2017, Vol. 60, No. 2, VARRONIAN MOMENTS (2017), pp. 61-75 Published by: Oxford University Press Stable URL: https://www.jstor.org/stable/10.2307/48554659 REFERENCES Linked references are available on JSTOR for this article: https://www.jstor.org/stable/10.2307/48554659?seq=1&cid=pdfreference#references_tab_contents You may need to log in to JSTOR to access the linked references. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Oxford University Press is collaborating with JSTOR to digitize, preserve and extend access to Bulletin of the Institute of Classical Studies

This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms

VARRO, ATTICUS, AND ANNALES R. M. A. MARSHALL Abstract: This paper explores the scholarly relationship between Varro and Atticus by focusing on the attribution of a Varronian fragment (Gell. 17.21.24). Taking into account Gellius’s citation practices, it is argued that the fragment (concerning the execution of the tyrant Manlius) was mediated via Atticus, not taken directly from Varro. The evidence for Varro’s and Atticus’s friendship and scholarly cooperation undermines traditional attempts to identify any ultimate written source for this material, and prompts a re-evaluation of the relationship between the two friends’ antiquarian interests to outline the consequences of scholarly and social exchange. This investigation suggests that, like many apparent points of Attico-Varronian overlap (including Rome’s foundation-date), the fragment’s origins might owe more to personal acquaintance than secluded book-learning. Their antiquarian choices had the potential to be politically charged, and the version of Manlius’s death endorsed by Varro may thus indicate his political stance towards Caesar. Keywords: Varro, Atticus, Annales, Cicero, Caesar, Manlius, Gellius, 753 bc, antiquarianism, scholarship

1. Introduction [...] and at Rome Marcus Manlius, who during the siege of the Capitol had repulsed the Gauls as they were climbing up its steep cliffs, was convicted of having formed the design of making himself king. Marcus Varro says that he was condemned to death and hurled from the Tarpeian Rock; but Cornelius Nepos has written that he was scourged to death. In the very same year, which was the seventh after the recovery of the city, it is recorded that the philosopher Aristotle was born.1 (Gell. 17.21.24–25) The note that Manlius was hurled from the Tarpeian Rock has been assigned, thanks to its subject matter, to the sixth book of Varro’s Res diuinae or twentieth book of the Res humanae.2 Other arguments, based on the context in which the note is found, ascribe it to Varro’s Annales I or III.3 A hypothetical case could also be made for other Varronian works, including De gente populi Romani and De uita populi Romani. However, we cannot guarantee — at least given the traditional constraints of source criticism — that any of these identifications is correct, because our ‘fragment’ need not depend on a statement written in any Varronian treatise. In the present paper, I will argue that Gellius may not have found Varro’s note on Manlius in his own reading of Varro, but borrowed this from Atticus’s Annalis. If so, one must then confront a methodological problem: Atticus, Cicero’s famed correspondent, knew Varro personally; why should we presume this report depends on a work read by Atticus, rather than an opinion offered to him, say, at dinner? If it proves impossible in principle to tie down the mention of Manlius 1

Trans. Rolfe 1948–54. Latin quoted p. 64, below.

Res diuinae: Merkel 1841: cxxiv (ad F6), cxxxiv; Brunetti 1874: 1329–30; Semi 1965: 2.153. Res humanae: Deschamps 1995: 16–20. 2

3

Ritschl 1877a: 449; HRF F2; HRR F2; FRHist 52 F2.

BICS-60-2  2017

61 © 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms

BICS 60-2 Nov14-2 FINAL-FINAL.indd 61

14/11/2017 14:56:16

BICS-60-2  2017

62

to any one Varronian work, future editions will have to find some way of signalling that Manlius’s execution may belong in several Varronian works simultaneously. Perhaps, in addition to ‘fragment’ and ‘testimonium’, it is time to introduce a third category of sententia to our editions of authors whose fragmentary corpora present similar challenges to the definite attribution of vaguely referenced material.4 Rather than thinking of Roman antiquarianism as a purely textual phenomenon, argued through books written by scholars working in splendid isolation, this paper will examine the ramifications of personal acquaintance, even collaboration, for our understanding and reconstruction of this scholarly tradition. To begin with, I will argue that Gellius depended not on Varro directly, but on Atticus. The opening debate over fragment ascription and Gellius’s source(s), as well as being valuable in its own right, serves as an entrée to the wider questions of scholarly collaboration and the political significance of antiquarian research I wish to explore, with reference to the different traditions about the execution of Manlius. Taking account of Varro’s and Atticus’s intellectual and social connections in the third section of this paper, my new arguments concerning source ascription may not only shed light on the collaborative, even competitive, nature of Varro’s and Atticus’s research, but also open up questions concerning the potential politicization of antiquarian lore about tyrants and tribunes in the late Republic. This possibility is examined in the concluding section of the paper, which places the disagreement concerning the mode of execution of Manlius in its politically charged context. 2. Gellius’s source(s) 5 The context in which our note is found, rather than its contents, led Ritschl to identify its source as Varro’s Annales. It appears in the midst of Gellius’s synchronic chapter of ‘the flowers of history’ (historiae flosculi), which relied upon excerpts from chronici, gathered from various places and reassembled hastily (cursim): Gell. 17.21.1. In a still-fundamental study, Leuze demonstrated that Gellius’s chapter was indeed assembled cursim. One set of dates are calculated according to the assumption that Rome was founded in the second year of the seventh Olympiad, i.e. 751–750 bc.6 The source for these dates must have been Nepos’s Chronicon, known to have employed this foundation date as a basis for calculation (Solin. 1.27 = FRHist 45 F3): besides our passage, Nepos is named elsewhere in Gellius’s chapter (17.21.3, 8), and mention of chronici in its introduction is a clear allusion to the title of Nepos’s work.7 The dates of a second group of individuals and events, however, were calculated from a different foundation date, namely Ol. 6.3: 754–753 bc. This is conventionally labelled the ‘Varronian’ foundation date, chiefly thanks to Cicero’s praise of Varro (Acad. 1.9; published 45 bc), and to Censorinus, DN 21.6.8 Yet Varro’s use of the date now bearing his name is 4

This may seem self-evident to those used to working with ancient philosophers and doxographical evidence.

The following section summarizes arguments made in Marshall 2014: 264–71. On Gellius’s use of Varro see also Todisco in this volume.

5

6

Leuze 1911. The date already in use by Polybius’s day: Dion. Hal. 1.74.3 (= Polyb. 6.11a.2).

For the material attributable to Nepos, see Leuze’s table (1911: 269). Fantham (2001) has recently vindicated his findings.

7

A date expressed auc at DN 17.11 is not certainly Varro’s own. On Censorinus as a source for Varro see also Piras, p. 13–14, 20, in this volume.

8

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 62

14/11/2017 14:56:16

R. M. A. MARSHALL: VARRO, ATTICUS, AND ANNALES

63

only directly attested once, and in a rather unexpected place: Res rusticae 1.2.9,9 published 38 or 37 bc.10 It is widely assumed that Varro employed this date earlier, but the question will be reopened below. Atticus is also known to have used 754–753 bc as Rome’s foundation date (Solin. 1.27 = FRHist 33 F2), and wrote a work called the Annalis, a chronicle of Roman magistrates, laws, and wars: Atticus was a great imitator of the customs of the men of old and a lover of the early times [...] he gave a full account of them in the work in which he set down the chronological order of the magistrates. For there is no law, no treaty of peace, no war, no illustrious deed of the Roman people, which is not mentioned in that work at its proper date, and — a most difficult task — he has so worked out the genealogies of the families, that from it we can learn the descendants of our famous men.11 Atticus’s work was in Cicero’s hands by September 47 bc, and exercised an immediate effect on the quality of his historical exempla, witnessed chiefly by the Brutus, written in spring 46 bc.12 The dates of his Roman material were synchronized with events and people from Greek history, likely drawn chiefly from Athens, given his long association with that city (cf. Cic. Brut. 27–28 [= FRHist 33 F4]) and known predilection for all things Athenian (note his cognomen!). Feeney’s observation that the Athenian exempla in Gellius’s synchronic chapter form ‘the tip of a large Athenocentric iceberg’ thus adds weight to my following arguments for Gellius’s dependence on Atticus.13 Faced with a choice between Varro and Atticus, Leuze avoided identifying either as Gellius’s source for material adhering to the 754–753 bc foundation date, preferring to speak of a work/works employing ‘Varronian’ chronology. To avoid confusion, it will be more convenient — and (as we will see) more accurate — to refer to this chronology as ‘Attico-Varronian’. Gell. 17.21.4–7, 12–13, 16–18, 20–25 (the material immediately surrounding our Manlius note), 28–36, 40–41, and 43 can be associated with this second ‘Attico-Varronian’ dating system. Building on Leuze’s work, D’Anna has demonstrated the definite presence of Atticus in parts of Gellius’s chapter,14 though his findings have met with resistance. Drummond follows Leuze in dismissing Atticus from contention as Gellius never names him.15 This prejudice is ultimately traceable to the nineteenth-century Quellenforscher, who worked from the premise that Gellius only used sources that he actually references: as an a priori assumption the theory is worthless, not to mention that the loss of Gellius’s eighth volume should

Assuming Varro used ‘Varronian’ chronology, C. Licinius’s tribunate is dated to 145 bc (corroborated by Cic. Amic. 96). The calculations of Censorinus at DN 21.6 also imply that Varro used the ‘Varronian’ foundation date, though the praise of Varro at DN 21.4–5 relates to the synchronization of this date with other calendars, and tells us nothing about who actually discovered it. There is no reason to relate this testimonium, as some have done, to the Res humanae.

9

10

Marshall 2016.

11

Nep. Att. 18.1–2, trans. Rolfe 1984.

12

Fantham 2001; Feeney 2007: 25–28.

13

Feeney 2007: 32–41.

14

D’Anna 1973–74; 1975.

15

Drummond 2013b: 1.420.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 63

14/11/2017 14:56:16

BICS-60-2  2017

64

automatically invalidate any such hypothesis!16 Sometimes Gellius is frank concerning his use of intermediaries (e.g. Gell. 20.11.4); elsewhere, significantly, he cites Varro through Verrius Flaccus without hinting that theft has been committed.17 There is a second answer to those who deny that Atticus could be Gellius’s source, namely that technical treatises such as Atticus’s chronology were liable to be reworked by later hands, added to, excerpted, or recast for other purposes. It is entirely possible that Atticus’s Annalis fell into the orbit of the schoolmen and epitomators, and reappeared shorn of his name and with any number of concretions and deletions.18 Gellius may have been unwittingly working with an Attican core. This hypothesis might explain, incidentally, why such a useful work fell into obscurity so quickly. Atticus’s Annalis is not securely attested after Asconius.19 The presence of Atticus in Gellius’s chapter is most clearly demonstrated through a comparison of Gellius and Cicero. As noted, Cicero relied upon Atticus as his chief source for the historical material in the Brutus (information found only in Gellius is underlined): annis deinde postea paulo pluribus quam uiginti, pace cum Poenis facta, consulibus C. Claudio Centhone, Appii Caeci filio, et M. Sempronio Tuditano, primus omnium L. Liuius poeta fabulas docere Romae coepit post Sophoclis et Euripidis mortem annis plus fere centum et sexaginta, post Menandri annis circiter quinquaginta duobus. [43] Claudium et Tuditanum consules secuntur Q. Valerius et C. Mamilius, quibus natum esse Q. Ennium poetam M. Varro in primo De Poetis libro scripsit eumque, cum septimum et sexagesimum annum haberet, duodecimum Annalem scripsisse, idque ipsum Ennium in eodem libro dicere. (Gell. 17.21.42–43)

atqui hic Liuius [qui] primus fabulam C. Claudio Caeci filio et M. Tuditano consulibus docuit anno ipso ante quam natus est Ennius, post Romam conditam autem quarto decumo et quingentesimo, ut hic ait, quem nos sequimur. est enim inter scriptores de numero annorum controuersia. Accius autem a Q. Maxumo quintum consule captum Tarento scripsit Liuium annis xxx post quam eum fabulam docuisse et Atticus scribit et nos in antiquis commentariis inuenimus. (Cic. Brut. 72)

A little more than twenty years later, when peace had been made with the Carthaginians and the consuls were C. Claudius Centho, son of Appius the Blind, and M. Sempronius Tuditanus, the poet Lucius Livius was the very first to put on plays at Rome, more than 160 years after the death of Sophocles and Euripides and about fifty-two years after the death of Menander. The consuls Claudius and Tuditanus were followed by Q. Valerius and C. Mamilius, in whose year the poet Ennius was born, as Marcus Varro has written in the first book of De poetis; and he adds that at the age of sixty-seven Ennius had written the twelfth book of Annals, and that Ennius says so in that same book.20

And yet this Livius produced his first play in the consulship of C. Claudius, son of Caecus, and M. Tuditanus, as late as the very year before the birth of Ennius, 514 years after the founding of Rome, according to the authority whom I follow; for there is a dispute among writers about the chronology. Accius however stated that Livius was taken captive from Tarentum by Q. Maximus in his fifth consulship, thirty years after Livius had produced his first play, as both Atticus writes and we ourselves find in the early records.21

Gellius’s account is more detailed than Cicero’s, implying independent use of Cicero’s own source, i.e. Atticus.22 16

Holford-Strevens 2005: 72.

17

Holford-Strevens 2015: 149–50.

18

Cf. the afterlife of Eusebius’s and Jerome’s chronica (Brincken 1957; Croke 2007: 577–80).

19

Byrne 1920: 49.

20

Trans. Rolfe 1948–54, adapted.

21

Trans. Hendrickson–Hubbell 1971, adapted.

22

D’Anna 1973–74: 183–86; 1975: 345–47. Similarity also noted by Fantham 2001: 355.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 64

14/11/2017 14:56:16

R. M. A. MARSHALL: VARRO, ATTICUS, AND ANNALES

65

Gellius’s reference to Varro’s De poetis (a work on literary history) can be explained in two ways. Neither explanation affects my overall argument, but one is of relevance to Varro’s and Atticus’s intellectual relationship. Either 1) Gellius added the consuls Valerius and Mamilius from his own reading in Varro’s treatise, or 2) Atticus cited the De poetis for this information (whence the reference was transmitted to Gellius). Another Ciceronian passage (see below) provides a t.a.q. for Varro’s poetical researches that allows this possibility. If Atticus cited Varro, he will have done so for the names of the consuls at Ennius’s birth, but independently calculated Ennius’s year of birth as a date from Rome’s foundation (this detail belongs in a chronicon, not a literary history). Of these explanations, the second is more likely:23 Gellius explicitly cites the De poetis twice more, but the quality of his references is uninspiring. One accompanies the supposed funerary epitaph of Plautus (Gell. 1.24.3–4), but the accompanying epitaphs for Naevius and Pacuvius suggest an anthology of similar spurious inscriptions.24 Later in his synchronic chapter, Gellius refers to the De poetis for Naevius’s service in the First Punic War (Gell. 17.24.45). Given the location, the mediation of Atticus is again a real possibility, as the following Ciceronian passage demonstrates. We have no direct evidence for the publication date of Varro’s De poetis, but the only time Varro is mentioned as a source in Cicero’s Brutus concerns the controversy surrounding the date of Naevius’s death: In the consulship of [Cethegus and Tuditanus, 204 bc], as early records show, Naevius died; though our friend Varro, with his thoroughness of investigation into early history, thinks this date erroneous and makes the life of Naevius somewhat longer. His reason is that Plautus, his contemporary, did not die until the consulship of Publius Claudius and Lucius Porcius, twenty years after the consuls named above, when Cato was censor [184 bc].25 This passage is important because it demonstrates that Varro had already conducted research into the lives of the poets by the time that Cicero composed the Brutus. In other words, the De poetis is likely to have been written before or concurrently with Atticus’s Annalis, and thus nothing prevents Atticus from using (or rejecting) Varro’s opinions on the poets.26 It is important too, because Varro’s later date is rejected by recourse to ‘early records’ (ueteres commentarii), the same mysterious records that were used by Atticus (according to Cicero) to disprove Accius in the passage quoted earlier (Cic. Brut. 72). The similar way in which Varro is rebuffed here suggests that Varro’s later date for Naevius’s death was noted and rejected by Atticus, and is not Cicero’s own observation. Besides noting the similarity between Gell. 17.21.42–43 and Cic. Brut. 72 (= Atticus), D’Anna highlighted similarities between Gell. 17.21.4–7, 28–29, and several other Ciceronian passages displaying the influence of Atticus, suggesting these were also derived

23

Dahlmann 1963: 29–34, 101–02.

On this material: Dahlmann 1963: 65–100; Lehmann 2002: 33–47, 91–106, 185–93. Even if the Naevius and Pacuvius epitaphs originally came from Varro’s De poetis, their citation by Gellius does not imply knowledge of the ultimate source. Suetonius’s De poetis is a potential intermediary (cf. Suet. Virg. 36 for Virgil’s ‘epitaph’), the conjectural source for the Republican poets at Gell. 19.8.10–14: Schmidt 2000: 28. 24

25

Cic. Brut. 60, trans. Hendrickson–Hubbell 1971.

26

Dahlmann 1963: 101–02.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 65

14/11/2017 14:56:16

BICS-60-2  2017

66

from his Annalis.27 D’Anna disregarded the remaining Gellian passages using ‘AtticoVarronian’ chronology (17.21.12–13, 16–17, 20–25) thanks to the lack of comparative passages in Cicero, though the failure to find further parallels can be wholly attributed to Cicero’s selective use of Atticus as a source for literary history (the material at Gell. 17.21.12–13, 16–17, 20–25 concerns wars and politics). We can finally approach the Varronian note itself. Gellius’s material conforms to the following structure: et M. Manlius Romae, qui Gallos in obsidione Capitolii obrepentes per ardua depulerat, conuictus est consilum de regno occupando inisse, damnatusque capitis e saxo Tarpeio, ut M. Varro ait, praeceps datus, ut Cornelius autem Nepos scriptum reliquit, uerberando necatus est; eoque ipso anno, qui erat post reciperatam urbem septimus, Aristotelem philosophum natum esse memoriae mandatum est. (Eng. trans. at the head of this chapter) Two notes are inserted between the synchronism of Manlius’s execution and birth of Aristotle, one from Varro, one from Nepos, neither of any chronological relevance. The synchronism with Aristotle’s birth makes it certain that Gellius was following his ‘AtticoVarronian’ chronology here:28 Aristotle was born in 384 bc, which provides a date for the Sack of Rome, counting inclusively after the ancient fashion, of 390 (Nepos presumably placed this event in 388/387: cf. Polyb. 1.6.2). Now, as D’Anna has already seen, it is unlikely that Gellius himself added the note from Nepos at this point:29 if this is his own addition, he has ignored the different dates Nepos gave to these events. As the structure of Gell. 17.21 demonstrates, Gellius’s excerpts from Nepos are not blended with those from Atticus, but are simply butted up against them. More significantly, however, Gellius otherwise avoids using Nepos as a source for the whole of the fourth century bc. D’Anna took the citation’s unusual form as a sign that Varro had mentioned Nepos, but did not follow the idea to its natural conclusion. Gellius’s intrusion of Varro’s name at this juncture is as much out of place as Nepos’s. D’Anna avoided linking Gell. 17.21.20–25 to Atticus in the absence of Ciceronian parallels, but if the whole passage is based on an excerpt taken directly from one of Varro’s works, why mention him at this particular point? The simplest solution is that the juxtaposed notes from Varro and Nepos are both second-hand references taken by Gellius from the source employing the ‘Attico-Varronian’ chronology, i.e. from Atticus. If we posit the mediation of Atticus, it is possible to reject several Varronian works when considering the note’s ultimate source. As already observed, Cicero had received a copy of Atticus’s Annalis by September 47 bc, which must exclude derivation from e.g. De gente populi Romani (published in or after 43 bc).30 We cannot ignore Varro’s Annales on this basis (its date is unknown), but the Antiquitates pose a conundrum. Given the t.a.q. of Atticus’s work and the generally

27

D’Anna 1973–74: 172–78, 188; 1975: 339–45.

28

Leuze 1911: 253.

29

D’Anna 1973–74: 169.

30

On the dating of DGPR see Todisco, p. 57, in this volume.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 66

14/11/2017 14:56:16

R. M. A. MARSHALL: VARRO, ATTICUS, AND ANNALES

67

accepted publication window attributed to Varro’s Antiquitates (47–46 bc),31 Atticus and Varro must have undertaken their chronological research simultaneously. Indeed, Varro’s Antiquitates may have only appeared after Atticus’s Annalis. This not only leaves the question of the ultimate source of the Manlius note in limbo, but also raises an equally vexing problem: who first published the so-called Varronian foundation date for Rome, Varro or Atticus? I would like to offer, however, a new approach to the question of the passage’s ultimate origin. Should we be tied to the notion that Gellius’s Varro fragment once depended, at whatever remove, on a written source, given the ample evidence for the close friendship between Varro and Atticus? 3. Varro and Atticus Varro, born in 117 or 116 bc,32 was several years older than Atticus.33 Although we cannot say how or when they first met, it was presumably at a relatively young age. Though Byrne conjectured Atticus may have shared the tutor Aelius Stilo with Cicero and Varro,34 the first intellectual acquaintance they certainly held in common was Antiochus of Ascalon, whom they heard lecture in Athens,35 though probably not concurrently. Atticus went East c. 86 bc and stayed in and around Athens until c. 65.36 Atticus’s departure roughly coincides with Varro’s thirtieth year, when he was presumably launching himself into Roman politics. If Cichorius’s inference of a quaestorship for Varro is correct (i.e., if Varro did not initially enter the senate thanks to his tribunate: Gell. 13.12.6),37 this will have been held around the time of Atticus’s departure. A passage from the Menippean satire Serranus, περὶ ἀρχαιρεσιῶν (‘Serranus, on the election of magistrates’),38 composed c. 81–67 bc,39 might have shed light on their early relationship, but the text is insecure: noster Atticus riualis, homo item lectus in curiam, cum macescebat40 Non. 137.3M = Sat. Men. F453B Our rival Atticus, a man likewise nominated to the senate, while he pined away.

Boissier (1861: 44–47) and Jocelyn (1982: 158–77) favour the 50s bc, downplaying any engagement with Caesar’s ‘religious programme’. Against this view (all favouring c. 48–46): Della Corte 1970: 123–34; Horsfall 1972; Lehmann 1986; Tarver 1996. See also Momigliano 2003.

31

32

Marshall 2016.

33

Probably born November 110 bc: Feger 1956: 503.

34

Byrne 1920: 23–24.

35

Varro: Blank 2012: 252–53. Atticus: Byrne 1920: 26.

36

Shackleton Bailey 1965: 3–4.

37

Cichorius 1922: 219–20.

38

See Cèbe 1972–99: 11.1829–46.

39

Dates supplied by internal references in other satires: Cichorius 1922: 207–13; Cèbe 1972–99: 1.xv–xvii.

accius F1 attius F3 amicus Laetus antiquus Gulielmius Attilius Vollbehr Axius Müller | ruralis Popma hilaris Riese | cum mac.] commacescebat Iunius macore macescebat Oehler cura macescebat Riese commacescebat macore Müller cum macore mac(r)escebat Della Corte.

40

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 67

14/11/2017 14:56:16

BICS-60-2  2017

68

Famously, Atticus refused to enter politics and was never enrolled in the senate.41 However, nothing prevents Varro from describing some fictive scenario: macescebat then forms the imagined response of the retiring Atticus when compelled to take an active part in politics.42 But in what sense was Atticus Varro’s rival c. 80–60 bc? Neither had made a start on their scholarly publications; one might hazard numerous conjectures, but it is unclear, in any case, whether Varro is talking here in propria persona. All one can conclude from this Menippean satire is that, if our Atticus is meant, then Varro (and his audience) knew enough of his unwillingness to enter politics to make this piece of fantasy effective. If the Res rusticae avoids anachronism, Varro and Atticus met in Epirus in 67 bc. Atticus bought property near Buthrotum in 68,43 and Varro claims to have visited the area when serving as Pompey’s legate in the Pirate War of 67, making this the occasion for the dialogue of book 2 (Atticus is given a speaking part).44 The first definite date for a connection is 54 bc. In July Atticus pressed Cicero to make Varro an interlocutor in De republica (Cic. Att. 4.16.2), though without result. Cicero refused, pleading historical verisimilitude, but promised to include an honourable mention of Varro in a preface. On what grounds is unclear, though Cicero had written to Atticus in May requesting that his household staff allow him to browse their master’s books in his absence, ‘those of Varro among the rest’ (Cic. Att. 4.14.1). The particular books at issue are unknown (they may, in fact, be unconnected with the writing of De republica),45 but a more fundamental point might emerge from this exchange. Dix has suggested that Cicero’s peculiar request, with its implicit suggestion that Varro’s books are somehow separate from the rest of Atticus’s library, may be indicative of their special status.46 Atticus might have had books to hand by Varro that had been sent for comment before public circulation. We know that Cicero made similar confidential use of Atticus, and will shortly see further evidence corroborating Dix’s suggestion. If Varro could not be found a part in De republica, Atticus’s persistence on his friend’s behalf did finally succeed: Cicero’s treatise on Academic Scepticism was rewritten to make Varro the major interlocutor. The work was rededicated to Varro in July 45 bc (Cic. Att. 13.25). Large portions of old and new versions still survive, as does the exchange of letters in which Atticus pleaded his friend’s case.47 Cicero was also doubtless influenced by the knowledge that Varro was writing a work to be dedicated to him, though complained of already waiting for several years (Cic. Att. 13.12.3). Cicero soon afterwards received Varro’s De lingua Latina, likewise originally intended for someone else (cf. Ling. 7.109).48 While often assumed to be identical with the work Cicero criticized for its belatedness, this is hardly certain. Perhaps Atticus was secretly behind this timely rededication as well, attempting to bring two friends together through reciprocal literary gifts. We cannot say, Thus the emendations (though the problem disappears if another Atticus is meant). Attius / Accius, found in minor branches of the tradition, cannot be defended. On Atticus’s avoidance of public life: Perlwitz 1992: 86–146. On alleged Epicureanism: Castner 1988: 57–61 (now of equivocal significance: Fish 2011).

41

42

Cf. Sat. Men. (Serranus) F452B: ‘to have been called, from a serene life, to the dregs of your Curia’.

43

Cic. Att. 1.5.7.

44

Varro, Rust. 2. pref. 6.

45

Horsfall 1972: 120.

46

Dix 2013: 221 n. 57.

47

Hunt 1998: 10–13.

48

Barwick 1957.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 68

14/11/2017 14:56:16

R. M. A. MARSHALL: VARRO, ATTICUS, AND ANNALES

69

but from the tenor of Cicero’s correspondence with both Atticus and Varro, it is clear that Atticus enjoyed the better relationship with Cicero’s δεινὸς ἀνήρ (homme terrible: Cic. Att. 13.25.3).49 A handful of stilted letters from Cicero to Varro survive,50 sent around the time of Caesar’s triumphal return to Rome in 46 bc — both Cicero and Varro fought for Pompey — but the series tellingly begins with a snapshot of Atticus acting as intermediary, reading Cicero extracts from his Varronian correspondence (Cic. Fam. 9.1). Book dedications from Varro to Atticus follow in later years, including De uita populi Romani (published c. 43 bc).51 The Atticus: de numeris (from the Logistorici) was named in his honour (Censorinus, DN 2.2). It is unclear whether this piece appeared before Atticus’s death in March 32 bc,52 but Varro certainly honoured the living Atticus elsewhere: he is given a speaking part in Res rusticae II (published in 37 or 36 bc), winning the epithet ‘Faustulus’ as an expert in rearing dogs and sheep. Work remains to be done on Varro’s and Atticus’s responses to Caesar and the ideological implications of their chronological research, but even a quick glance over their respective outputs demonstrates the breadth of their mutually informed interests. Atticus authored several works: Annalis; Imagines; three genealogical monographs; De consulatu Ciceronis (in Greek).53 With the exception of the memoir on Cicero’s consulate, all find their counterparts in Varro’s oeuvre. Besides the catalogue of notables that must have constituted the major part of Atticus’s Annalis (Nep. Att. 18.1–2, quoted above), Atticus produced a portrait-book of famous Romans, captioned with biographical sketches in verse.54 Varro’s Imagines (or Hebdomades) not only employed the same title, but followed exactly the same format, though adding Greeks alongside Romans (F106–24 Salvadore).55 The parallel inclusion of Greeks suggests that Varro’s work was undertaken on an altogether grander scale: his finished work filled fifteen volumes, and took several years to complete. There is a strong possibility that Atticus’s collection was created concurrently with Varro’s. Though Pliny thinks that Varro invented the format, he also notes Atticus as a witness to the contemporary passion for portraiture (NH 35.11). Atticus’s collection presumably relied on the fundamental research undertaken for the Annalis, and should thus be dated c. 47 bc or later.56 In late 44 bc, Cicero wrote to Atticus assuring him that ‘I don’t take it badly you approve of Varro’s Πεπλογραφία’ (Cic. Att. 16.11.3). This is a reference to Varro’s Imagines,57 though why Cicero was untroubled by Atticus’s good opinion is obscure, unless he jokingly anticipated a jealous outburst from Varro’s rival portraitist. The letter is revealing because 49

See generally Rösch-Binde 1998.

50

See Leach 1999; Wiseman 2009: 107–29.

Internal references supply a t.p.q. of 49 bc: Risposati 1939: 85. However, composition likely overlapped with De gente populi Romani (published c. 43 bc). On this relationship: Ax 2000: 357–67.

51

52 Nep. Att. 22.3. As the logistorici are apparently missing from Cicero’s résumé of Varro’s philosophical work to c. 45 bc (Acad. 1.8–9 with Morgan 1974: 117–22), and as most dealt with philosophical subjects, the majority (or whole) of the collection was presumably composed later. See also Horsfall 1972: 122–24. 53

For overviews: Byrne 1920: 23–51; Buckley 2002; Drummond 2013a: 1.344–53.

54

Nep. Att. 18.5–6 with: Byrne 1920: 36; Drummond 2013a: 1.346. No fragments survive.

55

Ritschl 1877b: 508–92; Dahlmann 1935: 1227–29; Cardauns 2001: 79–80.

56

Prokoph 2015 has argued for the priority of Atticus’s Imagines on this basis.

Πεπλογραφία, ‘tapestry of worthies’: Shackleton Bailey (1967: 301). Perhaps Cicero’s coinage? Cf. Jones 1939, offering the unconvincing gloss πεπλογραφεῖν Varronem.

57

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 69

14/11/2017 14:56:17

BICS-60-2  2017

70

it antedates by several years the collection’s final publication. In a long fragment from the work’s introduction, Varro recorded that its completion coincided with his seventy-eighth year (Gell. 3.10.17), implying final publication in 39 or 38 bc. Not only may composition of Varro’s and Atticus’s Imagines have been contemporaneous, but Atticus’s early access to Varro’s incomplete collection may add weight to his conjectured role as Varro’s literary confidant. Atticus also wrote monographs on the historical genealogies of the Iunii, Marcelli, Aemilii, and Fabii at the request of representatives of their lines Marcus Brutus, Claudius Marcellus, Cornelius Scipio, and Fabius Maximus (Nep. Att. 18.3–4). These works are deeply obscure. No fragments survive, though as they presumably grew out of Atticus’s research for the Annalis, they can be approximately dated by this work’s appearance in 47 bc. The identities of Claudius Marcellus and Cornelius Scipio are uncertain.58 The latter cannot be identified with Q. Caecilius Metellus Scipio without radical emendation. Cornelius Scipio Salvitto has been proposed,59 a rising star in Caesar’s administration, but ordinarily, one presumes, not the sort of man with whom Atticus associated. Q. Fabius Maximus was Caesar’s man too, and it is easy to imagine their joint genealogy may have served some propagandistic purpose for Caesar’s faction. Less clear is why Atticus should have agreed to write such a work, but Ann Marshall has suggested it may be connected with the bargaining over Buthrotum in Epirus, where Atticus owned estates.60 Caesar threatened to found a colony there, and could have extracted a genealogical endorsement of his lieutenants in return for abandoning the project.61 One might compare Atticus’s genealogical work with Varro’s treatise De familiis Troianis, on the mythical Trojan ancestors of the great Roman families.62 This work cannot be securely dated, but seems to respond to the new emphasis on Rome’s Trojan origins that arose under Caesar’s dictatorship. The surviving fragment, concerning the gens responsible for the care of the Palladium, is suggestive. Diomedes, believing he would never have peace unless he returned the Palladium to the Trojans, tried to give this to Aeneas, but handed it instead to Nautes, ‘whence the sacra of Minerva were the preserve of the Nautines, not the Iulii’ (Serv. Aen. 2.166 = FRHist 52 F3a). A Caesarian coin of 47–46 bc depicts Aeneas himself (legendary Julian ancestor) carrying the Palladium from Troy.63 The version told in De familiis Troianis may be an attempt to correct Caesarian propaganda, and if so, presumably post-dates Caesar’s assassination. If Varro’s De familiis Troianis was not a direct challenge to Atticus’s (Caesarian?) biographies — Varro’s title implies that he took a broader overview of the Roman elite’s ancestry — it is surely a response on some level. Perhaps, given the breadth of coverage implied by Varro’s title and the mention of the Nautii (if representatives of this gens survived into the first century bc, they were now deeply obscure),64 the work sought to redefine the Trojan connection as a common Roman, 58

Drummond 2013a: 1.350–51.

59

Billows 1982: 61; Marshall 1993: 313–15.

60

Marshall 1993: 313–15.

61

Atticus’s and Cicero’s lobbying halted the project: Cic. Att. 16.16a.

See: Dahlmann 1935: 1241–42; Bäumerich 1964: 41–62; Drummond 2013b: 1.421–22; 2.840. For the contemporary relevance of antiquarian research see Todisco, p. 58, in this volume. 62

63

Drummond 2013b: 3.515 (RRC 458; Assenmaker 2007: 394–405).

The family last held the consulship in 287 bc (C. Nautius Rutilus = RE 6), though a senatorial N(a)utius of 129/101 bc (Sherk 1969: no. 12) may be a relation. 64

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 70

14/11/2017 14:56:17

R. M. A. MARSHALL: VARRO, ATTICUS, AND ANNALES

71

rather than Caesarian, inheritance. Putting aside the conjectural Caesarian complexion of Atticus’s biographies, the distinction between his necessarily partisan family commissions and Varro’s more generous coverage is still obvious. The question is worth investigating further. The subject of what divine images were carried to Italy by Aeneas, incidentally, supplies the only instance from antiquity of a direct contrast between the views of the two experts: Varro in the second book of his Histories [i.e. Res humanae] says [...] Aeneas his father on his neck, and the Achaeans, on this devotion, granted him the opportunity of returning to Troy; and Aeneas their household gods [...] Atticus agrees about Aeneas’s father but not about the household gods, saying they were brought to Italy from Samothrace.65 Atticus thought the Penates had been acquired by Aeneas from Samothrace, apparently agreeing with the second-century historian Cassius Hemina.66 Varro, on the contrary, believed the Penates were first introduced to Troy from Samothrace by Dardanus, father of Troy’s founder (Macr. Sat. 3.4.7). The account favoured by Varro, rather than attributing agency to Aeneas, thus reduces him to the status of a custodian. One might compare the programmatic stress laid on such acts in Varro’s Res diuinae (F2a Cardauns), dedicated to Aeneas’s descendant Caesar. It would be wrong to lay too much stress on the ideological implications of this disagreement: Atticus was, after all, following the version endorsed by Cassius Hemina a century before Caesar’s rise to fame. It is potentially significant, however, that Varro privileged an alternative account, one with a subtly different emphasis. Sadly, the wording and structure of the above Virgil scholium makes it impossible to say definitively whether Atticus quoted Varro’s Res humanae, Varro quoted Atticus, or if the contrast was drawn by some later scholar. While Atticus’s Annalis (c. 47 bc) is the most obvious source for his contribution to the Penates debate, we simply do not know whether Varro’s Res humanae preceded or followed this work. Lastly, there is Atticus’s Annalis to consider. As noted earlier, as well as giving magisterial fasti and genealogical information, this work synchronized Roman with Greek history, using Olympiad 6.3 (754–753 bc) as its Roman year one. Given the evidence for Atticus’s interest in the origin of the Penates, however, his chronology may not have begun with Rome’s foundation, but with a preliminary survey of Roman prehistory. The publication date of Varro’s own Annales is a mystery. It is not certainly mentioned in Cicero’s résumé of Varro’s intellectual works (Acad. 1.9), and Atticus, rather than Varro, clearly exercised the major influence on the reformation of Cicero’s chronological ideas: it is thus likely to have appeared later.67 It may be suggestive that, if one accepts a minor emendation to the reference supplied with the unique fragment of this work, Varro did not treat the regal period until the penultimate book, suggesting a definite bias in favour of prehistory.68 This 65

Schol. Veron. Aen. 2.717 = FRHist 33 F1; trans. Drummond 2013a.

66

Cf. FRHist 6 F6–7 with Briscoe 2013: 3.162–64; Drummond 2013a: 3.457.

67

Cf. Drummond 2013b: 1.419.

Charisius 133.25B = FRHist 52 F2. Charisius’s reference erroneously implies that Varro’s four-volume treatise (Jer. Ep. 33) filled a single volume (idem in Annali). Despite Drummond (2013b: 3.513), van Putschen’s restoration — idem III Annali — deserves consideration: Charisius’s practice of providing book numerals after titles is not rigidly observed (cf. Charisius 102.14?, 145.22, 152.4, 158.14, 170.9, 171.2, 225.4, 246.4, 273.14B). Cf. the annalists Cassius and Cn. Gellius, who also devoted their opening volume(s) to pre-Roman history (Rawson 1991:

68

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 71

14/11/2017 14:56:17

BICS-60-2  2017

72

bias is also present in a work that can be definitely dated. Varro’s De gente populi Romani, published in 43 bc and containing the only evidence for a sustained Varronian synchronic chronology,69 did not reach Rome’s foundation until the fourth and final book.70 The latter work, thanks chiefly to our knowledge of its publication date, looks like a conscious attempt to write the prequel of Atticus’s earlier Annalis. Yet the intertwining of Varro’s and Atticus’s mutually informed chronological interests clearly antedates this work. Varro certainly worked on the age of the city before 45 bc (Cic. Acad. 1.9: ‘you have laid bare [...] the age of our state’), presumably detailing his findings somewhere in Res humanae XIV–XIX (‘On Time’: August. De civ. D. 6.3), though perhaps elsewhere too; as noted above, this work is generally dated to 48–46 bc. What this work does not seem to have contained, however, is any proper fasti: evidence for detailed listings of magistracies or a comprehensive chronology for Rome is entirely lacking.71 If Atticus’s Annalis did appear after Varro’s work, it cannot have been a crude epitome of his friend’s magnum opus, but required extensive and original research,72 a conclusion reinforced by Cicero’s admiring reaction in his Brutus. If Varro and Atticus were simultaneously engaged on chronological problems, resulting in highly complex and dissimilar works that appeared so soon together and that employed the same novel foundation date for Rome, we are surely justified in imagining that some scholarly collaboration has taken place, and in abandoning the label ‘Varronian’ in favour of ‘Attico-Varronian’. Whether or not Varro put in the legwork only for Atticus to publish first (or vice versa),73 our thoughts should turn away from dependence upon ‘hard’ sources and towards the ramifications of friendship, the exchange of learned letters, and conversations over the dinner table.74 The form of Atticus’s reference to Varro at Gell. 17.21.24 certainly implies that his readers could find Manlius’s fall recounted in one of Varro’s published works.75 But it does not guarantee that one of Varro’s works was its source. Atticus may well have included the reference in his Annalis having heard from Varro personally that he had written on the subject; or took it for granted, thanks to detailed discussions or exchanges of notes, that the question was being treated in the Antiquitates or some other work(s) in progress. 4. Manlius Thus far, two trends emerge from our comparison of the two men’s works. Whatever Atticus did, Varro had to do it bigger, and whether or not one subscribes to the Caesarian slant of Atticus’s genealogical research, Varro seems to have politely declined to accept the polish applied to the early Iulii by their most illustrious living representative. 245–71). Fraccaro 1907: 82–111; Dahlmann 1935: 1237–41. Calculations brought down to the consulship of Hirtius and Pansa (43 bc): Arn. Adv. nat. 5.8 (= HRR F9). 69

The first Olympiad or Rome’s foundation opened the final volume: Dahlmann 1935: 1239–40. On this work see Piras, p. 15–16, and Todisco, p. 57–58 in this volume. 70

71

Cf. Mirsch 1882: 36–45.

72

Cf. Drummond 2013a: 3.458.

Note Drummond 2013a: 3.458: ‘the new foundation date may be the result rather than the cause of Atticus’s innovations in the magisterial list’. 73

74

See MacRae, pp. 39, 46 in this volume for oral transmission in the field of sacred law.

75

As Prof. Tim Cornell reminds me.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 72

14/11/2017 14:56:17

R. M. A. MARSHALL: VARRO, ATTICUS, AND ANNALES

73

If we accept that the note on Manlius is ultimately derived from Atticus, and may well result from his and Varro’s friendly exchanges, the issue of how Manlius met his death starts to become more interesting: the traditions surrounding the execution of would-be kings were in flux during the late Republic, continually readapted to suit the waves of postGracchan political violence. Wiseman has done most to disentangle the various stories concerning the trial and downfall of Manlius as reported in Livy and others,76 who intermingle branches of the tradition into increasingly baroque constructs,77 but there is little agreement regarding their relative chronology, and various elements remain puzzling. Manlius is famed as the hero who saved the Capitol from the Gauls, grew too ambitious, stirred up the plebs, aimed at kingship, and met a deserved but tragic end. An early version of the story must have put Manlius on trial before the comitia centuriata, meeting under the duumuiri perduellionis outside the city. The Twelve Tables ordained that citizens could not be condemned to death except before such an assembly.78 The punishment for treason (perduellio) was to be nailed to an arbor infelix and scourged to death.79 Nepos’s statement in Gell. 17.21.24 that Manlius was beaten to death is, in fact, the earliest attestation of this tradition, which probably dates back to the second century bc or earlier. This version, Wiseman argues, was subsequently coloured by attempts to portray Manlius as a ‘protopopularis’. Some such trend is clearly visible in Livy’s account, which describes Manlius and his actions in terms strongly reminiscent of Sallust’s Catiline.80 In another version, however, Manlius was tried by the tribunes and executed in their signature manner. Without lictors or fasces, tribunes had to content themselves with throwing prisoners off the Tarpeian Rock.81 Pleasingly for the more sensationalist historians, the site of Manlius’s greatest triumph could now be depicted as the scene of his literal downfall.82 Varro clearly endorsed major parts of this version. The problem of the trial’s illegality was presumably excused as an instance of a mob convening a kangaroo court.83 A plebeian trial implicitly requires Manlius to have been punished as an enemy of the plebs, but the ‘proto-popularis’ theme is laid on so thickly that our surviving sources preserve no trace of earlier vilification, paradoxically depicting Manlius as the plebs’ friend and patron (the first patrician, in fact, to assume this role).84 Needless to say, both versions are anachronistic: the Livian account depicts the middleor late-Republican comitia centuriata, while a formal trial presided over by tribunes ignores the precarious status and limited agency of these officials in the early Republic, not to mention the procedure’s illegality.85 The duumuiri, on the other hand, appear in the surviving records Livy, 6.11.1–20.16. Cf. Val. Max. 6.3.1a; Gell. 17.2.14 (= Claudius Quadrigarius); Plut. Quaest. Rom. 285F, Cam. 36; App. Ital. 9; [Aur. Vict.] De vir. ill. 24; Amm. Marc. 21.16.13; Serv. Aen. 8.652. 76

77

Wiseman 1987. Further bibliography: Oakley 1999: 493.

78

XII tab. 9.2; Crawford 1996: 2.696–701.

79

Oldfather 1908.

80

Oakley 1999: 482–84. On elements of Manlius’s story embarrassing to Cicero, see Smith 2006a: 55.

81

Cadoux 2008: 215–17.

82

Wiseman 1987: 242.

83

Oakley 1999: 489, though no examples of early lynch justice are recorded.

84

Cf. Livy, 6.18.14.

85

Oakley 1999: 488–89. On the tribunate: Drummond 1989: 212–25.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 73

14/11/2017 14:56:17

BICS-60-2  2017

74

of precisely three pre-Augustan trials. Besides that of Manlius, they apparently participated in the parricide case of Horatius under the kings, and were notoriously resurrected from the history books for Rabirius’s trial in 63 bc.86 Though Oakley argues for an inversion of Wiseman’s sequence,87 the question of which version came first is largely unimportant here. Whenever the tradition began to branch out, subsequent retellings always had the potential to be politically charged. In one version, the one-time hero Manlius conceived tyrannical designs, was tried by the whole citizen body, and executed for treason. A simple warning against hubris. In the alternative version, the tribunes took charge: the representatives of the plebs destroyed the people’s enemy (with later ‘proto-popularis’ colouring on the people’s implicitly false friend). The only date we have for Nepos’s Chronicon is provided by a fleeting reference in Catullus 1.5–7 (composed c. 54 bc).88 Nepos’s endorsement of the duumviral tradition may not be unconnected with the recent revival of the office by Caesar et al. in 63 bc, which must have been preceded by a search for precedents, doubtlessly precipitating interest in (and invention of ?) historical parallels. As reported by Atticus in 47 bc, however, Varro endorsed a story in which a sometime vanquisher of Gauls, apparently setting his sights on tyranny, was overthrown by the tribunes. As Livy demonstrates, this version was not felt to be incompatible with a Catilinarian veneer, however odd the results; Varro’s Manlius may (or may not) have been similarly conceived of as the first patrician to befriend the plebs and purchase support with promised debt relief. These details, however, could only have added depth to a parallel, so obvious to us, which may also have been drawn by contemporaries: Caesar. Reasons why Varro shunned the duumviral tradition of Manlius’s execution are not hard to find. In 47 bc, following the Pompeians’ decisive defeat, the possibility of contriving a formal trial for Caesar was unthinkable; besides which, the last duumviral trial (actually involving a young Caesar) had proved farcical.89 Providing an exemplum for the tribunes, however, in which previous bearers of their illustrious office took the lead in ending a patrician’s tyranny — one who likewise based his reputation on fighting Gauls — might prick some consciences. It is thus worth noting that Caesar did indeed meet with sustained tribunicial obstruction: in 49 bc, the only opposition Caesar faced on entering Rome was from the tribune Caecilius Metellus, who tried to prevent the opening of the treasury;90 in 47 bc, tribune Trebellius opposed his Caesarian colleague Dolabella’s maverick legislation for debt remission so stoutly that Antony disavowed him;91 in 45 bc, Pontius Aquila refused to rise for Caesar’s triumph, supposedly eliciting the response: ‘Come then, take the Republic from me, tribune Aquila!’ (Suet. Iul. 78.2); finally, in 44 bc, the tribunes Caesetius Flavus and Epidius Marullus removed a diadem from Caesar’s statue, and tried to prosecute those who hailed him as king.92 Bauman 1969; Cloud 1977. Magdelain (1973) dismisses the office as annalistic fabrication, though one that must have antedated Caesar by a considerable margin. 86

87

Oakley 1999: 486–92, prompted by Magdelain’s scepticism (1973).

88

Briscoe–Drummond 2013: 1.399.

89

Goldsworthy 2006: 121–24; Tyrrell 1978.

Cic. Att. 10.4.8; Att. 10.8.6; Caes. BCiv. 1.33.3; Lucan, 3.114–68; Plut. Pomp. 62; Caes. 35.3–4; App. B Civ. 2.41; Dio Cass. 41.17.2.

90

91

Livy, Epit. 113; Plut. Ant. 8–9; Dio Cass. 42.29–33; cf. Plut. Caes. 51.1.

Livy, Epit. 116; Vell. Pat. 2.68.4–5; Val. Max. 5.7.2; Suet. Iul. 79.1; Plut. Caes. 61; Ant. 12; App. B Civ. 2.108; Dio Cass. 44.9–10. On Caesar’s tribunicial opposition: Yavetz 1974: 61–63.

92

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 74

14/11/2017 14:56:18

R. M. A. MARSHALL: VARRO, ATTICUS, AND ANNALES

75

Fantham has painted a Varro far more supine in defeat,93 but this picture neglects the carping author of Saturae Menippeae and Τρικάρανος: ‘der letzte Hauch des scheidenden guten Geistes der alten Bürgerzeit’, as Mommsen styled the former collection.94 Varro’s eventual collusion with Caesar’s regime (he became state librarian: Suet. Iul. 44.5) need not have been wholehearted, nor even accepted gracefully. Cicero wrote to Varro following Dyrrhachium on the basis that he was a fellow malcontent (Cic. Fam. 9.1–6). Varro did, after all, fight for his friend Pompey until outmanoeuvred by Caesar, and in youth was himself a tribune. Note, too, the reasons given by Appian for Varro’s proscription by Antony in 43 bc: ‘Varro was a philosopher and historian, a distinguished soldier and praetor, and probably for these reasons was proscribed as hostile to the monarchy’ (App. B Civ. 4.202).95 Even if Varro did prefer the tribunician version of the story because it suited his politics, why should it appear in Atticus’s Annalis? Atticus had remained aloof from the struggle between Pompeians and Caesarians. Apparently unconvinced by either account, he noted both versions but refused to endorse either. Could he have been put off by Varro’s partisanship; were Manlius and Caesar directly equated by him in some heated private exchange? Whatever the case, Varro’s preferred version needed to be mentioned somewhere in Atticus’s Annalis, because if tribunes had presided over the trial of Manlius, Atticus’s roll call of magistrates would be expected to name them.96 To Gellius, however, basking in the sunshine of the high empire, all these controversies, which must have seemed so vital to Varro and Atticus, were simply ‘flowers of history’. University of Glasgow

93

Fantham 2003: 109–17.

Mommsen 1856: 562. Reference from Wiseman (2009: 131–51), who nicely draws out Varro’s nostalgia for noblesse oblige and constitutional decorum. 94

Taylor 1934 argues that Varro’s De gente populi Romani intentionally promoted Caesar’s deification. Reconciliation with Octavian in 43 bc, however, tells us nothing about Varro’s views on Caesar’s dictatorship in 47. He may have felt the need to atone for past indiscretions and win powerful new friends, especially following his treatment by Antony (App. B Civ. 4.202–03; Cic. Phil. 2.103–05; Gell. 3.10.17). On Varro’s relations with Caesar see also sections 5 and 6 of Todisco in this volume. 95

M. Menenius and Q. Publilius: Livy, 6.19.5–20.1. Forsythe (1999: 84) discusses the significance of these names, clearly later inventions. 96

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:51:55 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 75

14/11/2017 14:56:18

ST AUGUSTINE AND THE DISAPPEARANCE OF VARRO1 DANIEL HADAS Abstract: This paper argues that St Augustine’s presentation of Varro’s Antiquitates rerum divinarum in City of God was pivotal to the latter text’s disappearance. It shows how Augustine used the Antiquitates’ tripartite theology (poetic, civil, and natural) to destroy Varro’s authority on traditional Roman religion. In Augustine’s reading, Varro’s open criticism of the gods of myth and poetry implied an equal rejection of the civil cult. This left the natural gods of the philosophers, but Augustine derided Varro’s attempts at philosophical theology. The result was that, to readers of the City of God, the Antiquitates rerum divinarum appeared as a failure: Varro had been incapable of justifying traditional Roman religion, while lacking the courage to attack it openly. Readers could then turn to the City of God itself as a better guide to the Roman gods, and there was no further need to read or copy the Antiquitates. Keywords: Augustine, Varro, civil, apologetics, theology, transmission, Christianization, conversion

Students of the classical sources used by St Augustine in his City of God are confronted with a puzzle: some relatively obscure texts, which he cites only in passing, have survived, whereas two major works, which play fundamental roles in parts of Augustine’s argument, have been lost from direct manuscript transmission: Cicero’s De republica and Varro’s Antiquitates rerum divinarum.2 In the centuries when the preserving and copying of books in the Latin West became the exclusive province of the Church, no author was more read and more revered than Augustine. The City of God, in particular, was copied assiduously, despite its great length.3 It is then reasonable to expect that, among classical works, scribes would have had some inclination to preserve those that engaged Augustine’s attention in his great book. This surely is some part of the explanation of why Apuleius or the Hermetic Asclepius were preserved.4 Yet the Antiquitates and the De republica disappeared. The latter, of course, subsequently returned, albeit mutilated, from the dead. But I will limit myself here to Varro. Augustine uses the Antiquitates rerum divinarum as the base text for his attack on Roman paganism in books 6 and 7 of the City of God: these books are a constant, often ad hominem, dialogue with Varro. Yet, I shall argue here that, for all the respect and attention he gave to Varro, Augustine’s handling of the Antiquitates rerum divinarum and of their author was instrumental in condemning the work to oblivion. 1 Ad maiorem Dei gloriam. Thanks to: the editors, the anonymous readers, Duncan MacRae, Carlotta Dionisotti (especially for the opening point on Apuleius). 2 As explained at De civ. D. 6.3, Varro wrote forty-one books of Antiquitates, twenty-five on res humanae and sixteen on res divinae. The whole work is lost (except for fragments), but I am here concerned, as was Augustine, only with res divinae. I also will not discuss Varro’s De cultu deorum, also used in De civ. D.: its fate was the same as that of the Antiquitates, and doubtless for the same reasons (for the fragments, see Cardauns 1960).

Nine manuscripts containing all or parts of De civ. D. survive from before 800, an exceptionally large number. See Lowe 1934–71, starting with the index entry in ‘Supplement’, 73. For an additional fifth- or sixth-century fragment, see Bischoff and Brown 1985: 344, 353 (for marginalia from the same period).

3

4

See O’Daly 1999: 253. Augustine uses Asclep., Met., De mun., Apol., De deo Soc. (extensively). All survive.

BICS-60-2  2017

76

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 76

14/11/2017 14:56:18

DANIEL HADAS: ST AUGUSTINE AND THE DISAPPEARANCE OF VARRO

77

It would of course be foolhardy to assert that it was entirely impossible for any given work of Latin literature, including any one of the seventy plus works of Varro, to survive. Still, for the Antiquitates rerum divinarum, we may put the question more precisely: the library of Hippo contained at least part of this book, and it also contained the hundreds of works Augustine wrote himself. In Hippo, the library did not survive the Vandals, but the bulk of its contents must have been brought across the Mediterranean.5 The many thousands of pages of Augustine’s own works then spread throughout Europe, with astonishingly few losses.6 Varro’s Antiquitates, whether they crossed the sea or not, soon passed into silence. Those who preserved Augustine could have preserved what he had of Varro, and we can legitimately ask why they did not. As will be seen, I believe the principal answer to this question to be that the City of God placed the Antiquitates rerum divinarum within a Christian framework where access to the original was no longer needed. But, before this, we may note the importance of Augustine’s damning verdict on Varro’s style. The need for models for the writing of formal and artistic Latin is among the factors that guaranteed the survival of classical texts. That compromised Varro, whose prose no one could call elegant. In an age less fond of fine writing, we may of course admire him as unfussy and terse, but late antiquity was no such age, and it is doubtful that any literary merit was found in Varro’s antiquarian compilations. This, indeed, is Augustine’s very specific judgment when he places Varro centre stage in the City of God: tametsi minus est suavis eloquio, doctrina tamen atque sententiis ita refertus est, ut in omni eruditione, quam nos saecularem, illi autem liberalem vocant, studiosum rerum tantum iste doceat, quantum studiosum verborum Cicero delectat. denique et ipse huic tale testimonium perhibet, ut in libris Academicis dicat eam, quae ibi versatur, disputationem se habuisse cum Marco Varrone, ‘homine’, inquit, ‘omnium facile acutissimo et sine ulla dubitatione doctissimo’. non ait ‘eloquentissimo’ vel ‘facundissimo’, quoniam re vera in hac facultate multum impar est. (6.2)7 Even though he is less sweet in his speech, still he is so stuffed with learning and ideas, that — when it comes to all that erudition which we call secular, but they call liberal — he instructs the student of facts just as much as Cicero delights the student of words. Indeed, that man himself also bears strong witness in his favour, by saying, in his Academic books, that he held the disputation they contain with Marcus Varro, ‘of all men’, he says, ‘easily the most acute, and without any doubt the most learned’. He does not say ‘the most eloquent’ or ‘the most well-spoken’, because truly in that skill he is greatly inferior. This remark on Varro’s failures in eloquence is an aside, irrelevant to the use Augustine makes of him in the City of God. It seems rather to show an instinctive urge to offer a caveat when affording praise and attention to a pagan author lacking in this vital quality. Clearly, Augustine expected those of his readers who knew Varro to share his distaste for how he

5

See Thompson 2013: 51–53.

Of the ninety-three titles listed in August. Retract., only eleven are lost (see Retract. 1.6, 21; 2.5, 11, 19, 27– 29, 32, 35, 46; but on 1.6, see Augustinus-Lexikon s.v. ‘Grammatica (De)’). Similar results for the Indiculum of Possidius: see Augustinus-Lexikon s.v. ‘indiculum, -us’, 4. Many letters and sermons are of course lost, but we have hundreds of each, with more still emerging.

6

7

Quoting a now lost portion of Cicero’s Academica.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 77

14/11/2017 14:56:18

BICS-60-2  2017

78

wrote.8 As for those who did not, then or later, they would have learnt from the City of God that Varro was no model for writing Latin.9 But, as Augustine says, what Varro lacked in elegance he made up for in learning. There are of course many secular Latin works, from Pliny the Elder to the late antique grammarians, that were preserved not as great literature, but because they contained valuable knowledge. Varro’s own surviving works belong in this category. However, teaching on agriculture and language will appear useful wherever there is farming and writing. How useful, in an increasingly Christian world, was an encyclopedic treatise on the religion of ancient Rome? As his title indicated, Varro had sought to record much that, in his own day, was already obscure or obsolete within Roman religion.10 Four centuries later, this could still hold an attraction for pagans who indulged in antiquarian piety or pious antiquarianism.11 But no such attitude was supposed to be compatible with conversion to Christianity — so what attraction could the Antiquitates still hold for Christians? Here, there can be no simple answer, because there was no simple or single attitude of the new Christians towards their pre-Christian past. To be sure, the books used in actual religious practice — e.g. Sibylline prophecies or the books of the pontiffs and augurs — could not be studied or preserved by Christians: they were too tainted with what was for Christians demonic worship. But didactic works on religion were another matter. At the very least, if the classics of ancient literature were to be preserved, some modicum of material on ancient religion was needed to understand them. Nor was the classroom the only place where such knowledge might be sought. Christians may have judged their past very harshly, but they did not wish to forget it. The whole project of the City of God shows that knowledge of the gods from whom they had turned away was part of how Christians sought to understand who they now were. This could then leave space not only for Christian apologetics, but for pagans’ own reports of the ancient beliefs. We may point here to the survival of two voluminous pagan works contemporary with the City of God: Macrobius’s Saturnalia and Servius’s commentary on the Aeneid. However, while the Servian commentary and Macrobius’s Saturnalia contained much on ancient religion, and much of it was owed directly or indirectly to Varro, neither was a work of divine antiquities.12 The only truly religious works of Roman antiquity to have been preserved are Cicero’s and Seneca’s, and both are suffused with a bitter philosophical scepticism, deeply congenial to Christians, towards the whole apparatus of traditional religious belief. For Seneca, the result is generally neglect of Roman tradition in favour of the abstracted theology of Stoicism. Cicero’s great religious treatises, De natura deorum and De divinatione, have on the other hand much to say about tradition, because they put this tradition openly on trial. Of Varro’s position vis-à-vis this project we will say more below. But we may already note excellent reasons why Cicero’s reports on the Roman gods were much more likely to be transmitted than those of his learned friend. Cicero too could be an antiquarian authority, but at the same time, he was fairly brief, eloquent, and, above all, concerned with fundamental questions, such as what the gods were, and how they 8

Similarly, already, Quint. Inst. 10.1.95 (reference owed to the editors).

The impression can only have been reinforced by Augustine’s exhausting summary, at De civ. D. 19.1–3, of Varro on the 288 possible philosophical sects. For Augustine on Varro’s style, see further Hagendahl 1967: 590–92.

9

10

See De civ. D. 4.22, 6.2, as quoted below.

11

See Vallat in this volume, esp. sections 8 and 9.

12

On Varro in Servius see Vallat in this volume.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 78

14/11/2017 14:56:18

DANIEL HADAS: ST AUGUSTINE AND THE DISAPPEARANCE OF VARRO

79

related to humans. Varro in contrast was very lengthy, ineloquent, and devoted primarily to collecting — not criticizing — traditions. Both Varro’s length and the nature of his interests were clear to whoever had access to the City of God. Augustine’s introduction of Varro in book 6 provided not only Cicero’s praise, but (in 6.3) an outline of the whole of the Antiquitates, both the Antiquitates rerum humanarum and the Antiquitates rerum divinarum. From this we learn that, of the sixteen books of the latter, the first is introductory, the final three discuss the gods, and the other twelve are divided into four triads: the first triad treats of priests, the second of temples and holy places, the third of the sacred calendar, and the fourth of rites and sacrifices. In other words, the bulk of Varro’s book was devoted not to the fundamental questions of theology, but to providing a detailed historical exposition of the practice of Roman religion. Augustine’s own interest in this topic was limited: Cardauns finds only a single potential fragment of books 2–12 of the Antiquitates in the City of God.13 We may argue about specific attributions, but the overall picture is certainly correct. Augustine could doubtless have found much to scoff at in Varro’s account of Roman religious practice, and in particular, could have built on his earlier fulminations against games in honour of the gods,14 to which Varro had devoted book 11. But emphasis on the games, and on Roman religious practice in general, belonged to the historical enquiry in books 1–5 of the City of God, where Augustine wished to show that worship of the old gods had not brought Rome its earthly power: for this, he had found the evidence of the historians to be sufficient. In books 6–10, he moves on to the question of which gods are to be worshipped for eternal happiness beyond this world.15 Accordingly, his primary concern is no longer how the Roman gods had been worshipped, but who they were in essence. This is the question to which he seeks answers from Varro, and this could be found only in the first and final three books of the Antiquitates rerum divinarum. Indeed, it is quite impossible to show that Augustine read, or even possessed, the other twelve books in their entirety. True, not every Christian shared Augustine’s exclusive commitment to theology: whoever read and copied Macrobius and Servius may still have wished to quarry the twelve books of Antiquitates that Augustine neglected. But there cannot have been many, as the Roman world faded, who wanted nearly as much as Varro offered. The Antiquitates rerum divinarum were not a handbook, but a magnum opus, whose copying, preservation and transportation would involve considerable time and expense. The same can of course be said of the City of God itself, but in the new world, Augustine’s appeal and authority far surpassed Varro’s. This brings us back to our opening question: should not the success of the City of God itself have promoted the survival of Varro? Even if readers saw that Augustine only engaged with parts of the Antiquitates rerum divinarum, would they not have wished to preserve those parts, just as, for instance, the ‘Dream of Scipio’ had been preserved from Cicero’s De republica, along with Macrobius’s commentary on it? We know the answer is negative. I will now seek to explain this through a review of Augustine’s presentation of Varro in the sphere that interested him: Varro as a theologian, a teacher about the nature of the gods. We will find here that Augustine conferred real value on Varro, not only by presenting him as the worthiest exponent of Roman theology, but also by seeking to show that Varro’s 13

Cardauns 1976: vol. 1, fr. 71. On Augustine’s selective reading of the Antiquitates, see Hagendahl 1967: 602.

14

See in particular De civ. D. 1.32; 2.8–13; 4.26–27.

15

For this division, see De civ. D. 4 praef. and 10.32 (fin).

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 79

14/11/2017 14:56:18

BICS-60-2  2017

80

exposition had in fact considerable common ground with Christian belief. But this paradox led inevitably to Augustine’s presentation of the Antiquitates as a failure: Augustine’s Varro had seen through the lies of Roman religion, but he lacked the courage to state his conclusions openly, and the wisdom and divine grace to replace the old religion with anything better. So the Antiquitates offered only half-truths. Whoever wanted the whole truth behind these could find it, and therefore everything of theological value in Varro, in the City of God itself. *** If Augustine ended up burying Varro, he is nevertheless lavish and persistent in praising him. His introductory chapter (6.2) on Varro’s theology opens: quis Marco Varrone curiosius ista quaesivit? quis invenit doctius? quis consideravit adtentius? quis distinxit acutius? quis diligentius pleniusque conscripsit? Who has enquired into these matters more carefully than Marcus Varro? Who has found out about them with more learning? Who has considered them more attentively? Who has defined them more acutely? Who has recorded them more diligently and copiously? This echoes the passage from Cicero quoted above, and which follows immediately. With the adverbs doctius and acutius, Augustine redeploys Cicero’s two laudatory superlatives: acutissimus and doctissimus. These two adjectives will then recur as a constant refrain when Varro is named in books 6 and 7, and indeed throughout the City of God. He is given some form of both epithets together seven times,16 in addition to eight instances of doctissimus17 and three of acutissimus (or variants) on their own,18 while a further five references to ‘most learned and acute men’19 in fact point again primarily or exclusively to Varro. So, in total, we are reminded some twenty-three times that Varro is doctissimus and/or acutissimus: as often in the City of God, repetition is used to hammer in an idea, and also gradually to refine it. The epithets serve first and foremost to establish common ground between Augustine and his adversaries:20 vos, qui tam doctorum hominum talia scripta legistis et aliquid magnum vos didicisse iactatis (7.22) You, who read writings of this sort by such learned men, and boast that you have learnt some great thing. The City of God was ostensibly addressed to pagans, but in reality, like any Christian apology, was also written for Christians who needed reassurance that they had chosen the right path, and guidance on how to think of the world they had left behind. Many readers in both camps, classically educated as was Augustine himself, would have absorbed a deep attachment to Roman tradition. We can grasp something of their formation through Servius, and of their mature taste through Macrobius. For such readers, Varro was a venerable authority, revered 16

6.2, 6 (bis), 6; 7.5, 9, 28, 30. See also De doctrina christiana 2.17.27.

17

3.4; 4.1; 6.7; 7.5 (bis), 25; 19.22; 21.8.

18

4.31; 7.5; 7.28.

19

6.2, 8; 7.18, 22, 27.

20

On the value of Varro for Augustine’s intended readership, see O’Daly 1996 and Markus 1996.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 80

14/11/2017 14:56:19

DANIEL HADAS: ST AUGUSTINE AND THE DISAPPEARANCE OF VARRO

81

if not actually read. Augustine did read (at least some) Varro, and then adopted Cicero’s praise of him, so that he could begin his dialogue with a show of understanding and respect for his readers’ cultural loyalty. Of course, no real concession was to follow. Rather, Augustine’s project will be to turn the qualities he praises in Varro against this same loyalty. If Varro is the most learned Roman, his facts on religion cannot be questioned, and if he is the most acute, it cannot be claimed that he has misunderstood those facts. Then, if his learning and acuity led to no plausible defence of Roman religious tradition, there would be no higher authority to which its defenders could appeal. The learnedness (doctrina) of Varro’s theology was chiefly manifest in his great catalogues of gods and their functions, and, already before book 6, Augustine signals his utter rejection of Varro’s religious justification for these catalogues: quid est ergo, quod pro ingenti beneficio Varro iactat praestare se civibus suis, quia non solum commemorat deos quos coli oporteat a Romanis, verum etiam dicit quid ad quemque pertineat? (4.22) Why then does Varro boast that he has done some great favour to his fellow–citizens, because he not only names the gods who are to be worshipped by Romans, but also says what pertains to each of them? Augustine is confident none of his readers will be able to give a satisfactory answer to this question, because he expects them to accept that no rational theology can be constructed from traditional polytheism, and that this lack of rationality constitutes a damning flaw. He will illustrate this at great length in books 6 and 7, seeking out logical impossibilities in the attempt to assign specific functions to specific gods, and sneering at the minor gods put in charge of the mundane or the obscene.21 From this perspective, the more one learnt from Varro about Roman religious tradition, the less one could believe that any ingens beneficium came from that knowledge. This use of Varro, or related antiquarian material, as evidence for the prosecution of Roman polytheism is a stock element of Latin apology, inherited by Augustine from Tertullian, Minucius Felix, Arnobius, and Lactantius.22 However, if Augustine had seen in Varro no more than a great compiler and promoter of Roman beliefs, he would hardly have granted him Cicero’s second epithet, acutissimus. In the City of God, Augustine frequently insists on the stultitia of those who accepted those beliefs.23 Varro’s wealth of religious learning would in no way have saved him from this reproach of foolishness, had Augustine claimed that teaching Romans which gods to worship for which purposes was the true nature of Varro’s religious project. In fact, Augustine maintained quite the reverse: whatever Cicero meant by calling Varro ‘most acute’, Augustine sees his acuity precisely in the great measure of religious truth he finds in Varro — truth that by its very nature ran contrary to the orthodoxies of pagan Rome. Thus, in the City of God, Varro’s testimony for the prosecution is by no means reduced to 21

See in particular 6.2, 9; 7.2–4.

There is, as MacRae (2016: 200, n. 6) points out, no overall study of Varro and the Church fathers. For Tertullian, see Barnes 1985: 196–97; for Minucius, Clarke 1974: 285, 292, 300, 305, 307, 402; for Arnobius, Champeaux 2007: xiii–xxiv; for Lactantius, Ogilvie 1978: 50–55. Firmicus Maternus does not seem to have used Varro: see Turcan 1982: 49–59. 22

23

3.20; 4. 23, 30; 6 praef. 1; 7.19, 27; 8.24.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 81

14/11/2017 14:56:19

BICS-60-2  2017

82

the evidence he can provide of the shameful character of the Roman gods: he is also to play that other great role of secular authors in the apologetic tradition, the pagan authority who has recognized Christian truths avant la lettre.24 This unexpected role was not entirely new to Roman antiquarianism. Clement of Alexandria25 and Tertullian26 had already noted the tradition that Numa Pompilius had forbidden the cult of idols. Augustine expands on this: the introduction of idols after Numa was, for Varro, the addition of new errors into an already erring religion. Already erring because there was in fact only one all-powerful God, who could be identified not only with the Roman Jupiter, but even with the nameless God of the Jews.27 Such statements were of course greatly to Varro’s credit: quapropter cum solos dicit animadvertisse quid esset Deus, qui eum crederent animam mundum gubernantem, castiusque existimat sine simulacris observari religionem, quis non videat quantum propinquaverit veritati? (4.31) Therefore, since he says that they alone understood what God was, who believed him to be a soul governing the world, and since he believes that religion is practised more purely without idols, who cannot see how near he came to the truth? To claim that a pagan author had approached the truth was, as just stated, a standard apologetic strategy. But, in the earlier apologists, we often find that, once such an author is named and quoted, he is left behind, with little discussion of how a given pagan came to make statements that bolstered Christian beliefs. Augustine was more ambitious: he believed he could show that, read as a whole, Varro’s theology, whose pre-eminent authority his adversaries had to acknowledge, was in fact profoundly at variance with their own beliefs. For this, the statements pointing towards monotheism and the rejection of idols were a useful foundation. But Augustine was still required to provide an explanation of how such statements could be reconciled with the ostensibly traditional piety of the Antiquitates rerum divinarum. Taking a cue from Tertullian,28 Augustine approached his problem of identifying Varro’s true beliefs by exploiting his celebrated division of theology into three types (genera), each with its allocated practitioners: the mythical (fabulosum) theology for poets, the physical (naturale) for philosophers, and the civil (civile) for the people.29 This has become Varro’s best-know religious idea, but it is far from clear that the prominence given to it by Tertullian and Augustine matches its importance in the Antiquitates,30 which after all divides gods not according to these three types, but into dei certi (‘certain gods’), dei incerti (‘uncertain gods’), and dei praecipui atque selecti (‘foremost and select gods’).31 Now, that division For this topos in the earliest period, see Bardenhewer 1913: 178–79. It blossoms with Clement of Alexandria and Lactantius. 24

25

Strom. 1.71.

26

Ad nat. 2.17.12; Apol. 25.12.

Against idols: 4.9, 31; 7.5. New error added to old: 4.31. Only one God: 4.9, 31. Jupiter and the Jewish God: 4.9; 19.22 (clarified by De consensu evangelistarum 1.22.30). 27

28

Ad nat. 2.1–8.

29

De civ. D. 6.5. The partition is not original to Varro: see Lieberg 1973. See also MacRae, p. 43 in this volume.

30

Cardauns (1976) plausibly attributes all material on the division to the introductory first book of the Antiquitates.

31

As set out in De civ. D. 6.3.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 82

14/11/2017 14:56:19

DANIEL HADAS: ST AUGUSTINE AND THE DISAPPEARANCE OF VARRO

83

could surely have provided much further fodder for Augustine’s mocking of Roman tradition.32 But the division of the three theologies brought him closer to the question on which his reappropriation of Varro would turn: what Varro believed to be the nature of the gods. With the first two theologies, it was simple enough for Augustine to establish common ground between Christian belief on divine nature and Varro’s own. Varro had followed the old Greek philosophical tradition to declare that the poets had spoken multa contra dignitatem et naturam immortalium ficta (‘many fictions against the dignity and nature of the immortals’, 6.5). That same tradition had already been adopted without hesitation by Christian apologists.33 Augustine himself had thus devoted much of the first five books of the City of God to arguing that the religious institution of ludi, with the poetic performances they entailed, showed the demonic activity behind the Roman cult.34 He was then content to declare that Varro’s own negative comments showed full agreement with Christians’ unqualified rejection of poetic myths as a source of theological truth. Conversely, Augustine sought to infer that the natural theology of philosophers was for Varro the truest of the three kinds of theology: ‘prima,’ inquit, ‘theologia maxime accommodata est ad theatrum, secunda ad mundum, tertia ad urbem.’ quis non videat, cui palmam dederit? utique secundae, quam supra dixit esse philosophorum. hanc enim pertinere testatur ad mundum, quo isti nihil esse excellentius opinantur in rebus. (6.5) ‘The first theology,’ he says, ‘is best suited to the theatre, the second to the universe, the third to the city.’ Who cannot see to which one he has given the prize? To the second, of course, which he said above belonged to the philosophers. For he testifies that this pertains to the universe, and they believe that nothing that exists is greater than the universe. Tertullian had responded to Varro’s philosophical theology by mocking the diversity of philosophical opinions, thus marking the beginning of much hostile language towards philosophy in the early Latin Church.35 But this was not Augustine’s approach. Rather, he chose to emphasize — and no doubt to exaggerate — Varro’s preference for the philosophers’ gods. This was again to bring Varro closer to Christian truth: Augustine himself believed that, of all the manifestations of non-Christian religious thought, philosophy alone had come nearest to that truth.36 This will be the subject of book 8 of the City of God. There Augustine will of course find pagan philosophy wanting, but his opening statement that verus philosophus est amator Dei (‘the true philosopher is the lover of God’, 8.1) points to the care and respect with which he will approach the philosophical tradition, entirely different from the scorn and disgust he reserves for all forms of traditional Roman religion. So, in attacking the poets and praising the philosophers, Varro could be enlisted on the side of truth. But this was far from enough, because the civil gods remained, and they were after all both the real subject of the Antiquitates rerum divinarum (dei certi, incerti, 32

Although at De civ. D. 7.1 Augustine frowns at Tertullian’s joke on the division (Ad nat. 2.9.5).

33

See G. van der Leeuw, RAC s.v. ‘Anthropomorphismus’ (to which add Pl. Resp. 377e–383c).

34

See n. 14 above.

Ad nat. 2.2–6. The Latin anti-philosophical tradition finds it fullest expression in the third book of Lactantius’s Divinae institutiones.

35

36

See already 6.1.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 83

14/11/2017 14:56:19

BICS-60-2  2017

84

and selecti are all civil),37 and the gods who were primarily at stake in the City of God. Its impetus was a controversy about which gods’ worship should receive state approval, and such worship was precisely the business of civil religion. Hence Augustine’s title: dei civiles are the gods of the civitas. And here Varro and Augustine surely had to part ways: was not the very purpose of the Antiquitates to document and enshrine the religious traditions of the Roman people, the very object of Augustine’s scorn and disgust? Augustine’s response shows the power of his focus on the tripartite scheme. He claims that by presenting his religious antiquities as the civil theology within this scheme, Varro had quite deliberately and fatally undermined them. Augustine argues as follows. Firstly, only the philosophical theology, which pertained to nature, described the gods as they really were.38 Varro had made this very clear by his choice to place his divine antiquities, i.e. his treatise on Rome’s civil theology, after the human antiquities: genuine gods should come before humans, so Varro was indicating that the civil gods were human creations.39 Then, by juxtaposing attacks on the gods of poetry with his ostensibly neutral accounts of Roman tradition, Varro was implicitly stating that there was no real difference in value between the two: the performance of ludi, with their reprehensible poems, was after all a civil institution,40 while civil traditions contained exactly the same sort of disgusting beliefs about the gods as the poets’ works.41 Augustine returns again and again to these conclusions in book 6, so that it becomes a long demonstration of the thesis on Varro announced at its beginning: si rerum velut divinarum, de quibus scripsit, oppugnator esset atque destructor easque non ad religionem, sed ad superstitionem diceret pertinere, nescio utrum tam multa in eis ridenda contemnenda detestanda conscriberet. cum vero deos eosdem ita coluerit colendosque censuerit, ut in eo ipso opere litterarum suarum dicat se timere ne pereant, non incursu hostili, sed civium neglegentia […] et tamen ea legenda saeculis prodit, quae a sapientibus et insipientibus merito abicienda et veritati religionis inimicissima iudicentur: quid existimare debemus nisi hominem acerrimum ac peritissimum, non tamen Sancto Spiritu liberum, oppressum fuisse suae civitatis consuetudine ac legibus, et tamen ea quibus movebatur sub specie commendandae religionis tacere noluisse? (6.2) If he had been the opponent and destroyer of the allegedly divine matters, about which he wrote, and had stated that they pertained not to piety, but to superstition, I don’t know if he could have written so much that was risible, despicable, and hateful about them. But since he so worshipped those same gods and recommended their worship, that in these very writings of his he says that he is afraid these things may perish, not by enemy invasions, but by citizens’ neglect […] and yet he handed down for reading to future generations things which are rightly judged by both the wise and the foolish to be repulsive and entirely inimical to true piety: what are we to think, except that this man — most clever and knowledgeable, but not set free by the Holy Spirit — was oppressed by the custom and laws of his city, and yet was 37

See 7.1, 23 (quoted below).

38

4.31; 6.4–6.

39

6.4.

40

6.5–6.

41

6.7–9.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 84

14/11/2017 14:56:19

DANIEL HADAS: ST AUGUSTINE AND THE DISAPPEARANCE OF VARRO

85

unwilling to keep silent about those things that troubled him, albeit under the guise of endorsing piety. In short, by the contents and structure of his Antiquitates, Varro had provided all the information an oppugnator atque destructor of the civil theology would need. Nor had he done so blindly: the true mark of his acuity was his implicit message that his great learning could offer no sustainable defence of the horrors of traditional paganism. We may of course doubt whether Augustine was right to read Varro this way.42 He surely did not see his three theologies as entirely incompatible, and may rather have believed each potentially served a necessary role. But our question here is not whether Augustine was right, but how his judgment affected Varro’s survival. If Augustine’s Varro was a friend to monotheism, to philosophy, and to the Jews, and a subverter of Roman religion and myth, were not the Antiquitates rerum divinarum just such a book on pagan religion as was likely to be sought and copied in the Christian centuries? A first answer is that Augustine makes it perfectly clear that his conclusion that Varro was repulsed by civil theology and civil cult is not bolstered by any explicit statement in the Antiquitates. In the passage just quoted, he claimed that Varro was not free (non […] liberum) to tell what he truly thought; accordingly he spoke with subtle slyness (subtili significatione, 6.4; subtilitate, 6.8), and his meaning had to be teased out by those who could match Varro’s own acuity (neque hoc aperte dicere voluisse, sed intellegentibus reliquisse; ‘he did not want to say this openly, but left it to those who could understand’, 6.4).43 The corollary was that, for the less alert, Varro did indeed seem a friend of the civil gods. Indeed, for Varro to play the role Augustine assigned him in the City of God, this had to be so: otherwise, he could not have been the authority on Roman religion whom Augustine’s adversaries were expected to acknowledge. Augustine is equally clear, and indeed emphatic, on the reason Varro hid his true beliefs: he was a coward.44 At the turn of the fourth century, the persecution of Christian martyrs was still a recent memory, and one diligently maintained by cult and writing. Christians had been persecuted precisely because they refused to conform to the Roman civil cult. So Augustine naturally assumed that, had Varro written openly against that cult, he would have put himself at risk. Varro lacked the courage to embrace this danger, and so he hid his beliefs. The truth could be deduced from his writings, but Varro himself remained silent (ipso tacente veritas clamat; ‘he is silent, but truth cries out’, 6.4). Augustine also understood perfectly well that Christians were persecuted not for private scepticism about the traditional gods, but for carrying that scepticism to the logical conclusion of refusing to participate in their worship. Scepticism itself, he knew, had been openly expressed by pagans who had dared to be freer than Varro. In book 6, Augustine makes this point by interrupting his discussion of the Antiquitates to contrast it with Seneca’s Contra superstitiones: libertas sane, quae huic defuit, ne istam urbanam theologian theatricae simillimam aperte sicut illam reprehendere auderet, Annaeo Senecae, quem nonnullis indiciis So e.g. O’Daly 1996. Kinder to Augustine is Rousseau (2009), but with no mention of Augustine’s claim that Varro secretly attacked the civic cult. 42

43

See also 7.17.

The language of fear is frequent in Augustine’s explanation of Varro’s practice: confugiat (4.27); vereris (6.6); metuebant, non audebant (6.8); non ausus est (6.10); timoris silentio (7.5); trepidus (7.17); trepidare (7.34).

44

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 85

14/11/2017 14:56:19

BICS-60-2  2017

86

invenimus apostolorum nostrorum claruisse temporibus, non quidem ex toto, verum ex aliqua parte non defuit. adfuit enim scribenti, viventi defuit. (6.10)45 Indeed the freedom, which failed in this man [Varro], so that he did not dare criticize openly the theology of the city, so similar to that of the theatre — for Annaeus Seneca, whom we find from some evidence to have flourished at the time of our Apostles, this freedom did not fail: not entirely, but to some extent. It was there for him when he was writing, but failed him when he was living. By viventi defuit, Augustine means that Seneca had still participated in the public worship of the gods he attacked in his writings. Elsewhere, he gives similar accounts of Socrates46 and Cicero.47 This too was cowardice, but of a lesser degree than Varro’s. These examples showed that pagans were capable of more truthfulness and bravery than Varro had dared to show. We note here again that, although we have only fragments of the Contra superstitiones, Cicero’s great texts on religion survive, and Seneca remains among the best transmitted of classical Latin authors. I do not mean that the survival of Seneca’s or of Cicero’s philosophical works is due primarily to these remarks in the City of God. Both benefited more generally from the gradual turn towards philosophical culture that characterized the Christianization of the Latin world. In itself, this turn was hardly favourable to the Antiquitates, which, as Augustine makes clear, contained no systematic exposition of Varro’s philosophical theology. On the other hand, Augustine had argued that Varro’s real preference was for this theology, and had given this preference his approval. Only, Augustine then undermined this approval by an extremely stringent judgment on the actual contents of Varro’s philosophical remarks on the gods. As his discussion of the Antiquitates progresses, Augustine shows that, despite their focus on civil religion, they did indeed contain elements of a philosophical reflection on religious tradition:48 the explanation of the pedagogical role of idols, exegeses of myth and cult as natural allegories (physiologicas interpretationes 6.8; interpretationes physicas, 7.5), and the attempt to define the divine’s true nature as the anima mundi. Clearly, this is not the sort of philosophy to which Augustine could give even partial assent: it contains none of the radical calls for reform of a Plato or an Epicurus, but tends rather to save the appearances of existing practice and even myth. So, in analysing this material, Augustine no longer seeks agreement between Varro and Christian teaching, but turns to straightforward polemic against Varro’s ideas: the justification of idolatry both contradicts Varro’s own statement that it was a form of error,49 and ignores the truths of Euhemerism;50 the allegories, to which

See all of 6.10–11. The juxtaposing of Seneca and the Apostles seems influenced by the apocryphal correspondence between Seneca and St Paul (see Encyclopedia of Ancient Christianity, s.v. ‘Seneca and Paul, Correspondence of’). 45

46

De vera religione 1–2.

De civ. D. 4.30. In contrast, Arnobius (3.6–7) sees Cic. Nat. D. as a fearless expression of scepticism, and claims pagans had campaigned for it to be destroyed (a story unknown to Augustine?). 47

At least part of this material was in the prologue to book 16 of the Antiquitates rerum divinarum: see De civ. D. 7.7, 17; Cic. Acad. 1.2.8.

48

49

7.5.

7.18, 27–28, 35. Varro of course admitted some Euhemerist accounts, as Augustine acknowledges elsewhere (De civ. D. 18.5; De consensu evangelistarum 2.12.27). 50

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 86

14/11/2017 14:56:19

DANIEL HADAS: ST AUGUSTINE AND THE DISAPPEARANCE OF VARRO

87

Augustine devotes many pages, are absurdities;51 the anima mundi theory both wrongly postulates God to be a soul, and serves as a specious justification for polytheism.52 Here we are faced with an inconsistency: Augustine had used Varro’s tripartite scheme to argue that his preference for the philosophical theology contained a hidden attack on the civil one. Yet, when Augustine addresses the details of the philosophical theology, he shows us a Varro who was using philosophy to justify civil practice and tradition. Augustine does not note this contradiction, or attempt to resolve it into any unifying presentation of Varro’s thought. Rather, as the more philosophical statements in book 7 are attacked, Varro’s singularity is increasingly elided: positions are attributed to an unspecified ‘they’ and Varro’s own ideas mixed with other sources.53 Now Varro is no longer a secret opponent of pagan beliefs, but their foolish representative. True, as the book progresses towards its conclusion, Augustine does begin once again to single out Varro, and address him directly.54 But when Augustine applies Cicero’s superlatives one last time, it is now in a tone of pity mixed with scorn — all Varro’s learning and wit have ultimately proved to be of no avail:55 quid igitur valet, quod vir doctissimus et acutissimus Varro velut subtili disputatione hos omnes deos in caelum et terram redigere ac referre conatur? non potest; fluunt de manibus, resiliunt, labuntur et decidunt […] quid solidum quid constans, quid sobrium quid definitum habet haec disputatio? […] quorum sacra Varro dum quasi ad naturales rationes referre conatur, quaerens honestare res turpes, quo modo his quadret et consonet non potest invenire. (7.28; 7.33)56 What good is it then, that the most learned and acute man, Varro, tries, as if by some subtle argument, to reduce and bring back all these gods to sky and earth? He cannot do so; they pour out of his hands, they jump back, slip away, and fall down [...] What in this argument is solid, what is steadfast, what is sober, what is established? [...] When Varro tries to bring their rites back to natural reasons, seeking to dignify disgusting things, he can find no way to make them square and agree. This shift in Augustine’s judgment of Varro, and in the importance he affords him, is no mere incoherence. There were two main props to the thinking of those of Augustine’s contemporaries who were unwilling that the Roman world become an entirely Christian one. The first was the mos maiorum, the belief — itself deeply ensconced in Roman tradition — that the altering or abolition of Roman religious practice could only be impious. The other was the exegetical tradition, complex, variegated, and itself ancient, that sought to read into traditional beliefs systematic, more or less philosophical, accounts of the workings of the universe. Augustine found both these props in Varro, identifying them respectively with the civil and philosophical theology, and he of course sought to destroy them both. But he does not treat them both equally. For the civil theology, he is content to dialogue with Varro 51

6.8; 7.9–13, 16, 19–27.

52

7.5–6, 9, 22 (and already 4.31).

53

Note the many third-person plural verbs (opinantur, dicunt, perhibent, etc.) at 7.1–3, 5, 8–14, 16.

The shift happens at 7.17, where explicant, inplicant, insiliunt, resiliunt are contrasted with ipse Varro. Augustine then briefly praises Varro’s admissions of uncertainty. But he will turn on him again in 7.19.

54

55

To some extent, this is anticipated in 6.6, but here Augustine says only that Varro was afraid to speak his mind.

See also 7.30, and note that, at the end of 7.33, Varro’s shaming of the civil theology via the poetical is apparently no longer intentional. Elsewhere in De civ. D., Varro is likewise presented as believing in the mythological material he presents (8.26; 18.9, 16–17, but contrast 18.10). 56

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 87

14/11/2017 14:56:20

BICS-60-2  2017

88

alone. As an antiquarian, Varro was unrivalled, so that it was triumph enough for Augustine to show that Varro’s teaching on Roman tradition cast profound doubts on its religious validity. Augustine could then close his discussion of Roman religious tradition in the City of God. But, as a philosophizing polytheist, Varro was merely one link in a chain of ideas that were still widely accepted, and here Augustine could not settle the matter only by dealing with Varro. Augustine does refute him, lest Varro’s defenders believe an appeal to his philosophy could be enough to save his and their polytheism. But above all, he dismisses Varro as an incompetent in this field: redeat ergo ab hac, quam theologian naturalem putat, quo velut requiescendi causa ab his ambagibus atque anfractibus fatigatus egressus est; redeat, inquam, redeat ad civilem. (7.23) Let him then go back from this, which he thinks to be natural theology, where he has gone out, as if to take a rest, exhausted by these twistings and turnings; let him go back, I say, let him go back to the civil theology. Indeed, Augustine could learn from Cicero that Varro did not really write philosophy,57 and Augustine’s own reading in any case confirmed that pagan philosophy offered far greater challenges to Christianity than anything to be found in Varro. So he could not expect those non-Christians who considered their beliefs to be essentially philosophical to be content with a polemic against Varro. Accordingly, book 7 marks the end of Augustine’s dialogue with Varro, but it is only the beginning of his engagement with pagan philosophy. The point is made very clearly in the opening to book 8: de theologia quippe, quam naturalem vocant, non cum quibuslibet hominibus [...] sed cum philosophis est habenda collatio […] hi iam etiam Varronis opinionem veritatis propinquitate transcendunt, siquidem ille totam theologian naturalem usque ad mundum istum vel animam eius extendere potuit, isti vero supra omnem animae naturam confitentur Deum. (8.1) Indeed, for the theology which they call natural, our discussion is to be held not with just any men, but with philosophers […] These men [the Platonists] have also already surpassed Varro’s beliefs, drawing nearer to the truth, since he was able to extend the whole of natural theology as far as this universe or its soul, whereas they confess a God above any soul’s nature. Whatever their subsequent errors, the best of the philosophers were able to start from the right first principles. Varro’s brand of philosophical theology, in contrast, remained hopelessly enmeshed in the base and earthly realities and imaginings of Roman tradition: ipsas physiologias cum considero, quibus docti et acuti homines has res humanas conantur vertere in res divinas, 58 nihil video nisi ad temporalia terrenaque Acad. 1.2.4–8. The De philosophia, which Augustine describes in De civ. D. 19.1–3, must in some sense have been a book of philosophy (its contents are unknown beyond what Augustine tells us. See Cardauns 2001: 69–71). Augustine may not have known De philosophia when he wrote De civ. D. 6–7, circa 8 years earlier than De civ. D. 19 (O’Daly 1999: 34–35). Nowhere does Augustine attempt to reconcile the Antiquitates with Varro’s claim in De philosophia to belong to the old Academy, although this matched the position assigned to him by Cicero in the Academica. Did Varro himself attempt such a reconciliation? 57

58

My emendation, based on Healey’s translation (1909: 221).

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 88

14/11/2017 14:56:20

DANIEL HADAS: ST AUGUSTINE AND THE DISAPPEARANCE OF VARRO

89

opera naturamque corpoream vel etiamsi invisibilem, tamen mutabilem potuisse revocari; quod nullo modo est verus Deus. (7.27)59 When I consider those physiologies, by which learned and acute men try to turn these human things into divine ones, I see that they could be referred to nothing except temporal and earthly works, and to bodily nature, which, even when invisible, is still changeable. In no way is this the true God. *** Let us now attempt to sum up how the religious teaching of the Antiquitates rerum divinarum would appear to the reader who met Varro’s book through the City of God. First of all, he would have learnt that the essence of Varro’s religious thought was to be found in the three theologies. He could then have considered each in turn. For the poetic theology, Varro offered an attack on mythology and poetry of which Augustine approved. However, as Augustine himself pointed out, the mythical theology was not easily separated from the civil. For this theology, Augustine had reinforced the authority of Varro’s learning. But he had also portrayed him to be a fundamentally half-hearted scholar: his real purpose, which he had not dared to confess, was as much to reject traditions as to record them. Augustine had then argued that Varro sought truth only in the theology of the philosophers. But this search had faltered, and even led Varro back to accepting both the myths and the civil traditions which he had rightly spurned. In such an account, the Antiquitates rerum divinarum emerges as a failed project: a product not only of error, but of cowardice and confusion. How many readers would have wished to seek out or copy sixteen books of such a work? Moreover, there is no better way of erasing a book than to replace it. Virgil supplanted Ennius; Livy eliminated the need for all previous annalists; the City of God itself came close to killing all earlier Latin apologies.60 It could at the same time be read as a new, Christianized Antiquitates rerum divinarum,61 covering all three of Varro’s theologies: the poetic in books 1–5, the civil in books 6–7, and the philosophical in books 8–10.62 Augustine reproduced enough of Varro to satisfy much antiquarian curiosity, while he was a surer and safer guide to the three theologies. He could better attack the poets’ lies, since his vision of God was truer and purer. He could say openly, indeed shout from the rooftops, what Varro had dared only to imply against the civil theology. For the philosophical theology, he was a better source, both for its pagan errors, where Varro was but a weak spokesman, and for its true manifestation in Christian revelation. And, beyond all this, there was the great purpose of the City of God: to provide a new Antiquities, a new past, to the Latin West, no longer the 59

Cf. also 8.5.

Arnobius and Minucius Felix survive in only two manuscripts, Firmicus Maternus and Tertullian’s Ad nationes in one. The traditions of Tertullian’s Apology and Lactantius’s Divinae institutiones are larger, but still dwarfed by that of the City of God. For Tertullian, Arnobius, and Lactantius, doubtful orthodoxy must also have harmed transmission. 60

So already MacRae (2016: 137–38) with references to earlier literature, and in particular to the pairing of Augustine and Varro in Sidonius Apollinaris (Epist. 2.9.4) and in St Isidore of Seville (Etym. 6.7.1), to which add Claudianus Mamertus, De statu animae 2.9 (see Shanzer 2005: 83). 61

Augustine himself divides books 1–10 otherwise: 1–5 deal with worship of the pagan gods for this life, 6–10 for the next (see De civ. D. 6. praef. 10.32; Retract. 2.43.1). But if we see Varro’s three theologies as a parallel substructure, we understand better why much of these books does not treat directly of what is ostensibly the question at hand. 62

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 89

14/11/2017 14:56:20

BICS-60-2  2017

90

sacred and profane history of Rome, but the history of the Jewish people, which was now the history of all who were called to be saved.63 We cannot know how Augustine would have judged the successes and failures of this project in the centuries that followed his death. By the same token, we cannot know whether he ‘would have relished the irony that most of what now survives of the works of Varro quoted by him in the City of God does so only through his quotations’.64 But we can see an eerie foreshadowing of Varro’s loss in book 7’s final chapters on Varro. Augustine concludes with Varro’s tale of the theological books of Numa Pompilius.65 These were buried by Numa, but later discovered by a ploughman, who took them to the senate. The senate decreed that they were to be burned, and Augustine infers that Numa had consigned to them either the secret demonic origins of Roman cult, or the shameful fact that Roman gods were in fact humans. This leads to Augustine imagining Varro’s own books being burnt: if Numa’s books had contained the same physiological explanations of the gods as Varro’s, the senate would have ordered the burning of the Antiquitates rerum divinarum. They did not, and this shows that Numa’s books hid other truer secrets. Augustine then praises God’s hidden providence (occulta Dei providentia, 7.35), which had neither allowed Numa to burn his books, nor the senate to abolish their memory, since Varro had preserved it. We may reflect similarly on the Antiquitates rerum divinarum: the Roman senate had no objection to them, but the early Christian centuries no longer wished to preserve what they taught about the gods. Whatever providence watches over scholars then saved their memory through the City of God. More circumspect than Augustine, we are less ready to guess what the lost books said. As a result, what Augustine has left us makes us long for the rest: ‘Among the lost works of republican prose, the Antiquitates is perhaps the one we most sorely miss.’66 The wish for more Antiquitates is no novelty: we find it already in the praise for Varro in John of Salisbury,67 or Petrarch.68 Such praise was based less on Varro’s extant works, which were little used by Petrarch and not even known to John,69 than on the shadow of the Antiquitates found in the City of God. In our more scientific age, praise is largely replaced by scholarship: painstaking attempts to reconstruct and interpret Varro, of which the present volume is a manifestation. This is an honourable enterprise, a search to know and understand the past. But we must not forget that it has its own history, to which the writing of the City of God and the loss of Varro are the ineluctable preludes. I do not think we can see these simply as inconvenient obstacles to the study of Varro: they are the conditions of that study. When Augustine wrote, the divine world Varro had documented retained for many its numinous force. The City of God’s polemic against the gods would otherwise have This aim is foreshadowed (among other places) at 7.31–34, as the discussion on Varro ends. Its realization is the purpose of the final twelve books of the City of God. 63

64

Vessey 2012: 23.

65

7.34–35. The source is De cultu deorum (see n. 2 above, p. 76), not the Antiquitates.

66

R. A. Kaster in OCD4, s.v. ‘Terentius Varro, Marcus’.

Entheticus maior 1177–90 (Laarhoven 1987: vol. 1, 182–83). Still, here and at Policraticus 7.9, John retains Augustine’s criticisms of Varro.

67

See Nolhac 1907: vol. 2, 110–15; Pacca and Paolino 1996: 444–45; Piras 2007–08. The culmination of Petrarch’s many laments for the loss of Varro is his letter to him (Familiarum rerum libri 24.6). 68

69

Webb 1909: vol. 1, xxix.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 90

14/11/2017 14:56:20

DANIEL HADAS: ST AUGUSTINE AND THE DISAPPEARANCE OF VARRO

91

been a needless project. As long as Roman gods lived, those who did not wish to worship them preferred to encounter them through Augustine, rather than through Varro. It was only once the gods were securely dead that boundless curiosity could emerge about all aspects of their worship, and the disappearance of the Antiquitates could be felt as a sorry loss. But what we have lost is more than a text: it is the beliefs, and indeed the doubts, that once caused that text to be written and preserved. After Augustine, after two thousand years of Christian history, these cannot be resurrected. The retrieval of the entirety of the Antiquitates rerum divinarum would not suffice. For better and for worse, we must always view Roman religion as outsiders. Augustine had quoted Cicero’s famous praise of the Antiquitates: nos, inquit, in nostra urbe peregrinantes errantesque tamquam hospites tui libri quasi domum reduxerunt (‘ “when”, he said, “we were sojourning and wandering like strangers in our own city, your books brought us back home” ’).70 Augustine’s aim was to make his readers strangers once again in Varro’s city, but citizens in the City of God. All may not be eager to enter the new city, but to the old city there is no returning. King’s College London

70

De civ. D. 6.2, quoting Acad. 1.9. Similarly, MacRae 2016: 137. See also MacRae in this volume.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:39 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 91

14/11/2017 14:56:20

VARRO IN VIRGILIAN COMMENTARIES: TRANSMISSION IN FRAGMENTS DANIEL VALLAT Abstract: This paper analyses the transmission of Varro in late antique Virgil commentaries. Various problems are identified and discussed: the reliability of authors’ names and titles of works in citations and testimonies; different forms of quotation; complications entailed by manuscript transmission; the delimitation of the fragments; the indirect transmission of Varro already in antiquity; the status and function of Varro in a Virgil commentary. Finally I suggest that Varro had a special if implicit status in fourth-century ideological debates, in the tacit rivalry of grammarians with Christian polemicists. Keywords: Virgilian commentaries; grammarians; Servius; manuscripts; textual transmission; quotation; fragments

Only a fraction of Varro’s voluminous oeuvre has survived. The only fully preserved work is the Res rusticae; only six books out of twenty-five of De lingua Latina have survived. For the rest, we have only fragments preserved through indirect transmission. Authors often drew on Varro’s works without acknowledging their source: Dionysius of Halicarnassus used the Res humanae for his history of the origins of Rome; Ovid brought the Res divinae into his Fasti; and Plutarch built on Varro’s works in his Roman Questions. These are tacit appropriations. Except for fragments transmitted by Pliny the Elder in the first century ad and Gellius in the second, most of our fragments of Varro come from fourth-century sources. Apart from fragments and testimonies preserved by Christian polemicists such as Augustine,1 the vast majority of our fragments of Varro come from grammarians, lexicographers, and commentators. I focus here on the Virgilian commentaries and review the various problems posed by the presence and the exploitation of Varro in these works. 1. Virgil commentators The commentaries on Virgil constitute the largest corpus of ancient exegesis on a nonChristian Latin author, even though only part of it has come down to us, and what has survived is heterogeneous in form and nature.2 Among Virgilian commentators we discern only one strong personality, Servius, and date him to the late fourth/early fifth century.3 Undoubtedly the most important ancient exegete of Virgil’s works, Servius’s commentary incorporates — but radically condenses — inherited critical material.4 Indeed, in line with 1

See Hadas in this volume.

For the lost material, see Ribbeck 1866: 114–200; Timpanaro 1986. I omit here the commentary of T. Claudius Donatus (Interpretationes Vergilianae) and Macrobius’s Saturnalia, two works which, each in their own way, depart from the grammarians’ world and provide, respectively, a rhetorical commentary and a typical scholarly study about the veteres in general and Virgil in particular — but not a line-by-line commentary.

2

3

On the date of Servius, see Brugnoli 1988; Murgia 2003; Velaza 2008.

4

See Lloyd 1961; Cameron 2011: 408–09; Vallat 2016.

BICS-60-2  2017

92

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 92

14/11/2017 14:56:20

DANIEL VALLAT: VARRO IN VIRGILIAN COMMENTARIES

93

ancient scholarly practice, Servius’s commentary is not original stricto sensu, but rather a compilation, enriched by more personal remarks. It is clear, for example, that Servius significantly abridged the (now lost) authoritative commentary of Aelius Donatus, rendering it suitable for use in the classroom by removing the erudite digressions. Vast as it is, Servius’s commentary still represents a digest of the first four centuries of Virgil criticism, but unlike, say, Macrobius’s Saturnalia, it is tailored to an audience of pupils. The Servius Danielis (SD) is an anonymous set of scholia: it consists of non-Servian additions found in some manuscripts of Servius.5 They are named after the editor Pierre Daniel, who first published most of the additions in 1600. Their origin is still debated, and their possible provenance from Donatus’s lost commentary was a major focus of twentiethcentury Servian scholarship, even though this theory has more recently aroused scepticism.6 Whatever their origin, the additions are typical of a commentum variorum, a stratified mix from different periods: some material appears to be later than, and drawing on, Servius, while other notes may be earlier, as they transmit passages from pre-Servian exegesis that Servius deliberately omitted. The Veronese scholia are marginalia in a palimpsest of Virgil (Verona, Bibl. Cap. XL.38, fifth–sixth century). Though they are very imperfect and full of lacunae, they nonetheless provide valuable material that seems to be independent of Servius and Servius Danielis.7 The other corpora are also incomplete. A commentary on the Bucolics and Georgics was falsely attributed to Valerius Probus of Berytus (first century ad). The attribution was probably fixed in the Renaissance, since there is no medieval manuscript for this text, but only manuscripts of the late-fifteenth century and the editio princeps of 1502.8 The preserved parts of pseudo-Probus’s text are erudite and elaborate, preserving scholia which are otherwise unattested, but apparently incomplete, since nearly all linguistic material has been removed. It appears to contain much ancient material, and may have been reorganized later. The last three texts (Philargyrii Explanationes in Bucolica, Brevis expositio Georgicorum, Scholia Bernensia) are interrelated and probably derive from a late antique common source; early Irish glosses attest to the insular circulation of the source text during the seventh and eighth centuries. These three texts point to a source containing a version of the Servian commentary augmented with material from other scholia.9 Philargyrius’s text on the Bucolics has survived in two versions (Explanationes I and II). The Brevis expositio Georgicorum is extant until G. 2.542.10 Finally, the Berne scholia are named after the main manuscript which transmits them (Bernensis 172).11 See Thomas 1880; Thilo 1881. SD additions to the text of Servius are sometimes ineptly stitched on (Goold 1970: 109–10). At times the text of Servius and SD are interwoven under one lemma; for example, both versions of the note on the secret name of Rome (A. 1.277) cite Varro, but that of SD is more complete and gives details missing in Servius: it might therefore seem older; but the extra details could have derived from other sources such as Solinus, Festus, etc., and so the SD note is not necessarily to be dated closer in time to Varro. For other notes imbricating Servius and SD which mention Varro, see A. 1.532; 2.166; G. 1.166; 1.270; 4.63.

5

6

See Lloyd 1961; Brugnoli 1988: 89; Cameron 2011: 573–74; Vallat 2012a.

7

See Daintree–Geymonat 1988; Baschera 1999.

8

See Gioseffi 1991.

9

See Daintree–Geymonat 1988.

10

For all these texts, I have used Hagen’s 1902 edition.

11

We still depend on Hagen’s 1867 edition.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 93

14/11/2017 14:56:20

94

BICS-60-2  2017

2. Varro’s name: presence and absence The presence of Varro in Virgilian commentaries is much more extensive than mere mention of his name might imply. Internal correspondences can help us identify a Varronian source where none is acknowledged; for example, Servius twice (A. 1.54 and 5.145) derives the word carcer (‘prison’) from the verb arceo, but only names Varro in the second instance.12 Without it, we would not know that Servius had borrowed and summarized (and perhaps misunderstood) the etymology from Varro (L. 5.151). Another example may illustrate the phenomenon of name-vanishing: in G. 1.99, Servius Danielis preserves the names of authors (Varro and Nigidius), while the Brevis expositio replaces them with indefinite pronouns (alii […] alii). Thus, names are easily dropped in the manuscript transmission, especially for non-literary texts, which can be freely reworded.13 Since this kind of tacit transmission was extremely common, it is difficult to identify unacknowledged sources and to quantify the phenomenon in general. We may surmise, however, that Varro was most often left unacknowledged where commentators were drawing on his primary fields of specialism: the Latin language, Roman history and traditions, and Roman religion. Source citations in ancient commentaries were not always transmitted with consistency. Proper names tended to disappear during medieval transmission, especially in the case of marginal scholia. There is evidence for this practice also in antiquity itself.14 Even though scholarly commentaries would generally preserve authors’ names, school commentaries often replaced them with pronouns, a phenomenon which is also found in the Greek tradition.15 Servius, and probably Donatus before him, greatly reduced the number of explicit references to his sources. Accordingly, the presence of Varro became progressively more anonymous from late antiquity onwards, such that even in modern editions many fragments which are not authenticated by Varro’s name are still attributed to him. Moreover, the name ‘Varro’ is potentially ambiguous when given as the author of poetic quotations, since it could refer either to the scholar, who also wrote poetry, or to his near contemporary, the ‘neoteric’ poet P. Terentius Varro Atacinus. Varro Atacinus is quoted by name six times in Servius and SD.16 The other 207 occurrences of Varro’s name refer to the scholar, and are distributed as follows:

12

For another example see Servius, A. 4.427 and 5.81.

13

Vallat 2016.

14

Goold 1970: 104, 110; Vallat 2012b: 250–51.

15

Dickey 2007: 111–12 for the use of indefinite pronouns.

See Lloyd 1961: 302 quoting Mountford–Schultz 1930; there are in fact more allusions than this, but without quotation. 16

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 94

14/11/2017 14:56:20

DANIEL VALLAT: VARRO IN VIRGILIAN COMMENTARIES Servius

84

Servius Danielis

99

Scholia Veronensia

5 (6)17

pseudo-Probus

7

Philargyrius

1

Breuis expositio G.

4 (7)

Scholia Bernensia

7 (16)

95

207

A striking preponderance of the citations occur in Servius and Servius Danielis. Since the text of Servius Danielis is only a fraction of the length of the Servian text (between five per cent and thirty per cent of the Aeneid commentary, depending on the book), the similar figures for the two texts may mislead: in fact, Varro is cited with much higher density in Servius Danielis, which apparently preserved a large number of quotations that Servius excluded. The same reasoning applies to the Scholia Veronensia, given their very fragmentary state.17 This suggests that Servius may be partly responsible for the disappearance of Varro’s name, as well as the names of other Republican authors, in the exegetical tradition. Lloyd convincingly demonstrated that Servius deleted many references to Republican authors in the final stage of a transformation that may have begun with Donatus and his efforts to edit scholarly exegesis for the perceived needs of a younger audience.18 The practice of anonymization contrasts sharply with Augustine’s ad hominem polemic in the City of God on Varro’s authoritative account of Roman religion.19 Taken as a whole, Varro’s name in Virgil commentaries is generally a reliable indication of authenticity: it does not seem that there is any pseudepigraphic attribution here, contrary to Greek traditions.20 Where comparison between quotations and Varro’s preserved works is possible, it emerges that the quotations are authentic, even if the detail of the text may sometimes vary (see below).21 3. Varro’s titles The titles of Varro’s works provide further evidence for the authenticity of the fragments, although they present specific problems. A title is given for a little under a third of the Varro quotations in Virgil commentaries (64/207), and only a small fraction of Varro’s works are represented, as shown in the following table, with titles by decreasing number of citations: The numbers in brackets give the total of occurrences in the work, but one should deduct quotations similar to what is found in Servius: in this case, except for the Scholia Veronensia, quotations probably come from Servius himself. Therefore, the first number indicates the number of original or independent occurrences.

17

18

See Thomas 1880: 182; Lloyd 1961: 296, 298, 323; Goold 1970: 135; Brugnoli 1988; Vallat 2016.

19

See Hadas in this volume.

20

Cf. Cameron 2004: 124–63.

There is one uncertain case: the Brevis expositio (2.168) attributes a note to Varro on the Ligures that Servius Danielis (A. 11.715) attributes to Nigidius Figulus. The Scholia Bernensia (G. 2.168) in Hagen’s edition, also refer to Nigidius. We may also have a rare case of contradiction, about the etymology of vates, between SD, A. 3.443 and Varro, Ling. 7.36. Either SD was wrong in attributing the fragment, or Varro allowed internal contradiction, which is not unusual.

21

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 95

14/11/2017 14:56:20

BICS-60-2  2017

96 Servius

SD

Ant. divinae

4

9

Logistorici

2

3

Ant. humanae

5

De lingua Latina

5

Gens P.R.

Phil.

BEG

Bern.

Probus

Veron.

14

4

1

10

2

1

1

5 3

Ora maritima

4

1

1

5 4

1

Vita P.R.

9 6

5

Saturae Res rusticae

Total

1

1

2

2

2

Aetia

2

2

De gradibus

1

1

De familiis Troianis

1

1

Epistolicae quaestiones

1

1

De saeculis (?)

1

1

De ludis theatralibus (?)

1

De scaenicis originibus (?) Titles/quotations

1 1

17/84

34/99

1 1/1

2/4

0/7

7/7

3/5

64/207

It should be noted that titles are not necessarily representative of the content of each quotation: for example, an extract from the De lingua Latina might refer not to a linguistic matter, but to a theological one (SD, G. 1.11). Titles pose various problems: they did not have fixed status in antiquity, and synonymy was common.22 If the Antiquitates divinae are usually designated by the same word divinarum, texts with multiple titles such as Logistorici were designated either by their generic name, or by their subject matter, or by the name of their dedicatee.23 A few titles remain uncertain. Some scholars include the De ludis theatralibus (Servius, A. 10.894) in the Antiquitates humanae;24 others (perhaps correctly) postulated that it was one of Varro’s books on theatre.25 Similarly, the De scaenicis originibus (SD, G. 1.19) is problematic: it is either a Logistoricus or an independent treatise. The question is complicated by the formulation de scaenicis originibus uel in Scauro, ‘in the Origins of the Theatre or in the Scaurus’ where it is not clear whether uel suggests an alternative title for the same work or the title of a different work.26

22

See, e.g. Schröder 1999.

For example: in logistoricis: Servius, A. 5.80; de pudicitia: SD, A. 4.45; in Logistorico, qui inscribitur Tubero de origine humana: Probus, Ecl. 6.31 (341 Hagen). In general, Servius seems to have used the generic title, SD the subject, and the other commentaries a more detailed convention of entitulature. According to Jerome, there were seventy-six books of Logistorici, which are not all identified; see Hendrickson 1911.

23

24

See Teuffel 1873: 244.

25

Brunetti (1874: 1425) classified it among the fragments from unknown books.

Riese (1865: 37, 256) emended uel to et in the belief that there were two works. Those who consider it a single work (a Logistoricus) include Chappuis (1868: 49), Brunetti (1874: 775, 902), and Lloyd (1961: 311). 26

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 96

14/11/2017 14:56:20

DANIEL VALLAT: VARRO IN VIRGILIAN COMMENTARIES

97

The referent of the expression de saeculis (SD, A. 8.526) is also uncertain. Is it a title (‘in the book On Centuries’) or an indication of subject matter (‘when talking about centuries’)? If the former, it is not clear whether the phrase refers to a separate treatise or to a book from the Antiquitates humanae, and the question has not really progressed since the nineteenth century.27 In A. 8.128, Servius attributes a passage to the Aetia of Varro (quod Varro in Aetiis ponit, ‘as Varro reports in the Origins’), while a manuscript of Servius Danielis (one of the three traditions of Servius’s text according to Murgia 1975) ascribes the same passage to the Antiquitates (quod Varro in antiquitatis libris ponit, ‘as Varro reports in his books on antiquity’). It is typical of the transmission of lesser-known texts that their titles are changed to titles still familiar to scribes or readers.28 In some cases the commentary gives the correct title but not the correct book number. For example, in his note on the balm melisphylla (G. 4.63), Servius refers to the first book of the Res rusticae: in fact, the reference occurs in the third book (R. 3.10.16). In A. 2.225, Thilo edited the text of Servius Danielis as it stands in the single manuscript: Varro autem rerum divinarum libro †XIX. delubrum esse dicit aut ubi plura numina sub uno tecto sunt, ut Capitolium, aut […], ‘Varro in book 19 of the Divine Antiquities says that the delubrum is where different gods are all under the same roof, as in the case of the Capitol, or […]’. The crux indicates a textual corruption in the view of the editor. Daniel had first read XXIX, and emended it to XIX, but the appendix of his edition of 1600 reads 27. Merkel proposed correcting the book number to VI, even though Macrobius quotes the same extract specifying that it comes from the eighth book, libro octauo.29 Thilo argued that what may have originally been written was libro XXXI antiquitatum qui est VI rerum divinarum, ‘in book 31 of the Antiquities, that is in book 6 of the Divine Antiquities’, but it is dubious whether so complex a formulation ever existed. In any case, it is clear that abbreviated numbers, like proper names, are sensitive to corruption in textual transmission.30 Attribution may be uncertain even in cases where all manuscripts are in agreement. Servius on A. 1.382 cites the second book of the Antiquitates divinae. But the fragment concerns Aeneas, and so to Krahner it seemed more likely that it belongs to the Antiquitates humanae. On such grounds Krahner proposed emending divinarum to humanarum — an emendation accepted by Mirsch,31 but not by Thilo or the Harvard editors, and rightly so: even if the attribution is wrong, Servius may already have found divinarum in his intermediate source — as illustrated below, he did not usually verify the original sources. These examples demonstrate some of the difficulties of attributing fragments to specific works of Varro. Even though Servius never quotes the Antiquitates by name, Mirsch credits him with transmission of twenty-three fragments from that work; some of these attributions are securely corroborated by correspondence with independent sources, but many are Riese (1865: 258), Chappuis (1868: 51), and Lloyd (1961: 311) considered it a Logistoricus; Brunetti 1874 was not consistent, attributing it to the Logistorici (col. 779) and to the eighteenth book of the Antiquitates humanae (col. 1285); Mirsch (1882: 129) also thought it was separate from the Antiquitates humanae. By contrast, other scholars writing at the same time tried to prove that de saeculis was the title of AH 15: Kettner 1868: 14–15; Gruppe 1876: 51–60. 27

28

See Kirchner 1910: 11–13.

29

Merkel 1841: 27, 123; Brunetti 1874: 1327–28; Boissier 1861: 224.

For example, see the textual variants in Servius’s note on the relative foundation dates of Rome and Carthage (A. 1.12).

30

31

Mirsch 1882: 89 (fr. 2.10).

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 97

14/11/2017 14:56:21

BICS-60-2  2017

98

questionable. A deduction is not a certainty.32 The material discussed here also shows the extent to which modern readers and editors of Varro remain dependent on nineteenthcentury scholarship.33 4. Forms of presence Varro is present in many forms in Virgil commentaries. The variety is partly due to the fact that in antiquity and the Middle Ages the commentary was not considered a literary work with fixed-text status, but rather a didactic tool that could be freely reworded, reworked, and abbreviated.34 Direct quotation of Varro in oratio recta would appear to be the most straightforward form, and might seem to guarantee authenticity, or at least a higher degree of accuracy than reported speech.35 Consider, for example, the following note on the spelling of (h)arena, ‘sand’: et Varro sic definit ‘si ab ariditate dicitur non habet, si ab haerendo […] habet’, ‘and Varro gives this definition: “if the word comes from ariditas, it is not written with an h-; if it comes from haereo, it gets one”’ (Servius, A. 1.172 = 280b Funaioli). Here, one would probably be right in thinking that Servius is repeating Varro’s own words. This is usually, but not always, the case. The appearance of authenticity may be deceptive, as the following example illustrates: Servius, A. 4.167: Varro dicit: ‘aqua et igni mariti uxores accipiebant’. Varro says: ‘the husbands used to receive their wives with water and fire’.

Varro, Ling. 5.61: igitur causa nascendi duplex: ignis et aqua. ideo ea nuptiis in limine adhibentur, quod coniungitur hic, et mas ignis, quod ibi semen, aqua femina, quod fetus ab eius humore. Therefore the conditions of procreation are two: fire and water. Thus they are used at the threshold in weddings, because there is union here, and fire is male, which the seed is in the other case, and the water is the female, because the embryo develops from her moisture. (tr. Kent)

One cannot establish with absolute certainty that the De lingua Latina was Servius’s source here; it may have been the Aetia or another work which treated the same subject in different words.36 Quotations introduced in oratio obliqua involve a higher degree of modification, resulting from syntactic integration.37

See Canetta 2016. Moreover, Varro was sometimes repetitive in his works: this makes untitled fragments more difficult to attribute. 32

33

See also de Melo in this volume.

34

See Murgia 1975: passim; Vallat 2016.

Lloyd 1961 organized his study of Varronian quotations on this basis. See also Uría Varela–Gutiérrez González 2011: 58–59. 35

See also Servius, A. 11.787; SD, A. 1.108; 1.112; 9.52 etc. Moreover, there are dubious cases of direct quotation: in Servius, A. 1.52, about Aeolus, some manuscripts read: sed, ut Varro dicit, rex fuit insularum, ‘but, as Varro says, he was king of the islands’ (paraphrase) and others read: sed Varro dicit ‘rex fuit insularum’, ‘but Varro says: “he was king of the islands”’ (direct quotation).

36

37

See Servius, A. 12.7 and Varro, Ling. 7.52.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 98

14/11/2017 14:56:21

DANIEL VALLAT: VARRO IN VIRGILIAN COMMENTARIES

99

At the other end of the spectrum lies the simple allusion: the reader is told that Varro treated such-and-such a topic, but not given Varro’s ipsissima verba. 38 Between quotation and allusion, there are a variety of forms, including paraphrase and summary. For example, Servius (G. 3.273) explains the episode of mares becoming pregnant by the winds, and says that Varro also recorded this phenomenon; but the words he attributes to Varro in oratio obliqua do not literally match the corresponding passage of Res rusticae (2.1.19). Servius adds elements (nimio ardore; frigidiores ventos; ad sedandum calorem) that are not in Varro, but that are in Virgil (G. 3.272–79). So Servius attributes to Varro what in fact is a paraphrase from Varro and Virgil. In SD, A. 3.366, the notice gives Varro’s etymologies on the Latin names of wonders. But the phrasing is less expansive than what is typically found in De lingua Latina, and so here it seems likely that Servius Danielis may have transmitted a summary, rather than the original text.39 And the secundum-formulation (secundum Varronem, passim) may introduce both verbatim and paraphrased quotations. In sum, no citation of Varro in Virgilian commentaries commands certainty about the exact phrasing of a fragment, since all scholarly material is subject to rewriting during the process of transmission. The phenomenon emerges with greater clarity when the same fragment is reported multiple times in Virgil commentaries. In general, the same formulation rarely appears twice. The second occurrence may summarize the first one,40 or present a similar but not exact repetition.41 So it would be wrong to think that we have Varro’s original words; rather the Virgil commentaries preserve testimonies that exhibit different degrees of reliability.42 5. Problems of textual transmission In addition to these problems of form, those who wish to establish the text of Varro must also contend with problems of manuscript transmission. One must distinguish between cases in which Varro’s text is extant and may be compared with that of the indirect transmission, and (far more often) cases in which the original is lost, and we are confronted with textual problems in trying to establish what Varro wrote and what the Virgilian commentator wrote. The first case is illustrated in the following table:

38

Cf. Servius, A. 9.600; SD, A. 1.122 sic et Varro; 8.600; G. 2.533, etc.

The situation is in fact more complex if one takes into consideration cross-references in Isidore and Paulus; see Vallat 2014: 156–57. 39

40

See Servius A. 3.444 and 6.74 (about the Sibylla); A. 5.409 and 6.304 (on the meaning of senior).

41

See, e.g. Servius, A. 5.80 and 11.97.

See also Servius, A. 6.36 and 6.72; Servius, A. 5.4 and SD, A. 4.682; Servius, A. 1.382 and SD, A. 2.801; SD, A. 1.378 and 3.148; SD, A. 4.59 and 4.166 (one or two fragments?); Servius, Ecl. 7.55 and Schol. Bern. G. 1.448. 42

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 99

14/11/2017 14:56:21

BICS-60-2  2017

100

SD, G. 1.11 (Thilo): Varro ad Ciceronem ita ait ‘ dii Latinorum, ita ut Faunus et Fauna sit. †per ex versibus quos vocant Saturnios, in silvestribus locis solitos fari futura atque inde faunos dictos’. ‘Varro says in his book dedicated to Cicero: “ are divinities of the Latins, of both sexes, so that there are Faunus and Fauna; the story has come down from the so-called Saturnian verses that they were accustomed in well-wooded spots fari ‘to speak’ those events that were to come, whence they were called Fauni.”

Ling. 7,36 (Kent): Fauni dei Latinorum, ita ut et Faunus et Fauna sit; hos versibus quos vocant Saturnios in silvestribus locis traditum est solitos fari quo fando faunos dictos. ‘Fauni “Fauns” are divinities of the Latins, of both sexes, so that there are both Faunus and Fauna; the story has come down that they, in the so-called Saturnian verses, were accustomed in well-wooded spots fari ‘to speak’ , from which speaking they were called Fauni.’ (tr. Kent)

In the standard editions used here, there are slight textual differences between the two versions (in bold): it is obvious that each of the two texts has been emended by comparison with the other.43 A feature of indirect transmission is that it may preserve a more accurate text than the direct tradition. In this case, there is every reason to believe that neither version transmits the original text of Varro, but that each has retained a part of it; it would, however, be methodologically unsound to emend either. In the second case, the corresponding work of Varro is no longer available, and modern editors depend entirely on the manuscript tradition of Virgil commentaries, a tradition which is not always consistent.44 For example, in Servius G. 3.446 (musmonem dicit ducem gregis, quem ita et Varro commemorat, ‘Virgil calls the leader of the herd musmo, and so too does Varro’), Thilo’s apparatus (musmonem AV: simonem H ē nomen M) points out that the original text (musmonem, corroborated by musmonum at Pliny HN 8.199) had been misunderstood and caused incorrect variants based on medieval reinterpretation (simonem may have been suggested by a Christianizing interpretation of ducem gregis). But even when manuscripts agree, the text can still be difficult to understand.45 Finally, sometimes fragments of Varro are indirectly transmitted by a single manuscript, and so there is no external control. This is generally true for Servius Danielis’s commentary on the first two books of the Aeneid. For example, the word liberatae at SD A. 1.448 (Varro. fr. 577.2 Bücheler) was emended by Thilo to libratae, while Riese preferred to restore liberatae.46 Each fragment of Varro in Virgilian commentaries has its own story. Each source, and sometimes each part of each source, has its own particular tradition, and so in this area we cannot rely on the analogical method alone. The editor must examine the specifics of each case and consider each source from the point of view of its intellectual context, its palaeographical and editorial traditions, and its reception in modern times. In this paper the main problems involved in extracting fragments from Virgil commentaries are addressed; but different representatives of Varro’s secondary tradition would call for different editorial criteria: for example, Nonius Marcellus’s aims were purely lexicographic; his manuscript tradition is independent of the Servian tradition; and his quotation of verse fragments has caused challenges for modern editors. Compare also SD, A. 1.43 and Varro, Ling. 7.23, where the last sentence is an editorial addition from the text of SD. Cf. also SD, A. 1.505 and Varro, Ling. 5.161. 43

44

See Murgia 1975.

45

See SD, A. 3.349; 8.330; G. 1.70; 4.265.

46

Riese 1865. See another similar problem, still unsolved in SD, A. 1.727.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 100

14/11/2017 14:56:21

DANIEL VALLAT: VARRO IN VIRGILIAN COMMENTARIES

101

6. A matter of boundaries Another important problem is the delimitation of fragments.47 While it is usually clear where quoted fragments begin, it can be difficult to tell where they end and the commentator resumes in his own voice. At times fragments of Varro are embedded in Virgilian exegesis in ways that involve excessive stratification. At Servius, A. 5.409, an Ovid quotation is inserted within Varro’s text. At A. 1.449, on the different kinds of doors, it is not clear where the Varro quotation ends: valvae autem sunt, ut dicit Varro, quae revolvuntur et se velant; ianua autem est primus domus ingressus, dicta quia Iano consecratum est omne principium. cetera intra ianuam ostia vocantur generaliter, sive valvae sint sive fores, quamuis usus ista corruperit. As for the valvae, as Varro says, they are doors which go back (revolvuntur) and hide themselves; as for ianua, it is the first entrance of a house, thus named from Janus, to whom all beginnings are dedicated. The other ones, inside the ianua, get the generic name ostia (‘opening’), whether they are valvae or fores — however, common usage has confused those terms. Thilo attributed to Varro only the first definition of valvae and ended the quotation after se velant. But one might argue that the quotation continued until principium. The third sentence does contain typically Servian wording (sive […] sive), but the opening words (cetera […] generaliter) could also be a summary of Varronian material. The difficulty is typical of fragmentary transmission, especially within texts which were not punctuated and which incorporated other texts.48 This leads into the problem of a series of citations on one matter, which I shall now discuss. When we encounter Varro’s name alongside other ancient authors in the Virgil commentaries,49 or alongside indefinite pronouns such as alii,50 we cannot tell whether the cluster of sources was assembled by the commentator, or whether Varro had already gathered them. This problem is typical of the commentum variorum, which obscures the chronology of scholarly stratification.51 7. What kind of knowledge? Did Servius have direct knowledge of Varro? There are clues that suggest he did not. Some of these have already been discussed above, such as uncertainty over titles. Then there is also the paradox, well known from J. Fontaine’s studies of Isidore, that the more accurate a reference is, the more likely it is to be a second- or third-hand reference.52 Other citation patterns point to indirect knowledge and late-stage integration of Varro: the interrupted quotation noted above (Servius, A. 5.409); in the Brevis expositio Georgicorum 1.1, two 47

For fragment delimitation in the grammarians, see Uría Varela–Gutiérrez González 2011.

48

See also Servius, A. 1.449; 2.81; 3.445; 4.167; Ecl. 7.21; G. 3.18; SD, A. 1.378; 2.512; 3.67; 8.363; Ecl. 8.12.

49

E.g. SD, A. 3.334.

50

E.g. SD, A. 3.12: id est Varro et alii complures, ‘that is Varro and many others’.

51

On horizontal and vertical stratification, see Béjuis-Vallat 2012: 308–10.

52

Fontaine 1959: 745.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 101

14/11/2017 14:56:21

BICS-60-2  2017

102

consecutive sentences from R. 1.29 are treated as if they were distinct fragments; and as noted above, Servius’s report of Varro at G. 3.273 is simply inaccurate. Beyond these points, serial citations on the same problem systematically obscure the ultimate source of the data. The first kind of serial citation, ascribing a point to Varro and indefinite others (alii), leaves it uncertain whether Varro was the first or last in a series to maintain the point.53 The deletion of the other authors’ names suggests that Varro may have been not only the most famous authority to have held the view, but the first. This may be inferred from other serial formulations which have not been truncated, and which allow us to trace the genealogy of the fragment, as in the following example: quam rem a Varrone tractatam confirmat et Plinius Pliny too confirms this meaning treated by Varro (Servius, A. 6.304) quod Varro et Suetonius commemorant as Varro and Suetonius record. (Servius, G. 3.24) In these examples it seems likely that Pliny and Suetonius quoted Varro and were the first steps in the transmission. This kind of series is usually in chronological order,54 and there is every reason to believe that most of the quotations from Varro were transmitted on the same principle, and that the name of the intermediate source was summarized by a pronoun such as multi, ceteri, alii...55 Moreover, we have one example (SD, G. 3.313) which explicitly reports indirect transmission through an intermediary: the reader is informed that Varro’s opinion is cited by Celsus (probably the first-century encyclopedist rather than the thirdcentury grammarian Arruntius Celsus).56 It seems likely that grammarians transmitting the Virgil commentaries did not verify their sources directly, and that they did not feel that they were expected to. In addition, we do not know how widely Varro’s works circulated in late antiquity, and even if some of them were extant, it is hard to imagine a grammaticus trawling through Varro’s monumental oeuvre to verify a three-word quotation. Perhaps the strongest argument in favour of the Virgil commentators’ indirect knowledge of Varro rests on the well-known working methods of the grammarians who transmitted these mostly variorum commentaries: they tended to copy (and sometimes modify) the most recent sources at their disposal.57

53

Cf. Servius, A. 6.733; 8.51; SD, A. 1.277; 1.415; 3.12; 3.85; 3.113 = 11.306; Ecl. 8.99.

54

See also: Servius, A. 5.45; 6.638; 7.563; 8.233; 9.600; SD, A. 8.600; 11.143.

Thilo 1881: xiv–xv showed that we must be cautious in the case of such plural pronouns, which can refer to a single scholar; see also Cameron 2004: 106. 55

56

See Zetzel 1981: 38.

Thomas 1880: 13, 204; Thilo 1881: xxi–xxii, lxxiv–lxxv; Halfpap-Klotz 1882: 44; Moeller 1892: 24; Kirchner 1910: 20; Lammert 1912: passim; Bährens 1917: 107; Lloyd 1961: 301, 315, 326; Goold 1970: 106; Brugnoli 1988: 809; Kaster 1988: 169; Maltby 2005, etc. 57

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 102

14/11/2017 14:56:21

DANIEL VALLAT: VARRO IN VIRGILIAN COMMENTARIES

103

8. Why quote Varro? The art of digression The presence of Varro in Virgil commentaries is the result of compilation through the centuries, the compound product of numerous scholars’ labours. The surviving texts present merely the final edifice without telling the history of its development. Quotation is a pervasive technique in the Virgilian commentaries, in line with Alexandrian exegetical practice (for example of the Greek epics). Quotations support and illustrate points, and furnish proof by example, as well as introducing students to the authoritative discourse, which is one of the foundations of the ars grammatica. The length, number, and shape of quotations all vary greatly. Sources may be either literary (auctores, often cited by name, for illustrative purposes) or scholarly (used primarily as supplements, and often difficult to identify). The grammarians’ canon of classical authors was revised in the fourth century and included, in particular, the poets and a few prose authors.58 The grammarians focused primarily on linguistic matters, and less often on literary ones.59 Varro is not a major figure in their canon. According to an ancient principle believed to have originated with Aristarchus, the author should be explained by reference to his own works (Ὅμηρον ἐξ Ὁμήρου σαφηνίζειν),60 and so Virgil is by far the most-cited author in Virgil commentaries. Explicit use of ‘scholarly’ sources is uneven and restricted in works intended for school use, such as Servius. By contrast, SD is more explicitly learned, while Macrobius remains the model for late antiquity. Varro is the only true scholarly source cited explicitly, because his name was prestigious and because he enjoyed a strong authority in conservative grammatical circles in three key areas: grammar, history, and religion.61 What was the purpose of quoting Varro in a Virgil commentary? Was Varro really useful for any interpretation of Virgil? To answer these questions, a glance at the distribution of Varro citations in a sample of commentary, on Aeneid 1, may be helpful. There are twelve occurrences in Servius62 and fifteen63 in Servius Danielis, distributed across the following subjects: Aeneid 1

Servius

Servius Danielis

Linguistics

5

6

Religion

2

5

Historia64

3

1

Mos antiquus

2

2

Geography

2

0

64

58

See for example Kaster 1978; Holtz 1981: 83.

59

Lazzarini 2013.

60

Nünlist 2015.

61

Lloyd 1961: 311.

62

Servius, A. 1.22; 52; 172; 246; 277; 382; 408; 449; 532; 648; 697; 740.

63

SD, A. 1.42; 43; 108; 112; 122; 182; 378 (bis); 415; 448; 505 (bis); 595; 649; 727.

64

Defined here as the ancient and sometimes legendary history of Rome, see Lazzarini 1984 and Dietz 1995.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 103

14/11/2017 14:56:21

BICS-60-2  2017

104

Some of the citations belong in two fields, such as Servius, A. 1.532 on the origin of Oenotria, which concerns both etymology and historia. There is not necessarily any correspondence between the work from which the fragment derives and the use to which it is put it in the commentary: for example, of the two extracts from De ora maritima (SD, A. 1.108 and 112), the first one is an antiquarian note on geography, while the second concerns semantics. We can extrapolate the same pattern for untitled fragments: a remark by Varro on a deity does not necessarily come from the Antiquitates divinae; a linguistic note does not necessarily come from De lingua Latina. The editor must be wary of overhasty attributions. In general, one may distinguish two levels of commentary for scholarly and school use respectively.65 Of course the boundary between the two is sometimes blurred, but longer notes are usually more erudite, and while linguistic scholia aim at a school audience, those on religion and ancient customs exhibit higher-level learning. One might divide them up as follows: Erudite scholia

School scholia

Servius

5

7

Servius Danielis

13

2

The notes of SD appear to be on the whole more learned, as one might have expected: SD is a commentum variorum, whereas Servius’s commentary is essentially a school text. This being the case, how many of these twenty-seven references to Varro really help us to understand Virgil better? In fact, very few: I found only four of them helpful to interpretation: -SD, A. 1.108: a scholarly note on geography to illustrate a learned allusion in Aen. 1.108–9; -Servius, A. 1.246: to explain a difficult passage in Virgil about the River Timavo and the word mare; -Servius, A. 1.382: to explain how Venus led Aeneas to Italy; -Servius, A. 1.532: on the origin of the word Oenotri. The rest offer no direct help in understanding Virgil, and so we might suggest that the majority of our Varro quotations in Aeneid 1 commentaries serve the art of digression. This in turn illustrates the special status of Virgil commentaries as distinct from other commentaries in ancient Latin pedagogy: they provided both an aid to understanding Virgil and a general foundation of universal knowledge, sometimes with more learned scholia. To be sure, the scholia always maintain a link with Virgil’s words, but it is sometimes a loose one based on freewheeling associations. Moreover, after four centuries of Virgilian exegesis, the scholia often reflect and respond to previous debates, as Georgii demonstrated.66 One dominant (and overused) interpretative expedient was to claim that Virgil always said and meant more than was actually in the text.67 On this basis, Virgil’s habit of alluding to arcane lore was used to justify grammarians’ digressions, even if the presumed allusion 65

Lloyd 1961: 311.

66

Georgii 1891; See also Vallat 2012b.

67

See Vallat 2013 on the per transitum technique. See e.g. Servius, A. 1.282.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 104

14/11/2017 14:56:21

DANIEL VALLAT: VARRO IN VIRGILIAN COMMENTARIES

105

was sometimes fanciful. For example, at A. 1.172 (egressi optata potiuntur Troes harena), Servius comments on the etymology of harena and quotes Varro, whose words are used to illustrate previous discussions on orthography, but not to explain Virgil’s verse. The same pattern appears in SD, A. 1.42, where the note explores the different kinds of fulmina and their links with deities: Virgil’s line had not posed any problem, but scholars still succeeded in creating one.68 In this particular case, the verb quaeritur indicates an earlier scholarly quaestio (i.e. Greek ζήτησις), thus suggesting a stage of interpretation before the extant redaction which involved quoting Varro. In sum, then, Varro is not normally cited for direct explanation of Virgil, but rather invoked to bolster the authority of the commentary, and to create a repository of information for students which is generally relevant to Virgil and his culture, though often quite indirectly, but sometimes in fact entirely marginal to Virgil. The use of Varro also exemplifies the digressive tendency of Virgilian commentators, as well as pointing to the broad readership at which these texts were aimed: since Virgil was a staple of education, Virgil commentaries were teaching aids designed to supply a broad foundation of knowledge. 9. Varro, Virgil, and the grammatici: authority and ideology in the fourth century In conclusion, I will focus on Servius’s commentary, the only one which we can date with some precision. We have examined above some of the ways in which Varro was used in Virgil commentaries. It now remains to investigate when Varro entered the tradition of Virgilian exegesis, and what his particular status was in the scholarly culture of the lateAntique grammaticus. There are three interwoven strands in my argument: one concerns the vicissitudes of the reception of Varro, a second the evolution of attitudes to Virgil, and the third is a tentative suggestion about ideological debates in the fourth century.69 From his own time, Varro was a widely exploited authority, and this is still the case in Virgil commentaries (e.g., in Servius A. 8.233 he is an authority on Latin grammar). But beyond this, what is his function in scholarly texts which do not, in fact, themselves call for primary scholarly enquiry? Although Servius, in contrast to SD, reduced the Varronian presence, he did not delete it entirely. The first reason for this is Varro’s intellectual prestige, which rests partly on his own antiquity:70 it can be proven that Servius generally represses the names of scholars later than Varro, such as Festus and Nonius; true enough, he cites Donatus, but mainly in order to disagree with him. But, as we have discussed, there is no reason to believe that Servius had access to full texts of Varro; rather his role was to select and copy from the previous centuries’ scholarship. So why does he retain Varro to the extent that he does? Varro is first and foremost a figurehead for an intellectual legacy. Although his work was integrated into the Virgilian exegetical tradition by many scholars at different times, I believe that we can date this integration later than the intermediary sources discussed above: Pliny the Elder and Suetonius, and even Gellius. This coincides with an upswing in attitudes to Virgil. Early Virgil criticism had been marked by detractors (obtrectatores), still in evidence at the time of Probus, who faulted various aspects of Virgil’s composition. By Servius’s time this kind of criticism was no longer practised. The divinus poeta was always 68

See Thomas 1880: 247–49 on this convoluted exegetical practice.

69

This section should be read closely with Hadas in this volume.

70

Vallat 2015.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 105

14/11/2017 14:56:21

BICS-60-2  2017

106

right and his words true, and it was the grammarian’s duty to explain why.71 This reversal in Virgil’s fortunes began in the second century and finally coincided with the acme of grammarians in the fourth century. Through their teaching, Virgil in turn became a figure of authority, not only in poetry, but in scholarship and in all fields of knowledge. Virgil competed with Varro’s authority, and exceeded it where there were points of conflict, even where language was concerned (cf. Servius, A. 8.233). Paradoxically, then, the competitive dynamic between Virgil and Varro itself bolsters Varro’s status, just as Virgil was becoming the foundation of Roman pedagogy and considered infallible. This may be illustrated by the example of Anchises’s death: Servius (A. 4.427) relates the different versions of Varro and Cato, both of which involve Anchises reaching Italy, which Virgil’s learned reader probably knew. But once the Virgilian vulgate, which has Anchises die in Sicily, prevailed in schools, it was worth quoting Varro or Cato as a reminder of the traditions that had been overshadowed by Virgil’s authority. Thus the rise of Varro’s presence in Virgil commentaries probably parallels Virgil’s reception as a scholar-poet in the middle of the second century ad. But, at the end of the fourth century, Varro’s legacy acquires a new and unexpected relevance in the ideological debates through the use made of him by Christian polemicists.72 Arnobius and Augustine, among others, used Varro as a target for their attacks on pagan religion, and the cults of the veteres. Servius had to reckon with this new situation, and he did so with the usual silence which he maintained on the religious question. Silence, however, does not mean indifference. Servius was contemporary with the various prohibitions of the pagan cults by Theodosius, with the Battle of the Frigidus, the controversies over the Altar of Victory, the sack of Rome by the Visigoths in 410, the polemics that followed, and even with Augustine’s anti-pagan use of Varro. And yet he made no reference to any of this, continuing to present Roman cults and gods as if nothing had happened, sometimes even defending the ancient gods against Virgil.73 This is not, I think, a mark of indifference or a refusal to take sides: pace Cameron,74 Servius does adopt a position, howbeit implicitly, in favour of the traditional religion, and his use of Varro is part of this attitude. Tacitly but firmly, Servius implies the authority of Varro, which puts the manipulations of a polemicist like Augustine into stark relief. The explicit presence of Varro allows Servius to remind his pupils that Varro’s works on religion were neither polemical nor descriptive of the religious landscape of the fourth century. Finally, Servius emphasizes that Varro was a valuable scholarly source in all areas of knowledge, not only in religion. In sum, through the commentaries, the pagan grammatici also validated the legacy of Varro — just as they did for the ancient disciplina Etrusca —75 and intended to rescue him from the appropriation of the Christian polemicists. Varro thus becomes a symbol of traditional Roman knowledge and also a cipher for cultural assertion and resistance. 71 Georgii 1891 for the fading of the anti-Virgil critics; Vallat 2015 for the critical reversal; cf. Augustine’s humorous remark at De utilitate credendi 6.13. 72

Cameron 2011: 614–21.

A. 1.4 is fundamental here: at the very beginning of his commentary, Servius, by a semantic change, exonerates Juno from the savagery (saeva) which Virgil attributed to her. 73

Cameron 2011: 207, 609, 621. The different papers in Garcea–Lhommé–Vallat 2016 also challenge this point of view. See also Pellizzari 2003: 73. 74

75

See Briquel 1997; Santini 2008.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 106

14/11/2017 14:56:21

DANIEL VALLAT: VARRO IN VIRGILIAN COMMENTARIES

107

One could not have guessed from Servius how Varro had been exploited in fourthcentury ideological polemics, but I think this is deliberate. Just as it could be proved that Servius was a pagan, while never in fact contributing to the religious debate, likewise his limited but varied use of Varro can be regarded as a silent but efficient transmission of ancient pagan knowledge. He seems to confer a timelessness on Varro’s knowledge, above and beyond trends or fashions. For grammarians, Varro is business as usual, but their use of him masks some deeper issues. To conclude: Varro’s presence in Virgil commentaries tells us much more about the commentators than about Virgil. Université Lumière, Lyon 2

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:53:18 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 107

14/11/2017 14:56:21

A TYPOLOGY OF ERRORS IN VARRO AND HIS EDITORS: A CLOSE LOOK AT SELECTED PASSAGES IN THE DE LINGUA LATINA1 WOLFGANG D. C. DE MELO Abstract: Varro’s De lingua Latina, our first grammatical treatise of any length written in Latin, is a problematic text: Varro’s linguistic theory and practice are often at odds with what later grammarians do, and the text has come down to us in a very poor state. This article examines how modern scholars have often approached the transmitted text with preconceived notions, and how this has influenced editorial choices and subsequent interpretations. The piece also looks at Varro’s own linguistic practices and predilections, and how he sometimes reaches conclusions that are logically inconsistent and indefensible. Keywords: Archaism, emendation, etymology, morphology, orthography, pronunciation, textual corruption

Varro’s De lingua Latina is a unique work written by a unique scholar. The Romans had been thinking about various elements of grammar for at least a century before Varro. Before Varro, Lucilius used to comment on points of grammar and spelling; Varro’s contemporary Caesar wrote two books on analogy in Latin while crossing the Alps with his army; and not much later, Verrius Flaccus compiled an enormous dictionary. However, it is Varro’s work that represents the first large-scale discussion of Latin, dealing with all aspects of the language. The De lingua Latina is full of intelligent insights as well as remarkable blunders, many of which cannot be excused by saying that Varro was a child of his time; had he written fewer books and spent more time on thinking things through and revising, a fair number of these errors could have been avoided. Much the same could be said about modern scholarship on Varro. Side by side with clever, thoughtful contributions, we find pieces that could have benefited from revision. In the former category there are such excellent articles as Gitner (2015); the second category will not be discussed here at all. In this chapter, I want to look at cases where good scholars, whose contributions to Varro are generally esteemed and recognized, were led astray by their own preconceived notions; I also wish to highlight cases where such scholars almost rewrote Varronian passages, but where such heavy interventions can be justified or at least accepted. Later on in my piece, I shall look at Varro’s own biases or prejudicial tendencies. I shall begin with cases of biased attitudes towards Varro’s language. This is followed by instances of intelligent and convincing rewriting of the Varronian text, and then instances of intelligent but less secure rewriting. 1. False assumptions in relation to Varro’s language 1.1. The spellings ae and e The text of the De lingua Latina has reached us in a very corrupt state. Of the original twentyfive books, we have only books 5–10 in direct transmission, giving us Varro’s practical 1

This work has been generously supported by the Leverhulme Trust.

BICS-60-2  2017

108

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 108

14/11/2017 14:56:21

WOLFGANG D. C. DE MELO: A TYPOLOGY OF ERRORS

109

etymologies (5–7) and his theoretical musings on morphology (8–10). The introductory book 1, the three books on the theory of etymology (2–4), the three books on the practical applications of morphology (11–13), and the final twelve books on syntax are lost. Only snippets transmitted by grammarians and men of letters survive.2 We are lucky to have books 5–10, found in the Codex Laurentianus LI. 10, folios 2–34, commonly abbreviated to F. This is not only our oldest manuscript, but also the one on which all others are based. Sadly, F is full of mistakes made by the person who produced this manuscript, but also by generations of his predecessors. While the text is clearly in need of often drastic emendation, sometimes we may wonder whether what seems to be corrupt is actually correct. One such case concerns spelling. Most of the spellings we find in F are medieval, but Kent (1938: i.xviii) argues that ‘Varro, as a countryman, may in some words have used E where residents of the city of Rome used AE’. As becomes clear from what follows, Kent is speaking about spelling as well as pronunciation, and his theoretical discussion has no bearing on his edition, where he mostly sticks to the standard orthography found in modern dictionaries. His assertion is based on the well-known fact that in pronunciation, areas outside Rome monophthongized the inherited ae to e earlier than Rome itself, where the spellings e for ae and hypercorrect ae for e begin to occur only in the early Empire.3 The question is not a purely theoretical one. We normally print the subordinator quom as cum when F transmits it as such, even though we know that Varro wrote quom.4 But we tend to correct transmitted e to ae where our dictionaries have the diphthong. Are we right in doing so? In 6.11, Varro derives saeclum ‘century’ from senex ‘old man’, a derivation which might make more sense if Varro actually said and wrote seclum. However, we need to be careful. In 7.96, Varro discusses words in which some people use ae and others e. Five pairs are listed: scaena / scena ‘stage’ (with obscaenum / obscenum ‘indecent’), scaeptrum / sceptrum ‘sceptre’, faeneratrix / feneratrix ‘female moneylender’, faenisicia / fenisicia ‘mown hay’, and Maesius / Mesius, a personal name. Only in the last pair does Varro ascribe the variant with e to rustics, and with good reason. The first two pairs are Greek loans with original η, so ae is clearly hypercorrect.5 The next two pairs are native words, but e is inherited, so the variants with ae are hypercorrections as well (see Walde and Hofmann 1965: i.479). Interestingly, Gellius (16.12.5–8) tells us that according to Varro, Cato wrote these words with e, and that Varro derived them from fetus ‘birth’.6 We must conclude that Varro was perfectly able to make a distinction between ae and e, at least in writing. The manuscript confusion between e and ae is medieval negligence: both were pronounced identically, but the spelling differed; the diphthong was written ę in Beneventan script, and the diacritic was often simply forgotten. Kent was one of the best editors of Varro, but in this instance he was led astray by his own learning; he knew about rural monophthongization as well as Varro’s provenance, and drew a prejudiced conclusion which looks logical only if we ignore the fact that Varro was a highly educated person.7 2

On the indirect transmission of Varro’s texts see esp. Hadas and Vallat in this volume.

For details see Adams 2007: 78–88. Varro himself points out in 5.97 that the word for ‘goat’ is pronounced hedus in the country, but haedus in Rome; unlike us, he believes that the urban speakers added a vowel.

3

4

F also transmits quom quite often.

There are instances where Varro does not recognize Greek loans, as in 5.120, where he derives magida ‘a type of vessel’ from magnitudo ‘large size’ rather than μαγίς, but here Greek origin must have been obvious to him.

5

6

On Gellius’s knowledge of Varro, see Marshall and Todisco in this volume.

Varro is interested in regional variation in the lexicon, but not normally in regional pronunciations. His comments on ae and e are made only because there is a clear sociolinguistic dimension to this variation.

7

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 109

14/11/2017 14:56:22

BICS-60-2  2017

110

1.2. Cum-clauses Another example of linguistic prejudice appears in 8.43: quare cum ignorarent quemadmodum similitudo debeat sumi, de analogia dicere non possunt. Therefore, since they do not know how similarity ought to be applied, they cannot speak about analogy. Ignorarent is clearly in the wrong tense; if we want to respect the sequence of tenses, we need a present. Rholandellus changed the form to ignorent, thereby abiding by the classical rules that state that cum-clauses which are to be interpreted as causal need to be in the subjunctive (Menge, Burkard, and Schauer 2000: 856–57). However, why should ignorent be corrupted to ignorarent? Goetz and Schoell emend the transmitted form to ignorant. The indicative is normal in early Latin cum-clauses, even if they are to be taken as causal.8 The usage persisted into the classical period in authors whose style is not as restrictive as that of Cicero and Caesar. But of course a medieval copyist would not be familiar with such niceties of usage, and would emend to a subjunctive; given the ignorance of our copyist and his predecessors, it should come as no surprise that the emendation was only partially successful, violating the sequence of tenses. Rholandellus was a sensible and sensitive editor, but again we can see editorial partiality at work: there is an unspoken assumption that Varro, being Cicero’s contemporary, should follow Ciceronian rules. In this case the ultra-conservative editors Goetz and Schoell prove to be superior. We can now turn to cases where editors have effectively rewritten the transmitted text, a procedure sometimes unavoidable with an author as poorly transmitted as Varro. 2. Intelligent rewriting of the transmitted text 2.1. Templum, tescum, and L. Spengel’s addition in 7.10 In 7.8, Varro informs us about the augural formula used on the citadel: [i]tem tesca[que] me ita sunto, quoad ego ea rite lingua nuncupauero. Textual notes: [i]tem tescaque Turnebus, item testaque F que del. Crawford ea rite L. Spengel, caste Scaliger, eas te F lingua p Victorius, linquam F Temples and wild lands shall be mine in this way, up to where I announce them with my tongue using a ritual formula. The restoration of the text is far from certain. Tesca, restored by Turnebus for transmitted testa, is unproblematic; not only do t and c look very similar in Beneventan minuscule, Varro See de Melo 2011: lxx for a Plautine example in the indicative quoted in the subjunctive by Cicero. For the syntax of cum-clauses in general see Hofmann and Szantyr (1977: 620 and 624–25).

8

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 110

14/11/2017 14:56:22

WOLFGANG D. C. DE MELO: A TYPOLOGY OF ERRORS

111

also discusses what tesca are over several chapters. Hesitantly, I follow Kent (1938.i: 274) in taking me as an old-fashioned spelling for the dative mi (see also ThLL V.2.255.8–16). In this part of the formula, the augur establishes a contrast between templum and tescum. Templum is the area for the observation of birds created by the augur’s words; tescum is the land outside.9 The transmitted text co-ordinates templa and tesca (or rather the nonsensical testa) with -que, which is the regular co-ordinator in very early texts. However, as Michael Crawford rightly pointed out to me in the discussion after the talk this paper is based on, we must delete -que and restore an asyndeton because of the following discussion: quod addit templa ut sit tesca, aiunt ‘sancta’ esse qui glossas scripserunt. id est falsum. nam curia Hostilia templum est et sanctum non est. sed hoc ut putarent aedem sacram esse templum esse, factum quod in urbe Roma pleraeque aedes sacrae sunt templa, eadem sancta, et quod loca quaedam agrestia, quae alicuius dei sunt, dicuntur tesca. Textual notes: sit Laetus tesca aiunt Turnebus, dextra aiunt F et sanctum templum add. L. Spengel quae L. Spengel, quod F dicuntur Bentinus, dicentur F As for the fact that he adds that temples must be tesca ‘wild lands’, those who have written glosses say that tesca means sancta ‘inviolable’. This is wrong. For the Hostilian curia is a temple and is not inviolable. But this belief that a temple is a consecrated building and that a temple is inviolable seems to have come about because in the city of Rome most consecrated buildings are temples, and the same are inviolable, and because certain wild places which belong to one god or another are called tesca. This chapter as presented here already contains the addition by L. Spengel that I wish to discuss. But first we need to look at some preliminaries. The subject of addit must be the augur from 7.8. In our passage, Varro claims that the augur says that templa must be tesca, a clear misunderstanding of the ancient text, but one which is only conceivable if tesca and templa are not co-ordinated; hence the deletion of -que in 7.8. Perhaps Varro got the idea that the two things must be identical from a misunderstanding of the function of the future imperative sunto in the passage: he interprets the text as prescribing that templa must be tesca, while in reality the prescription is an existential one, namely that there should be templa and tesca. Varro then rightly points out that tescum does not mean sanctum ‘inviolable’. It is interesting that he attributes this misunderstanding to ‘those who have written glosses’; this shows that lexicography was already a thriving field of research in his day.10

9

On the importance of these and other prescriptions in religious law, see MacRae in this volume.

10

On Varro’s research methods see also Piras, MacRae, and Todisco in this volume.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 111

14/11/2017 14:56:22

112

BICS-60-2  2017

In order to understand this chapter properly, we need to clarify four terms. Sanctus means ‘inviolable’, and more specifically ‘inviolable because of religious sanctions’. The second term, sacer, overlaps with sanctus to some extent.11 It means ‘sacred, consecrated to a deity’. Thus, our third term, aedis sacra, is a room or house consecrated to a deity; this corresponds to a shrine or a temple in its modern meaning, but not necessarily to a templum in the ancient sense. A templum is simply a stretch of land separated from another stretch of land. An augural templum can therefore just be a piece of uncultivated land, but a templum can also be a temple in the modern sense or, by metaphorical extension, refer to the heavens. Varro refutes the idea that tescum means sanctum by discussing the Curia Hostilia. This is an assembly hall established by Tullus Hostilius. As an assembly hall, it is separated from the surrounding land, and counts as a templum in the ancient sense; but being a building used for worldly discussions, it cannot be or become sancta. Varro’s explanation of how the misunderstanding of tescum as sanctum came about is equally persuasive. However, it only works if we accept L. Spengel’s addition of et sanctum templum. This addition is highly convincing for two reasons: first, without it Varro would merely be saying that it is a misunderstanding that an aedis sacra is a templum, but while not every templum is an aedis sacra, every aedis sacra is a templum,12 so this statement would be patently false. And second, Varro goes on to speak about three categories: aedis sacra, templum, and sanctus, so we need the three terms in the preceding discussion as well. Spengel may have rewritten the text here more than some editors are comfortable with, but I am happy to accept the addition because now everything falls into place; even if Spengel did not manage to reproduce Varro’s exact words, he must have come very close. The way Varro explains the genesis of the ancient misunderstanding is now simple. While these terms overlap only to a limited extent in principle, within Rome the overlap is near-complete. By definition, all aedes sacrae are templa; Varro says pleraeque ‘most’ because he is making a three-way comparison, and not all templa are sancta at the same time; for instance, an augural space is separated from the surrounding area, but is not inviolable by religious sanction. However, within Rome such augural spaces are atypical templa, as most templa there are temples in the modern sense. In the countryside, by contrast, augural spaces and sacred groves are much more common than big temple buildings. Therefore, in Rome the overlap between the three terms is such that most templa are sancta, and so the terms became associated. Outside the city, wild places, tesca, are often sacred to a deity. Because they are separate from the area around them, these tesca are also templa — a crucial difference from the augural formula used in the city in 7.8, where the augural space is the templum because it is separate from the rest, but the surrounding area is the tescum ‘wild land’, and this tescum is not sacred to any deity. Now a tescum outside the city may be a templum and sacrum to a deity but not sanctum; however, because tescum is an obsolete word, because tesca in the country are templa, and because templa in the city are sancta, city dwellers came to believe that tescum is no more than a synonym of sanctum. I hope I am not stretching the evidence Because of this overlap, sacer has a suppletive comparative sanctior, from the paradigm of sanctus; see 4.2 below.

11

12 Actually, things are slightly more complicated. Gell. 14.7.7 tells us that according to Varro, not all aedes sacrae are templa, which at first sight contradicts my statement. However, in this Gellius passage, Varro means that not all aedes sacrae are templa in the sense of being augural spaces. This is quite different from the passage under discussion, where a templum is simply a piece of land that is cut off, whether for augural purposes or for other religious or civic reasons. Every aedis sacra is a templum in this second broader sense.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 112

14/11/2017 14:56:22

WOLFGANG D. C. DE MELO: A TYPOLOGY OF ERRORS

113

if I state that Varro, who is proud of his origins in the Sabine countryside, is trying to correct a corrupt city usage by looking at the uncorrupted earlier usage preserved elsewhere. 2.2. What does a praefica do? Varro quotes a large amount of secondary sources, especially in the etymological part of his work, and here in particular in book 7. An important question is how accurately Varro reports such texts. For many quotations, Varro is our only source, which makes it hard to answer this question. But Varro also quotes a large amount from the canonical plays of Plautus, and here we are able to compare this indirect transmission with a direct tradition. Sometimes there are discrepancies, and in such cases the Varronian text is by no means consistently superior (Deufert 2002: 139–50). For instance, in 7.50 Varro quotes Plaut. Amph. 275 in order to illustrate the Latin name for the constellation Orion. The direct transmission has the form iugulae, which is regular and scans perfectly. Varro has the rarer singular iugula both in the quotation and in his discussion, which makes it unlikely that this is a mere copying mistake; this form can be made to scan, but the result is clumsy: we have to assume a locus Jacobsohnianus (breuis in longo) in the third element, or assume a violation of Ritschl’s law. Regardless of scansion, the meaning of the variants is identical, but the directly transmitted text is metrically superior. Elsewhere, we may have to admit early corruption in the Plautine text because both the direct transmission and Varro contain oddities. For instance, Leo deletes Merc. 619–24 as an interpolation. Leo generally deletes too much, but in this case the deletion is convincing, as Merc. 625 fits perfectly with Merc. 618. The only problem is that in 7.60, Varro quotes Merc. 619. The interpolation must have been so early that it could make it into a version of the text that predates Varro. What I want to look at here, however, is 7.70, because it showcases the more unusual situation where the citation Plaut. Truc. 495 is correct, while the explanation has to be corrected: In Truculento: ‘sine uirtute argutum ciuem mihi habeam pro prefica.’ dicta, ut Aurelius scribit, mulier ab luco quae conduceretur quae ante domum mortui laudis eius caneret. hoc factitatum Aristoteles scribit in libro qui scribitur Νόμιμα βαρβαρικά, quibus testimonium est quod Freto est Neuii: ‘haec quidem hercle, opinor, praefica est; nam mortuum collaudat.’ Claudius scribit: ‘quae praeficeretur ancillis, quemadmodum lamentarentur, praefica est dicta.’ utrumque ostendit a praefectione praeficam dictam. Textual notes: praefica add. B Aldus scribitur B Augustinus Νόμιμα βαρβαρικά Turnebus, nomina barbarica F Freto est Goetz et Schoell, Freto est Canal, fretum est F In the Truculentus: ‘I’d consider an eloquent citizen without bravery to be like a praefica “woman leading the mourning”.’

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 113

14/11/2017 14:56:22

BICS-60-2  2017

114

Praefica ‘woman leading the mourning’, as Aurelius writes, is what a woman from the grove was called who was hired to sing the praises of a deceased man in front of his house. Aristotle, in his book entitled Foreign Customs, writes that this was done regularly; what is in The Strait of Naevius is a testimony to these customs: ‘She is, I think, a praefica “woman leading the mourning”; for she is extolling a dead man.’ Claudius writes: ‘A woman who praeficeretur “was to be put in charge” of the slave-girls, as to how they should lament, was called a praefica “woman leading the mourning”.’ Both passages show that the praefica was called thus from her praefectio ‘appointment as leader’. What we have here is the text with some minor corrections. It is syntactically correct and at least at first sight makes reasonably good sense. However, on closer inspection things do not add up. Five authorities are cited: Plautus, Aurelius, Aristotle, Naevius, and Claudius. At the end of the chapter, we learn that two of these sources derive praefica from praeficere ‘to put in charge’ or praefectio ‘appointment as leader’; incidentally, this is also the modern etymology of the word. Of our five authorities, Plautus and Naevius do not present us with any derivation, and Aristotle merely speaks of the frequency of the custom. The etymology accepted by Varro must thus have been advanced by Claudius and Aurelius. In the transmitted text, Claudius does indeed propose the desired etymology, but Aurelius does not. Luckily, A. Spengel found a solution to the problem that is as easy as it is convincing. Passive -tur is normally abbreviated -t~, and combinations of vowel plus nasal are often written with a diacritic as well, for instance -ẽ- for -en- or -em-. Such diacritics are commonly left out or wrongly inserted. Spengel corrected conduceretur to conduceret, and caneret to canerent. What Aurelius says can now be translated as follows: ‘Praefica is what a woman from the grove was called who hired women to sing the praises of a deceased man in front of his house.’ The praefica undoubtedly sang the praises of the deceased man as well (see Non. 145.25 M and Paul. Fest. p. 223), but this is not what she got her name from; she got her title because she was in charge of the other singers. Two changes of diacritics may already count as rewriting, but without these two changes we would have a text in which Varro is unable to count to two, as he would state that there are two sources for the same etymology, but would proceed to give us only one, with Aurelius having a different opinion altogether. 2.3. How many conjugation classes does Varro accept? The system of conjugations resembles that of declensions insofar as we today generally speak of four conjugations and five declensions, even though in practice we recognize five and six, respectively. Thus, we consider agere ‘to act’ and facere ‘to do’ to belong to the third conjugation, even though they are diachronically and synchronically distinct, as can be seen from the first persons agō versus faciō. Similarly, we treat dux ‘leader’ and turris ‘tower’ as third-declension nouns, even though the former continues a consonant stem and has duce as its ablative, while the latter continues an i-stem and has turrī as its ablative. The system of declensions is discussed in 10.62: Sin ab singulari quis potius proficisci uolet, initium facere oportebit ab sexto casu, qui est proprius Latinus. Nam eius casuis litterarum discriminibus facilius reliquorum

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 114

14/11/2017 14:56:22

WOLFGANG D. C. DE MELO: A TYPOLOGY OF ERRORS

115

uarietate discernere poterit, quod ei habent exitus aut in A, ut hac terra, aut in E, ut hac lance, aut in I, ut hac febri, aut in O, ut hoc caelo, aut in V, ut hoc uersu. Textual notes: initium Groth, id illum Goetz et Schoell, inillum F casuis A. Spengel, casus his F uarietate Augustinus febri Goetz et Schoell in adnotationibus, leui F But if someone wishes to begin from the singular instead, he ought to make a start from the sixth case, which is the case peculiar to Latin. For through the differences in the letters of this case he will be able to distinguish the variation of the remaining ones more easily, because the ablatives have their endings in A, as ablative terra ‘earth’, or in E, as ablative lance ‘platter’, or in I, as ablative febri ‘fever’, or in O, as ablative caelo ‘sky’, or in U, as ablative uersu ‘verse’. The text of this chapter is relatively secure. Even if we do not accept Goetz and Schoell’s febri, some other i-stem noun must be concealed behind leui. What is clear from this is that Varro uses the ablative singular to distinguish declension classes, but without taking quantities into account. That is why he has two separate declensions for lanx and febris where we today speak of the third declension, while he lacks our fifth declension, which ends in -ē (diē) rather than -ĕ (lancĕ). This is particularly interesting because Varro does seem to take vowel length into account when he discusses conjugation classes in 9.109: utrum in secunda forma uerborum temporalium habeat in extrema syllaba -as an -ĭs at -īs ad discernendas similitudines interest. quocirca ibi potius index analogiae quam in prima, quod ibi abstrusa est dissimilitudo, ut apparet in his meo, neo, ruo. ab his enim dissimilia fiunt transitu, quod sic dicuntur meo, meas, neo, nes, ruo, ruis, quorum unumquodque suam conseruat similitudinis formam. Textual notes: persona add. L. Spengel temporalium L. Spengel, temporale F an -es add. L. Spengel an -ĭs at -īs L. Spengel,13 anis at si F Whether in the second person the verb paradigm has in its last syllable -as or -es or -ĭs or -īs is relevant for distinguishing similarities. Thus it is there rather than in the first person that an indication of analogy lies, because in the first person dissimilarity lies hidden, as is clear in these: meo ‘I go’, neo ‘I sew’, and ruo ‘I fall’. For from these, dissimilar forms arise in the transition, because one speaks in this way, meo, meas, neo, nes, and ruo, ruis, each of which keeps its own paradigm of similarity. This is how L. Spengel presents the chapter; according to him, Varro recognizes our first, second, third, and fourth conjugations, but does not draw a distinction within the third conjugation between two subclasses. This makes sense because Varro is talking about the second person singular, where the two subclasses of the third conjugation look identical. Taylor 1977: 130 argues against L. Spengel’s conjecture, but the use of diacritics in Varro was already proved by Roth 1861. 13

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 115

14/11/2017 14:56:22

BICS-60-2  2017

116

But does the chapter in this form reflect Varro’s ideas or Spengel’s? In the first sentence of my quotation, the second-conjugation ending is added by Spengel and not found in the manuscript. Similarly, the garbled anis at si that we find in F need not reflect two separate endings. Does Spengel make Varro say things that he did not dream of saying? Despite the problematic transmission of this chapter, I am inclined to accept Spengel’s text, for three reasons. First, Spengel is not trying to foist modern ideas onto Varro, as can be seen from the fact that the third conjugation is not divided into two subtypes by adding more material. Second, the addition of the second-conjugation ending is sensible in view of the rest of the chapter, where Varro mentions neo, nes; he can hardly mention this conjugation here without having been aware of it just a few words earlier. And third, the division of the nonsensical anis at si into two conjugations is sensible if we accept that Varro considered quantities linguistically significant and that he marked them occasionally. This is not obvious from the discussion of declension classes above, but it is beyond doubt because of the discussions in 9.104 or 10.25 (see also Roth 1861). We can now turn to some instances of rewriting where I have greater doubts. 3. More problematic cases of rewriting 3.1. The Sabines and the Ides Varro often mentions the Sabines, no doubt because he himself was born in Sabine territory and had some local pride. But the linguistic identity of these Sabines is somewhat nebulous. What did they speak? Latin with some pre-Latin elements? A variety of Oscan? Or a variety of Umbrian? The question becomes relevant in 6.28, where he discusses the origin of the Ides: idus ab eo quod Tusci itus, uel potius quod Sabini idus dicunt. Textual note: idus (Fv), Eidus Giardina, Edus Flobert The Idus ‘Ides’ got their name from the fact that the Etruscans call them Itus, or rather because the Sabines call them Idus. This is the text as transmitted, and it must be corrupt. By the first century bc, Romans would pronounce the word as īdūs, but in inscriptions and on coins the original diphthong was still written, so we often see eidus. We shall leave aside the question of how Varro would have spelled the word, and also the question of what the Etruscans would have said; at any rate the spelling t must be correct, since Etruscan did not normally use voiced stops outside alphabet tablets. But what did the Sabines say? They can only have said īdūs like their Roman contemporaries if they essentially spoke Latin, albeit with a few idiosyncrasies in other words. But in that case there would be no point in citing their form because to Varro it would be a Latin one. So we must look at Oscan and Umbrian. In Oscan, the ablative plural eídúís is attested,14 so Giardina restores eidus for Varro. In Umbrian, such an original diphthong would be monophthongized to ē, hence Flobert’s edus. Which option we choose ultimately depends on what we believe Varro’s Sabine to have been, and thus to some extent on our own prejudices. At NH 29.7.14, Pliny reproduces a letter by Cato in which he is annoyed that the Greeks think of Romans as Oscans. Do we believe that Varro also looked 14

At Cp 31 and at several other places; see Untermann 2000: 203–04.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 116

14/11/2017 14:56:22

WOLFGANG D. C. DE MELO: A TYPOLOGY OF ERRORS

117

down on the Oscans? Or was he more neutral? It is hard to say. I prefer Flobert’s solution for purely geographical reasons; Oscan was the language of the Italian south, while Umbrian was spoken in closer proximity to Rome. 3.2. Illicium and illex 6.86–95 constitutes a single long passage in which Varro quotes various ancient documents. He is illustrating what to his mind is the word inlĭcium or illĭcium ‘enticement / invitation’, while we nowadays follow Mommsen and interpret this as a phrase in līcium ‘into the circle’; without this modern interpretation, the phrase in licium uocare makes no syntactic sense. From 6.95 we learn that Varro’s misinterpretation is in fact not his own, but can already be found in the legal commentaries of Marcus Junius. The chapter concludes as follows in F: quod tamen ibidem est quod illicite illexit quae cum E et C cum G magnam habet comunitatem. The last sentence is clearly transmitted poorly. Varro points out the kinship between C and G, which we would nowadays describe in terms of identical place and manner of articulation; the difference is that the former is voiceless, while the latter is voiced, but neither the Greeks nor the Romans knew how voicing is produced. Varro also points out the kinship of something with e. This something must be a vowel, and it could be a (parco and peperci) or I (artifex and artificis); o is also conceivable, but less likely (tego and toga). As the word under discussion is illicium, and as it is closely related to illecebra, it is probably best to assume, with Ciacconius (as quoted in Augustinus’s edition), that the letters in question are i and e and that the letter which went missing is i. Since we are still on the subject of inlicium, and since Varro explicitly mentions inlex, the cleverest way to handle this kinship of letters is the way Goetz and Schoell did: quod tamen , ibidem est quod illicit[e] illex, it quod cum E et C cum G magnam habet comunitatem. Textual notes: inlex apud Plautum in Persa est qui legi non paret add. Goetz et Schoell in apparatu illicit[e] illex it quod Goetz et Schoell (quod iam Mueller), illicite illexit quae F I add. Ciacconius apud Augustinum Yet as to the fact that inlex in Plautus in the Persa is someone who legi non paret ‘does not obey the law’, and that in the same work illex is what illicit ‘entices’, it happens because I has a great kinship with E and C with G. Inlex / illex is ambiguous if we ignore vowel quantities. It could refer to someone who entices, in which case we are dealing with illĕx, a root noun in -s from lacere with the preverb in- ‘into’; or it could refer to someone who breaks the law, in which case we are dealing with illēx, from the same root we find in lēx ‘law’, but with in- being the negative prefix. Before -s, voiced stops devoice. This does not affect illĕx, where the root already ends in -k-, but it does affect illēx, which has an underlying -g-, as can be seen from the genitive illēgis. The kinship of E and I is visible in illĕx and its genitive illĭcis; original *-a- weakened to -e- in medial closed syllable, but to -i- in medial open syllable.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 117

14/11/2017 14:56:22

BICS-60-2  2017

118

Goetz and Schoell’s restoration, relegated to their apparatus, states that Plautus uses both words in the Persa. Of course we cannot know whether Varro really wrote this, but it is very clever; illĭce, the ablative of illĕx, occurs in Persa 597, and the vocative illēx is found in Persa 408. 3.3. The etymology of suffibulum In 6.21, Varro discusses the feast of Ops Consiva, held on 25 August. That day, the priest has to wear a kind of ceremonial headdress called a suffibulum. Varro discusses its etymology, but the transmitted text makes little sense: id diciturne ab suffiendo subligaculum? Is this called subligaculum ‘tie below’ from suffire ‘to fumigate’? This cannot stand for various reasons. First, Varro does sometimes express doubts about etymologies, but he does not present them in the form of a question.15 Second, the word we need is suffibulum; subligaculum, though well formed, is attested only one other time (Cic. off. 1.129). And third, Varro would not explain subligaculum by something that sounds as different from it as suffire. Diciturne needs to be changed. By contrast, subligaculum, being morphologically well formed, is bound to be correct; we should not change it to subfigabulum with Kent (who also takes Skutsch’s suffiendo on board), even though such a change is palaeographically acceptable and the result is also morphologically well formed. However, subligaculum needs an explanation. Flobert resolves the issue as follows: id dicitur non ab suffiendo, subligaculum. Textual notes: dicitur non Flobert (duce Mueller), diciturne (Fv) sed ab suffigendo quod subtus ligetur ut add. Flobert This is said not from suffire ‘to fumigate’, but from suffigere ‘to fasten down’, because it subtus ligatur ‘is tied below’, as if a subligaculum ‘tie below’. If we accept Flobert’s corrections, Varro derives suffībulum from suffīgere ‘to fasten down’, which is correct by modern standards. Both words have the root fīg- ‘to fasten’, which goes back to *dheigw- ‘to fix’, with a cognate in Lithuanian díegti ‘to plant, sting’; de Vaan (2008: 219) suggests that Tocharian B tsākaṃ ‘he may bite’ could also be related, but expresses justified doubt. The regular outcome should be fīuere, and such a form is attested in Paul the Deacon excerpting Festus (p. 81 L.); -g- was reintroduced from the perfect, but suffībulum still shows the regular reflex of the labiovelar. Now we have a text that does not only make sense, but even provides us with an etymology that is correct from our modern perspective. The question that remains unanswered is, however, to what extent this text reflects Varro and to what extent it reflects Flobert, an intelligent editor who would not want to ascribe untenable etymologies to our ancient scholar. If I go with Flobert in my forthcoming edition, I will do so more out of respect for the great scholar than out of confidence that he is right.

15

On Varro’s awareness of the limits of etymology see Piras in this volume.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 118

14/11/2017 14:56:22

WOLFGANG D. C. DE MELO: A TYPOLOGY OF ERRORS

119

We have seen how editors’ unconscious biases or assumptions can lead them astray, and how they sometimes even tend to rewrite the Latin text; however, the latter occasionally leads to remarkably good results. Let us now turn to Varro’s own biases and preconceived assumptions. 4. Varro’s own biased predilections 4.1. Etymology Taylor asserts that Varro ‘generally behaves as scientifically as possible for a first-century etymologist’ (2015: 21). This kindly view ignores the fact that Varro has his own biases, biases which he does not always manage to overcome. Having two alternative explanations for one word is in principle unproblematic. This is in fact common practice amongst ancient grammarians and shows either twofold connections exhibited by a single word, or sensible doubt about a proposal; in the latter case, even modern etymologists suggest different possibilities. However, on occasion double or triple etymologies are problematic, and not just from a modern perspective. A case in point is found in 7.52, where Varro quotes from the Cornicula, a play ascribed to Plautus: bc

In Cornicula: ‘qui regi latrocinatu’s decem annos Demetrio.’ Latrones dicti ab latere, qui circum latera erant regi atque ad latera habebant ferrum, quos postea a stipatione stipatores appellarunt, et qui conducebantur; ea enim merces Graece dicitur λάτρον. ab eo ueteres poetae nonnunquam milites appellant latrones. quod item ut milites cum ferro, aut quod latent ad insidias faciendas. Textual notes: stipatores uolgo, stipateres F λάτρον Victorius, CATPON F at nunc uiarum obsessores dicuntur latrones add. Kent (ex Paul. Fest. p. 105 L.) sunt add. Goetz et Schoell (ex Seru. Dan. Aen. 12.7) In The Little Crow: ‘You who latrocinatu’s “have served” King Demetrius “as a mercenary” for ten years.’ Latrones ‘mercenaries’ were called thus from latus ‘side’, as the ones who were on either side of the king and had a sword at their sides — later they called them stipatores ‘bodyguards’ from stipatio ‘close attendance’ — and who were hired; for this pay is called λάτρον ‘soldier’s pay’ in Greek. From this the old poets sometimes call soldiers latrones. Yet now those who infest roads are called latrones ‘highwaymen’, because like soldiers they are with a sword, or because they latent ‘lie hidden’ in order to make an ambush. In this paragraph, Varro presents us with the etymology of latro, which we have to extract from latrocinari ‘to serve as a mercenary’. The text is not unproblematic. I accept Kent’s addition, which is based on Festus, who may have copied from Varro. The addition is sensible because Varro is clearly opposing an older usage, in which latro meant ‘soldier, mercenary’, with a more recent one, in which the word has come to mean ‘highwayman’;

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 119

14/11/2017 14:56:22

BICS-60-2  2017

120

the earlier usage is indicated by the use of the past tense and the phrase ueteres poetae ‘the old poets’, whose principal representatives in Varro’s work are Plautus and Ennius. The semantic development is interesting because it shows how soldiers and their behaviour were regarded by common people. What matters for our purposes is that two etymologies are given for the old usage and one for the new one. The ancient latro is derived from latus ‘side’, but also from Greek λάτρον ‘soldier’s pay’; the latter derivation is what we accept nowadays. The new meaning is then derived from latere ‘to lie hidden’, as if we were dealing with an unrelated homophone rather than the same word with a new semantic development. But Varro clearly knows that we are dealing with the same word, otherwise he would have indicated that there are two separate words, as he does in 10.7, where he distinguishes between suis, the genitive of sus ‘pig’, and suis, the second person of suere ‘to sew’.16 Varro also knows that the meaning ‘highwayman’ is more recent. So why a separate derivation? The only reason that makes sense is that, like many modern etymologists, he cannot handle keeping a form or meaning unexplained, and this drives him to come up with an etymology even where it is not required and is, in fact, against common sense.17 4.2. Morphology Books 8–10 are dedicated to morphology. Nowadays, it is generally accepted that inflectional morphology, consisting of noun and verb endings, is by and large regular, otherwise we could not learn languages. Derivational morphology, creating compounds or turning verbs into nouns and vice versa, also follows certain principles, but is not as regular. Varro follows the dialectic principle. He argues against regularity or analogy in morphology in book 8, he speaks for regularity in book 9, and he adopts a moderate stance in book 10, where he states that inflection is by and large regular, while derivation is less so. Book 8, in which Varro argues against regularity in morphology, is an oddity: Varro does not really identify with this approach, as is obvious from the fact that the next two books present morphology, and here especially inflectional morphology, as regular. What is more, even within book 8 a lengthy introduction (8.1–23) is dedicated to Varro’s true views, which favour analogy without excluding instances of irregularity entirely, especially within derivational morphology, where it has its place. That said, the points made in book 8 are not intended to be ridiculous. The arguments presented may have deliberate weaknesses, but at least the data on which they are founded are to be taken seriously. In 8.50–51, Varro discusses pronouns and the forms that ought to exist if analogy were a true guiding principle in language. He ends the discussion with a word of caution: de hoc genere parcius tetigi, quod librarios haec spinosiora indiligentius elaturos putaui.

A similar case occurs in 5.179. We today consider munus ‘present’ derived from munus ‘duty’, via ‘present given out of a sense of duty’, but for Varro these are separate words derived differently. 16

One could argue that Varro’s derivation of latro from latere comes from some other source and that Varro feels compelled to represent it even though he knows that ‘highwayman’ is just a later meaning of the same word. However, this seems unlikely to me because Varro regularly introduces others’ etymologies by giving their names (as in 5.101), or by at least referring to alii ‘others’ (as in 5.34) or to quidam ‘certain people’ (as in 5.53).

17

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 120

14/11/2017 14:56:22

WOLFGANG D. C. DE MELO: A TYPOLOGY OF ERRORS

121

Textual notes: spinosiora Victorius,18 sponsiora F indiligentius Scioppius, indulgentius F This topic I have touched on more sparingly because I believed that the copyists would transmit these thornier forms too carelessly.19 Some forms of particular interest occur in 8.50: primum si esset analogia [ut] in infeineiteis articulis, ut est quis, quem, quoius, sic diceretur qua[e], quam, quaius. [...] et ut est quem, quis, sic quos, ques. quare quod nunc dicitur qui homines, dici oportuit ques. Textual notes: ut del. L. Spengel quoius H Augustinus, cuius F qua[e] Laetus quam quaius Augustinus, quamuis F First, if there were analogy in the indefinite articles, such as is quis, quem, quoius ‘who, whom, whose’ (masc.), one would in this way say qua, quam, quaius ‘who, whom, whose’ (fem.). [...] And as there is quem, quis ‘whom, who’ (masc. sg.), so there ought to be quos, ques ‘whom, who’ (masc. pl.). Thus what is now said qui homines ‘which men’, ought to be said ques. The masculine singular paradigm of ‘indefinite articles’, our pronouns, is treated as the basis for analogy, and feminine singular and masculine plural forms are derived from it. As is Varro’s usual practice, the forms that are presented last should exist but do not. Thus, in the feminine singular paradigm one ought to say quaius, and in the masculine plural paradigm one ought to say ques, but the forms do not exist. Or do they? Quaius is of course restored by editors; Varro was right in saying that copyists were bound to mistreat such forms. However, the restoration is sensible, and the form is indeed non-existent. Ques, on the other hand, is actually reasonably well attested, as in the indefinite si ques ‘if any people’ (CIL I2.581.3 and 24). The question then arises whether ques, archaic in Varro’s time, was not known to him, or whether he somehow forgot about it in the heat of the argument. Since ques is not infrequent and occurs in both inscriptions and high literature (Pacuv. trag. 221), Varro is bound to have come across it. However, Varro would not compile his linguistic works with a dictionary or a handbook of diachronic morphology on his desk, and so he is unlikely to have treated this form wrongly on purpose. It is more probable that his own linguistic intuitions trumped his learning. Native-speaker intuitions are of course a good thing, but they should never replace corpus work because they can easily be distorted by our own biases and assumptions.20 Another example, which I shall merely mention in passing, occurs in 8.77. We are presented with three adjectives: macer ‘lean’, sacer ‘sacred’, and tener ‘tender’. These Goetz and Schoell’s sponsiora has the downside of having an -n- before -s- that is probably post-Varronian, but the upside of being closer to what we find in F and showing how the corruption to sponsiora came about. 18

19

On the oral dimension of scholarly research, see Marshall in this volume.

20

For a good discussion see Labov 2008: 300–23.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 121

14/11/2017 14:56:23

BICS-60-2  2017

122

are then said to differ in their comparatives, in that the first two have trisyllabic forms like macrior, while the last form has a tetrasyllabic comparative tenerior. Our text leaves out sacrior, but the context demands such a form, and Dahlmann was right to restore it. However, while sacrior is morphologically well formed, and while Varro vouches for its existence, it does not occur in any transmitted text until the Carolingian period, being replaced by sanctior. In the process of forming paradigms, Varro’s intuitions became distorted.21 5. Conclusions Varro’s De lingua Latina is an inherently problematic text. We have looked at various cases of bias or prejudiced assumptions: bias towards Varro’s own linguistic usages, cases of rewriting the transmitted text, and Varro’s personal linguistic biases and predilections. Much more could be said on each of these topics. Ultimately, we have to accept that this work is so poorly transmitted that any emendation is bound to be influenced by our own prejudices; we can try to minimize this influence, but we cannot eliminate it entirely. At least we are not the only ones who are biased: Varro was by no means a neutral scholar himself. Wolfson College, Oxford

It is in principle possible that sacer is no more than the corruption of another adjective that had a trisyllabic genitive, but there are no other signs of corruption, and Varro does occasionally create otherwise unattested forms, such as the accusative spontem ‘free will’ in 6.72; the noun is defective and only forms the genitive and the ablative in Varro’s time. 21

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:55:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 122

14/11/2017 14:56:23

BIBLIOGRAPHY Adams, J. N. 2007: The Regional Diversification of Latin 200 BC–AD 600, Cambridge. Ando, C. 2010: ‘The ontology of religious institutions’, HR 50, 54–79. Arena, V. 2012: Libertas and the Practice of Politics in the Late Roman Republic, Cambridge. ——— 2014: ‘Informal norms, values, and social control in the Roman participatory context’, in D. Hammer (ed.), A Companion to Greek Democracy and the Roman Republic, Chicester, 217–38. Aricò Anselmi, G. 2012: Antiche regole procedurali e nuove prospettive per la storia dei comizi, Turin. Assenmaker, P. 2007: ‘“Pinguis salutis atque imperii”: l’enjeu du Palladium dans les luttes politiques de la fin de la Republique’, Les études classiques 75, 381–412. Astarita, M. L. 1993: La cultura nelle ‘Noctes Atticae’, Catania. Ax, W. 2000 (= 2001): ‘Dikaiarchs Bios Hellados und Varros De vita populi Romani’, RhM 143, 337–69. Repr. in Fortenbaugh and Schütrumpf (eds) 2001, 279–310. Badian, E. 1996: ‘Princeps senatus’, in OCD (3rd edn), Oxford and New York. Bährens, W. A. 1917: Studia Serviana ad litteras Graecas atque Latinas pertinentia, Ghent. Baier, T. 2007: ‘Dicearco e il De vita populi Romani di Varrone: il ruolo di utilitas e di voluptas’, Aevum Antiquum N. S. 7, 259–68. Baldwin, B. 1975: Studies in Aulus Gellius, Lawrence, KS. Bardenhewer, O. 1913: Geschichte der altkirchlichen Literatur, vol. I (2nd edn). Barnes, T. D. 1985: Tertullian: A Historical and Literary Study (2nd edn), Oxford. Barwick, K. 1957: ‘Widmung und Entstehungsgeschichte von Varros De lingua Latina’, Philologus 101, 298–304. Baschera, C. (ed.) 1999: Gli scolii veronesi a Virgilio, Verona. Bauman, R. A. 1969: The duumviri in the Roman criminal law and in the Horatius legend, Historia Einzelschr. 12, Wiesbaden. ——— 1983: Lawyers in Roman Republican Politics: A Study of the Roman Jurists in their Political Setting, 316–82 BC, Munich. ——— 1985: Lawyers in Roman Transitional Politics, Munich. ——— 1989: Lawyers and Politics in the Early Roman Empire, Munich. Bäumerich, H. J. 1964: Über die Bedeutung der Genealogie in der römischen Literatur, Diss., Cologne. Beard, M. 1990: ‘Priesthood in the Roman republic’, in M. Beard and J. North (eds), Pagan Priests: Religion and Power in the Ancient World, London, 19–48. Beard, M., J. North, and S. Price 1998: Religions of Rome, 2 vols, Cambridge.

BICS-60-2  2017

123 © 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms

BICS 60-2 Nov14-2 FINAL-FINAL.indd 123

14/11/2017 14:56:23

124

BICS-60-2  2017

Béjuis-Vallat, M. 2012: ‘Servius Danielis et la “double ascendance” d’Énée et des Troyens’, Eruditio Antiqua 4, 289–311. Bettini, M. 2011: The Ears of Hermes: Communication, Images, and Identity in the Classical World, trans. by W. M. Short, Columbus, OH. Billows, R. A. 1982: ‘The last of the Scipios’, AJAH 7, 53–68. Bischoff, B., and V. Brown 1985: ‘Addenda to Codices Latini Antiquiores’, Medieval Studies 47, 317–66. Blank, D. 2008: ‘Varro and the epistemological status of etymology’, Histoire, épistémologie, langage 30, 49–73. ——— 2011: ‘“Plutarch” and the sophistry of “noble lineage”’, in J. Martínez (ed.), Fakes and Forgers of Classical Literature = Falsificaciones y falsarios de la literatura clásica, Madrid, 33–59. ——— 2012: ‘Varro and Antiochus’, in D. Sedley (ed.), The Philosophy of Antiochus, Cambridge, 250–89. Blösel, W. 2000: ‘Die Geschichte des Begriffes Mos Maiorum von den Anfängen bis zu Cicero’, in B. Linke and M. Stemmler (eds), Mos Maiorum: Untersuchungen zu den Formen der Identitätsstiftung und Stabilisierung in der römischen Republik, Stuttgart, 25–97. Bodei Giglioni, G. 1986: ‘Dicearco e la riflessione sul passato’, RSI 98, 629–54. Boissier, G. 1861: Étude sur la vie et les ouvrages de M. T. Varron, Paris. Bona, F. 2003: ‘Ius pontificium e ius civile nell’esperienza giuridica tardo-repubblicana: un problema aperto’, in Lectio Sua: Studi editi e inediti di diritto romano, vol. II, Padua, 965–1013. Bonnefond-Coudry, M. 1989: Le Sénat de la République romaine de la guerre d’Hannibal à Auguste: pratiques délibératives et prise de décision, Rome. ——— 1993: ‘Le princeps senatus: vie et mort d’une institution républicaine’, MEFRA. Antiquité 105, 103–34. ——— 2007 (= 2014): Loi Iulia sur les séances du Sénat, in Lepor. Leges Populi Romani, sous la dir. de J.-L. Ferrary et de Ph. Moreau, Paris. http://www.cn-telma.fr/lepor/ notice466/ (Date de mise à jour: 28/11/14). Borges, J. L. 1999: Selected Non-Fictions, trans. by E. Weinberger, New York. Boscherini, S. 1985: ‘A proposito della tradizione del “pro Nobilitate” pseudo-plutarcheo’, in R. Cardini et al (eds), Tradizione classica e letteratura umanistica: Per Alessandro Perosa, Rome, 651–60. Bourdieu, P. 1977: Outline of a Theory of Practice, trans. by R. Nice, Cambridge. Bowie, E. L. 2013: ‘Libraries for the Caesars’, in König, Oikonomopoulou, and Woolf (eds), 237–60. Boys-Stones, G. R. 2001: Post-Hellenistic Philosophy: A Study of its Development from the Stoics to Origen, Oxford. Bremer, F. P. (ed.) 1896: Iurisprudentiae antehadrianae quae supersunt, vol. I, Leipzig. ——— (ed.) 1898: Iurisprudentiae antehadrianae quae supersunt, vol. II.1, Leipzig. Bretone, M. 1971: Tecniche e ideologie dei giuristi romani, Naples.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 124

14/11/2017 14:56:23

BIBLIOGRAPHY

125

Brincken, A.-D. von den 1957: Studien zur lateinischen Weltchronistik bis in das Zeitalter Ottos von Freising, Düsseldorf. Bringmann K., 2002: ‘Von der res publica amissa zur res publica restitute. Zu zwei Schlaggworten aus der Zeit zwischen Republik und Monarchie’, in Spielvogel J. (hrsg.), Res publica reperta. Zur Verfassung und Gesellshaft der römischen Republik und des frühen Prinzipats. Festschrift für Jochen Bleicken zum 75. Stuttgart, 113123. Briquel, D. 1997: Chrétiens et haruspices: la religion étrusque, dernier rempart du paganisme romain, Paris. Briscoe, J. 2013: ‘L. Cassius Hemina’, in Cornell et al. (eds), vol. I, 219–23, vol. II, 244–70, vol. III, 160–84. Briscoe, J., and A. Drummond 2013: ‘Cornelius Nepos’, in Cornell et al. (eds), vol. I, 395– 401, vol. II, 806–11, vol. III, 497–504. Broughton, T. R. S. 1952: The Magistrates of the Roman Republic, vol. II, New York. Bruck, E. 1945: ‘Cicero vs. the Scaevolas: law of inheritance and decay of Roman religion (De legibus II, 19–21)’, Seminar 3, 1–20. Brugnoli, G. 1988: ‘Servio’, in Enciclopedia Virgiliana, vol. IV, Rome, 805–13. Brunetti, F. (ed.) 1874: ‘I Frammenti’, in P. Canal and F. Brunetti (eds), M. Terenzio Varrone libri intorno alla Lingua Latina; I Frammenti, Venice, cols. 617-1482. Bryen, A. Z. 2014: ‘Law in many pieces’, CPh 109, 346–65. Buckley, M. 2002: ‘Atticus, man of letters, revisited’, in K. C. Sidwell (ed.), Pleiades Setting: Essays for Pat Cronin on his 65th Birthday, Cork, 14–32. Bücheler, F. (ed.) 1963: Petronii saturae. Adiectae sunt Varronis et Senecae saturae et similes reliquiae (8th edn), Berlin. Butterfield, D. J. 2014: Oxford Bibliographies. Marcus Terentius Varro. ——— (ed.) 2015: Varro Varius: The Polymath of the Roman World, Cambridge. Byrne, A. 1920: Titus Pomponius Atticus: Chapters of a Biography, Bryn Mawr, PA. Cadoux, T. J. 2008: ‘The Roman carcer and its adjuncts’, G&R 55, 202–21. Cairns, J. W., and P. J. du Plessis 2007: ‘Introduction: themes and literature’, in Cairns and du Plessis (eds), 3–8. Cairns, J. W., and P. J. du Plessis (eds) 2007a: Beyond Dogmatics: Law and Society in the Roman World, Edinburgh. Cameron, A. 2004: Greek Mythography in the Roman World, Oxford. ——— 2011: The Last Pagans of Rome, Oxford. Campbell, G. 2003: Lucretius on Creation and Evolution: A Commentary on De Rerum Natura, Book Five, Lines 772–1104, Oxford. Canetta, I. 2016: ‘Una fonte per il commentario di Servio a Virgilio: le Res Humanae di Varrone’, in A. Garcea, M.-K. Lhommé, and D. Vallat (eds), Fragments d’érudition: Servius et le savoir antique, Hildesheim, 17–29. Carcopino, J. 1932: ‘Sur la loi romaine du monument de Paul-Émile à Delphes’, in Mélanges Gustave Glotz, vol. I, Paris, 117–32.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 125

14/11/2017 14:56:23

126

BICS-60-2  2017

Cardauns, B. 1960: (ed.) Varros Logistoricus über die Götterverehrung (Curio de cultu deorum), Würzburg. ——— (ed.) 1976: M. Terentius Varro: Antiquitates Rerum Divinarum, Wiesbaden. ——— 1982: Stand und Aufgaben der Varroforschung (mit einer Bibliographie der Jahre, 1935–1980), Wiesbaden. ——— 2001: Marcus Terentius Varro: Einführung in sein Werk, Heidelberg. Castner, C. J. 1988: Prosopography of Roman Epicureans: From the Second Century bc to the Second Century ad, Frankfurt am Main. Cèbe, J.-P. (ed.) 1972–99: Varron, Satires Ménippées: édition, traduction, et commentaire, 13 vols, Rome. ——— (ed.) 1994: Varron, Satires Ménippées: édition, traduction, et commentaire, vol. X, Pappus aut indigena – Pransus paratus, Rome. Cenderelli, A. 1973: Varroniana: Istituti e terminologia giuridica nelle opere di M. Terenzio Varrone, Milan. ——— 1976a: ‘Interesse delle opere di Varrone come fonte di notizie per lo studio storico del diritto romano’, in Atti del congresso internazionale di studi varroniani, vol. II, Rieti, 321–26. ——— 1976b: ‘Varrone giureconsulto? (Riflessioni sulla figura del giureconsulto romano)’, in Seminario romanistico gardesano, 19–21 Maggio 1976, Milan, 33–41. Champeaux, J. (ed.) 2007: Arnobe: Contre les gentils, vol. III, Paris. Chappuis, Ch. (ed.) 1868: Fragments des ouvrages de M. Terentius Varron intitulés Logistorici, Hebdomades uel de imaginibus, De forma philosophia, Paris. Chastagnol, A. 1992: Le Sénat romain à l’époque impériale: Recherches sur la composition de l’Assemblée et le statut de ses membres, Paris. Chelotti, M., and M. Silvestrini (eds) 2016: Epigrafia e territorio, politica e società: Temi di antichità romane, vol. X, Bari. Cichorius, C. 1922: Römische Studien: historisches, epigraphisches, literargeschichtliches aus vier Jahrhunderten Roms, Leipzig. Clarke, G. W. (ed.) 1974: The Octavius of Marcus Minucius Felix, New York. Cloud, J. D. 1977: ‘Livy’s source for the trial of Horatius’, LCM 2, 205–13. Cohee, L. L. J. 1994: Responsa and Decreta of Roman Priesthoods during the Republic, PhD Diss., University of Colorado. Cole, T. 1967: Democritus and the Sources of Greek Anthropology, Chapel Hill. ——— 2004: ‘Ovid, Varro, and Castor of Rhodes: the chronological architecture of the Metamorphoses’, HSPh 102, 355–422. Collart, J. P. (ed.) 1954: Varron, De lingua Latina livre V: texte établi, traduit, et annoté, Paris. Cornell, T. J., and K. Lomas (eds) 1995: Urban Society in Roman Italy, London. Cornell, T. J., et al. (eds) 2013: The Fragments of the Roman Historians, 3 vols, Oxford. Crawford, M. H. 1996: Roman Statutes, vol. II, London. Croke, B. 2007: ‘Late antique historiography, 250–650 ce’, in J. Marincola (ed.), A Companion to Greek and Roman Historiography, Oxford, 567–81.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 126

14/11/2017 14:56:23

BIBLIOGRAPHY

127

Cugusi, P. 1967: ‘Le epistole di Varrone’, RCCM 9, 78–85. Dahlmann, H. 1935: ‘M. Terentius Varro’, in RE Suppl. 6, 1172–277. ——— (ed.) 1940: Varro, De lingua Latina Buch VIII, Berlin (repr. Berlin-Zürich-Dublin 1966; Hildesheim 2003). ——— 1963: Studien zu Varro, De poetis, Wiesbaden. Daintree, D., and M. Geymonat 1988: ‘Scholia non Serviana’, in Enciclopedia Virgiliana, vol. IV, Rome, 706–20. Daly, Ll. W. 1943: ‘The entitulature of pre-Ciceronian writings’, in Classical Studies in Honor of William Abbott OIdfather, Urbana, 20–38. D’Anna, G. 1973–4: ‘Alcune osservazioni sulle fonti di Gellio, N.A. XVII, 21 e sulla cronologia geronimiana dei poeti latini arcaici’, ArchClass 25–26, 166–237. ——— 1975: ‘Alcune considerazioni sulla fortuna del “Liber Annalis” di Attico: Attico fonte di Gellio?’, Studi Urbinati B 49, 331–47. Della Corte, F. 1970: Varrone: il terzo gran lume romano (2nd edn), Florence. ——— 1976: ‘L’idea della preistoria in Varrone’, in Atti del congresso internazionale di studi varroniani, Rieti, 111–36 (= Opuscula, Genova, VI (1978) 231–56). De Martino, F. 1974: Storia della costituzione romana, vol. IV, Naples. de Melo, W. (ed.) 2011: Plautus, vol. I: Amphitryon, The Comedy of Asses, The Pot of Gold, The Two Bacchises, The Captives, Cambridge, MA and London. Deschamps, L. 1987: ‘Temps et histoire chez Varron’, in Filologia e forme letterarie: Studi offerti a Francesco Della Corte, Urbino, 167–92. ——— 1995: ‘Trois nouveaux fragments des Antiquités Humaines de Varron?’, BStudLat 25, 16–21. Deufert, M. 2002: Textgeschichte und Rezeption der plautinischen Komödien im Altertum, Berlin. de Vaan, M. A. C. 2008: Etymological Dictionary of Latin and the Other Italic Languages, Leiden and Boston. Dickey, E. 2007: Ancient Greek Scholarship, Oxford. Dietz, D. B. 1995: ‘Historia in the commentary of Servius’, TAPhA 125, 61–97. Dingmann, M. 2007: Pompeius Magnus: Machtgrundlagen eines spätrepublikanischen Politikers, Rahden. Dix, T. K. 2013: ‘“Beware of promising your library to anyone”: assembling a private library at Rome’, in König, Oikonomopoulou, and Woolf (eds), 209–34. Dodds, E. R. 1973: The Ancient Concept of Progress, Oxford. Drummond, A. 1989: ‘Rome in the fifth century II: the citizen community’, in F. W. Walbank et al. (eds), The Cambridge Ancient History, vol. VII.2, Cambridge, 172–242. ——— 2013a: ‘T. Pomponius Atticus’, in Cornell et al. (eds), vol. I, 344–53, vol. II, 718– 25, vol. III, 457–62. ——— 2013b: ‘M. Terentius Varro’, in Cornell et al. (eds), vol. I, 412–23, vol. II, 836–43, vol. III, 513–17. du Plessis, P. J. 2013: ‘Introduction’, in P. J. du Plessis (ed.), New Frontiers: Law and Society in the Roman World, Edinburgh, 1–5.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 127

14/11/2017 14:56:23

128

BICS-60-2  2017

Edelstein, L. 1967: The Idea of Progress in Classical Antiquity, Baltimore, MD. Fantham, E. 2001: ‘The synchronistic chapter of Gellius (N.A. 17.21) and some aspects of Roman chronology and cultural history between 60 and 50 bce’, in Roman Readings: Roman Response to Greek Literature from Plautus to Statius and Quintilian, Berlin, 343–56. Reprint of LCM 6 (1981), 7–17. ——— 2003: ‘Three wise men and the end of the Roman republic’, in F. Cairns and E. Fantham (eds), Caesar Against Liberty? Perspectives on his Autocracy, Cambridge, 96–118. Feeney, D. 2007: Caesar’s Calendar: Ancient Time and the Beginnings of History, BerkeleyLos Angeles-London. Feger, R. 1956: ‘T. Pomponius Atticus’, in RE Suppl. 8, 503–26. Finley, M. 1973: The Ancient Economy, London. Fish, J. 2011: ‘Not all politicians are Sisyphus: what Roman Epicureans were taught about politics’, in J. Fish and K. R. Sanders (eds), Epicurus and the Epicurean Tradition, Cambridge, 72–104. Flach, D. (ed.) 1996: Marcus Terentius Varro, Gespräche über die Landwirtschaft: Buch 1, Darmstadt. ——— (ed.) 1997: Marcus Terentius Varro, Gespräche über die Landwirtschaft: Buch 2, Darmstadt. ——— (ed.) 2002: Marcus Terentius Varro, Gespräche über die Landwirtschaft: Buch 3, Darmstadt. ——— (ed.) 2006: Marcus Terentius Varro, Über die Landwirtschaft, Darmstadt. Flower, H. 1996: Ancestor Masks and Aristocratic Power in Roman Culture, Oxford. Fontaine, J. 1959: Isidore de Séville et la culture classique dans l’Espagne wisigothique, Paris. Fontanella, F. 2012: Politica e diritto naturale nel De Legibus di Cicerone, Rome. Forsythe, G. 1999: Livy and Early Rome: A Study in Historical Method and Judgement, Stuttgart. Fortenbaugh, W. W., and E. Schütrumpf (eds) 2001: Dicaearchus of Messana: Text, Translation, and Discussion, New Brunswick, NJ and London. Foucault, M. 1970: The Order of Things: An Archaeology of the Human Sciences, English translation, New York. Fraccaro, P. (ed.) 1907: Studi Varroniani: De gente populi Romani libri IV, Padua. Franceschi, F. 1954: ‘Censorino e Varrone’, Aevum 28, 393–418. Frier, B. W. 1985: The Rise of the Roman Jurists: Studies in Cicero’s Pro Caecina, Princeton, NJ. ——— 2000: ‘Roman law’s descent into history’, JRA 13, 446–48. Fuhrmann, M. 1987: ‘Erneuerung als Wiederherstellung des Alten: zur Funktion antiquarischer Forschung im spätrepublikanischen Rom’, in R. Herzog and R. Koselleck (eds), Epochenschwelle und Epochenbewusstein, Munich, 131–51.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 128

14/11/2017 14:56:23

BIBLIOGRAPHY

129

Gabba, E. 1979: ‘Sulle strutture agrarie dell’Italia romana fra III e I sec. a.C.’, in M. Pasquinucci and E. Gabba (eds), Strutture agrarie e allevamento transumante nell’Italia Romana (III–I sec. a.C.), Pisa, 15–73. ——— 1988: ‘La pastorizia nell’età tardo-imperiale in Italia’, in C. R. Whittaker (ed.), Pastoral Economies in Classical Antiquity, Cambridge, 134–42. ——— 1991: Dionysius and the History of Archaic Rome, Berkeley. ——— 1996: Dionigi e la storia di Roma arcaica, Bari. Garcea, A. 2008: ‘Consule ueritatem: Cicéron, Varron, et un chapitre de l’histoire de la vérité à Rome’, Revue de métaphysique et de morale 57, 93–110. Garcea, A., M. K. Lhommé, and D. Vallat (eds) 2016: Fragments d’érudition: Servius et le savoir antique, Hildesheim. Georgii, H. 1891: Die antike Aeneiskritik, Stuttgart. Gioseffi, M. 1991: Studi sul commento a Virgilio dello Pseudo-Probo, Florence. Giovannini, A. 1993: ‘Il passaggio dalle istituzioni monarchiche alle istituzioni repubblicane’, in Bilancio critico su Roma arcaica fra monarchia e repubblica in memoria di Ferdinando Castagnoli (Roma, 3–4 giugno 1991), Rome, 75–96. Gitner, A. 2015: ‘Varro Aeolicus: Latin’s affiliation with Greek’, in Butterfield (ed.), 33–50. Goldsworthy, A. K. 2006: Caesar: Life of a Colossus, New Haven, CT. Goold, G. P. 1970: ‘Servius and the Helen episode’, HSPh 74, 101–68. Goetz, G., and F. Schoell (eds) 1910: M. Terenti Varronis De lingua Latina quae supersunt. Accedunt grammaticorum Varronis librorum fragmenta, Leipzig. Goulet R. 1994: Dictionnaire des philosophes antiques, tome II, Babélicos d’Argos à Dyscolius, Paris. Grafton, A. T., and N. M. Swerdlow 1985: ‘Technical chronology and astrological history in Varro, Censorinus, and others’, CQ 35, 454–65. ——— 1986: ‘The horoscope of the foundation of Rome’, CPh 81, 148–53. Green, C. M. C. 1997: ‘Free as a bird: Varro De re rustica 3’, AJP 118, 427–48. Gruen, E. S. 1974: The Last Generation of the Roman Republic, Berkeley-Los AngelesLondon. Gruppe, O. 1876: ‘Über die Bücher XIIII bis XVIIII der Antiquitates humanae des Varro’, Hermes 10, 51–60. Guiraud, C. (ed.) 1985: Varron, Économie rurale: Livre II, Paris. Hadot, I. 2005: Arts libéraux et philosophie dans la pensée antique: Contribution à l’histoire de l’éducation et de la culture dans l’antiquité (2nd edn), Paris. Hagen, H. (ed.) 1867: Scholia Bernensia ad Vergilii Bucolica atque Georgica, Leipzig. ——— (ed.) 1902: Appendix Serviana, Leipzig. Hagendahl, H. 1967: Augustine and the Latin Classics, Gothenburg. Halfpap-Klotz, R. 1882: Quaestiones Servianae, Greifswald. Harries, J. 2006: Cicero and the Jurists: From Citizen’s Law to the Lawful State, London. Harrison, S. J. 1989: ‘Augustus, the poets and the spolia opima’, CQ 39, 408–14.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 129

14/11/2017 14:56:23

130

BICS-60-2  2017

Healey, J. 1909: The City of God […] a Translation into English by John Healey First Published in 1610, vol. I, Edinburgh. Hendrickson, G. L. 1911: ‘The provenance of Jerome’s catalogue of Varro’s works’, CPh 6, 334–43. Hendrickson, G. L., and H. M. Hubbell 1971: Cicero: Brutus, Orator (revised edn), London and Cambridge, MA. Heurgon, J. (ed.) 1978: Varron, Économie rurale: Livre I, Paris. Hoces Sánchez, M. C. 2012: ‘Mito y religión en el ‘de die Natali’ de Censorino’, in M. N. Muñoz Martín and J. A. Sánchez Marín (eds), Homenaje a la Profesora María Luisa Picklesimer (In memoriam), Coimbra, 143–62. Hofmann, J. B., and A. Szantyr 1977: Lateinische Syntax und Stilistik, Munich. Holford-Strevens, L. 2005: Aulus Gellius: An Antonine Scholar and his Achievement (2nd edn), Oxford. ——— 2015: ‘Varro in Gellius and late antiquity’, in Butterfield (ed.), 143–60. Holtz, L. (ed.) 1981: Donat et la tradition de l’enseignement grammatical: Étude sur l’ars Donati et sa diffusion (IVe–IXe siècle) et édition critique, Paris. Hooper, W. D., and H. B. Ash (eds) 1934: Marcus Porcius Cato, On agriculture; Marcus Terentius Varro, On agriculture, with an English translation, Cambridge, MA and London. Horsfall, N. 1972: ‘Varro and Caesar: three chronological problems’, BICS 19, 120–8. ——— 1981: ‘Some problems of titulature in Roman literary history’, BICS 28, 103–14. Hunt, T. J. 1998: A Textual History of Cicero’s Academici Libri, Leiden. Jocelyn, H. D. 1982: ‘Varro’s Antiquitates rerum diuinarum and religious affairs in the late Roman republic’, BRL 65, 148–205. Johnson, M. 2007: The pontifical law of the Roman republic, PhD Diss., Rutgers, NJ. ——— 2015: ‘The pontifical law and the civil law: towards an understanding of the ius pontificium’, Athenaeum 103, 140–56. Jones, J. H. 1939: ‘Πεπλογραφία Varronis tibi probari non moleste fero’, CPh 34, 148–49. Kantor, G. 2012: ‘Ideas of law in Hellenistic and Roman practice’, in P. Dresch and H. Skoda (eds), Legalism: Anthropology and History, Oxford, 55–83. Kaster, R. A. 1978: ‘Servius and Idonei Auctores’, AJP 99, 181–209. ——— 1988: The Guardians of Language: The Grammarian and Society in Late Antiquity, Berkeley. Keil, H. (ed.) 1889: M. Terenti Varronis rerum rusticarum libri tres, Leipzig. Keil, H., and G. Goetz (eds) 1929: M. Terenti Varronis rerum rusticarum libri tres, Leipzig. Kent, R. G. (ed.) 1938: Varro, On the Latin Language, with an English translation, 2 vols, Cambridge, MA and London. Kettner, H. 1868: Kritische Bemerkungen zu Varro und lateinischen Glossaren, Roßleben. Kirchner, J. 1910: De Servii, carminum Vergilianorum interpretis, commentario pleniore qui dicitur. Partic. I, Brzeg. König, J., and T. Whitmarsh (eds) 2007: Ordering Knowledge in the Roman Empire, Cambridge and New York.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 130

14/11/2017 14:56:23

BIBLIOGRAPHY

131

König, J., and G. Woolf (eds) 2013: Encyclopaedism from Antiquity to the Renaissance, Cambridge. König, J., K. Oikonomopoulou, and G. Woolf (eds) 2013: Ancient Libraries, Cambridge. Kronenberg, L. 2009: Allegories of Farming from Greece and Rome: Philosophical Satire in Xenophon, Varro, and Virgil, Cambridge. Kumaniecki, K. F. 1974–5: ‘De Varronis libro Isagogico ad Pompeium eiusque dispositione’, ACD 10–11, 41–44. Kunkel, W. 1952: Herkunft und soziale Stellung der römischen Juristen, Weimar. Laarhoven, J. van (ed.) 1987: John of Salisbury’s Entheticus Maior and Minor, 3 vols, Leiden. Labov, W. 2008: The Social Stratification of English in New York City: Origins, Evolution and Variation, Cambridge. Lacey, W. K. 1996: Augustus and the Principate: The Evolution of the System, Leeds. Lammert, F. 1912: De Hieronymo Donati discipulo, Commentationes Philologae Jenenses 9.2, Leipzig. Lausberg, H. 1990: Handbuch der literarischen Rhetorik (3rd edn), Stuttgart. Lazzarini, C. 1984: ‘Historia / fabula: forme della costruzione poetica virgiliana nel commento di Servio all’Eneide’, MD 12, 117–44. ——— 2013: ‘Servio: lezioni di stile. Citazioni di poeti fra esegesi e formazione’, in Stok (ed.), 101–24. Leach, E. W. 1999: ‘Ciceronian “Bi-Marcus”: correspondence with M. Terentius Varro and L. Papirius Paetus in 46 bce’, TAPhA 129, 139–79. Leeman, A. D., H. Pinkster, and H. L. W. Nelson (eds) 1985: M. Tullius Cicero, De Oratore Libri III. Kommentar. 2. Band, Buch I, 166–265, Buch II, 1–98, Heidelberg. Lehmann, A. 2002: Varron critique littéraire: Regard sur les poètes latins archaïques, Brussels. Lehmann, Y. 1986: ‘Religion et politique: autour des Antiquités divines de Varron’, REL 64, 92–103. ——— 1997: Varron théologien et philosophe romain, Brussels. Leuze, O. 1911: ‘Das synchronistische Kapitel des Gellius (Noct. Att. XVII 21)’, RhM 66, 237–74. Lewis, R. G. (ed.) 2006: Asconius. Commentaries on Speeches of Cicero, Oxford. Lieberg, G. 1973: ‘Die “theologia tripertita” in Forschung und Bezeugung’, ANRW 1.4, 63–115. Linderski, J. 1985a: ‘The dramatic date of Varro, De re rustica, book III and the elections in 54’, Historia 34, 248–54. ——— 1985b: ‘The libri reconditi’, HSPh 89, 207–34. ——— 1986: ‘The augural law’, ANRW 2.16.3, 2146–312. ——— 1989: ‘Garden parlors: nobles and birds’, in R. I. Curtis (ed.), Studia Pompeiana et Classica in Honor of Wilhelmina F. Jashemski, vol. II, New Rochelle, NY, 105–28. Lindsay, W. M. (ed.) 1903: Noni Marcelli de conpendiosa doctrina libri XX, Leipzig.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 131

14/11/2017 14:56:23

132

BICS-60-2  2017

Lintott, A. 1972: ‘Imperial expansion and moral decline in the Roman Republic’, Historia 21, 626–38. Lloyd, R. B. 1961: ‘Republican authors in Servius and the Scholia Danielis’, HSPh 65, 291–341. Lovejoy, A. O., and G. Boas 1935: Primitivism and Related Ideas in Antiquity, Baltimore. Lowe, E. A. 1934–71: Codices Latini Antiquiores, 11 vols + supplement, Oxford. Lowry, S. T. 1987: ‘The Greek heritage in economic thought’, in S. T. Lowry, Pre-Classical Economic Thought, Boston, 7–30. ——— 1998: ‘Economic and jurisprudential ideas of the Greeks’, in S. T. Lowry and B. Gordon (eds), Ancient and Medieval Economic Ideas and Concepts of Social Justice, Leiden, 11–37. ——— 2003: ‘Ancient and medieval economics’, in W. J. Samuels, J. E. Biddle and J. B. Davis (eds), A Companion to the History of Economic Thought, Oxford, 11–27. McConnell, S. 2014: Philosophical Life in Cicero’s Letters, Cambridge. MacRae, D. 2016: Legible Religion: Books, Gods, and Rituals in Roman Culture, Cambridge, MA and London. ——— 2018: ‘Diligentissumus investigator antiquitatis? “Antiquarianism” and historical evidence between republican Rome and the early modern republic of letters’, in Sandberg and Smith (eds), 137–156. Magdelain, A. 1973: ‘Remarques sur la perduellio’, Historia 22, 405–22. Maltby, R. 2005: ‘Donatus and Terence in Servius and Servius Danielis’, in Th. Fögen (ed.), Antike Fachtexte. Ancient Technical Texts, Berlin and New York, 207–20. Mantovani, D. 2008: ‘Leges et iura p(opuli) R(omani) restituit: principe e diritto in un aureo di Ottaviano’, Athenaeum 96, 5–54. Markus, R. A. 1996: ‘Response’ to O’Daly, in Sommerstein (ed.), 77–80. Marshall, A. 1993: ‘Atticus and the Genealogies’, Latomus 52, 307–17. Marshall, R. M. A. 2014: The Reception of Varro in Late Antiquity, Diss., Oxford. ——— 2016, ‘Varro nouus’, CR 66, 412–17. ——— forthcoming: The Reception of Varro in Late Antiquity. Martin, R. 1971: Recherches sur les agronomes latins et leurs conceptions économiques et sociales, Paris. Mayer, R. 2005: ‘Creating a literature of information in Rome,’ in M. Horster and C. Reitz (eds), Wissensvermittlung in dichterischer Gestalt, Stuttgart, 227–42. Menge, H., T. Burkard, and M. Schauer 2000: Lehrbuch der lateinischen Syntax und Semantik, Darmstadt. Merkel, R. (ed.) 1841: P. Ovidii Nasonis Fastorum libri sex, Berlin. Meyer, E. A. 2004: Legitimacy and Law in the Roman World: Tabulae in Roman Belief and Practice, Cambridge. Michels, A. 1967: The Calendar of the Roman Republic, Princeton. Mirhady, D. C. (ed.) 2001: ‘Dicaearchus of Messana: the sources, text, and translation’, in Fortenbaugh and Schütrumpf (eds), 1–142.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 132

14/11/2017 14:56:23

BIBLIOGRAPHY

133

Mirsch, P. (ed.) 1882: ‘De M. Terenti Varronis Antiquitatum rerum humanarum libris XXV’, Leipziger Studien zur classischen Philologie 5, 1–144. Repr. Hildesheim, 1972. Mitteis, L. 1908: Römisches Privatrecht bis auf die Zeit Diokletians, vol. I, Leipzig. Moatti, C. 1997 (trans. 2015): La Raison de Rome: naissance de l’esprit critique à la fin de la République (IIe–Ier siècle avant Jésus-Christ), Paris (= The Birth of Critical Thinking in Republican Rome, trans. by J. Lloyd, Cambridge 2015). Moeller, A. 1892: Quaestiones Servianae, Kiel. Möller, A. 2005: ‘Epoch-making Eratosthenes’, GRBS 45, 245–60. Momigliano, A. 1950: ‘Ancient history and the antiquarian’, JWCI 13, 285–315 (= Contributo alla storia degli studi classici, Rome 1955, 67–106 = Studies in Historiography, London 1966, 1–39). ——— 1966: ‘The consequences of new trends in the history of ancient law’, in Studies in Historiography, London, 239–56. ——— 2003: ‘The theological efforts of the Roman upper classes in the first century BC’, in C. Ando (ed.), Roman Religion, Edinburgh, 147–63. Reprint of CP 79 (1984), 199–211. Mommsen, Th. 1856: Römische Geschichte, vol. III, Berlin. ——— 1887: Römisches Staatsrecht (3rd edn), vol. I, Leipzig. ——— 1888: Römisches Staatsrecht (3rd edn), vol. III.1, Leipzig. Monaco, L. 1996: Persecutio piratarum: Battaglie ambigue e svolte costituzionali nella Roma repubblicana, Naples. Morgan, M. G. 1974: ‘Three notes on Varro’s Logistorici’, MH 31, 117–28. Mountford, J. F., and J. T. Schultz 1930: Index rerum et nominum in scholiis Servii et Aelii Donati tractatorum, Ithaca, NY. Müller, C. F. W. (ed.) 1898–1908: M. Tullii Ciceronis scripta quae manserunt omnia, 10 vols, Leipzig. Müller, L. (ed.) 1888: Noni Marcelli Compendiosa doctrina, Leipzig. Murgia, C. E. 1975: Prolegomena to Servius 5: The Manuscripts, Berkeley. ——— 2003: ‘The dating of Servius Revisited’, CPh 98, 45–69. Nelsestuen, G. A. 2011: ‘Polishing Scrofa’s agronomical Eloquentia: representation and revision in Varro’s De re rustica’, Phoenix 65, 315–51. ——— 2015: Varro the Agronomist: Political Philosophy, Satire, and Agriculture in the Late Republic, Columbus, OH. Neudecker, R. 2013: ‘Archives, books and sacred space in Rome’, in König, Oikonomopoulou, and Woolf (eds), 312–31. Newmyer, S. T. 2011: Animals in Greek and Roman Thought: A Sourcebook, London. Nicolet, C. 1970: ‘Le livre III des “Res rusticae” de Varron et les allusions au déroulement des comices tributes’, REA 72.2, 113–37. Noe, E. 1977: ‘I proemi del “de re rustica” di Varrone’, Athenaeum 55, 289–302. Nolhac, P. de 1907: Pétrarque et l’humanisme (2nd edn), 2 vols, Paris. North, J. A. 1995: ‘Religion and rusticity’, in Cornell and Lomas (eds), 135–50.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 133

14/11/2017 14:56:23

134

BICS-60-2  2017

——— 1998: ‘The books of the pontifices’, in La mémoire perdue: Recherches sur l’administration romaine, Rome, 45–63. ——— 2009: ‘An overview of Roman religious law’, in The Oxford International Encyclopedia of Legal History, vol. V, Oxford, 120–22. ——— 2014: ‘The limits of the “religious” in the late Roman republic’, HR 53, 225–45. Nünlist, R. 2015: ‘What does Ὅμηϱον ἐξ Ὁμήϱου σαφηνίζειν actually mean?’, Hermes 143, 385–403. Oakley, S. P. 1999: A Commentary on Livy, Books VI–X, vol. I, Introduction and Book VI, Oxford. O’Daly, G. 1996: ‘Augustine’s critique of Varro on Roman religion’, in Sommerstein (ed.), 65–75. ——— 1999: Augustine’s City of God: A Reader’s Guide, Oxford. Ogilvie, R. M. 1965: A Commentary on Livy, Books 1–5, Oxford. ——— 1978: The Library of Lactantius, Oxford. Oldfather, W. A. 1908: ‘Livy i, 26 and the supplicium de more maiorum’, TAPhA 39, 49–72. Osgood, J. 2006: Caesar’s Legacy, Cambridge. Owen, T. 1800: The Three Books of M. Terentius Varro Concerning Agriculture, Oxford. Pacca, V., and L. Paolino (eds) 1996: Francesco Petrarca: Trionfi, Rime Estravaganti, Codice degli abbozzi, Milan. Pani, M. 2006: ‘Sulle conoscenze costituzionali nell’aristocrazia romana’, in L. Labruna (ed.), Tradizione romanistica e costituzione, vol. I, Naples, 727–40. Pellizzari, A. 2003: Servio: storia, cultura e istituzioni nell’opera di un grammatico tardoantico, Firenze. Pépin, J. 1956: ‘La “Théologie tripartite” de Varron: essai de reconstitution et recherche des sources’, REAug 2, 265–94. Perlwitz, O. 1992: Titus Pomponius Atticus: Untersuchungen zur Person eines einflussreichen Ritters in der ausgehenden Römischen Republik, Stuttgart. Pernice, A. 1885: ‘Zum römischen Sacralrechte I’, Sitzungsberichte der preussischen Akademie der Wissenschaften, 1143–69. ——— 1886: ‘Zum römischen Sacralrechte II’, Sitzungsberichte der preussischen Akademie der Wissenschaften, 1169–203. Peter, H. (ed.) 1906: ‘De gente populi Romani libri III’, in H. Peter (ed.), Historicorum Romanorum Reliquiae, vol. II, Leipzig, 10–24. Piras, G. 1998: Varrone e i ‘poetica verba’: Studio sul settimo libro del ‘De lingua Latina’, Bologna. ——— 2007–08: ‘Nuove testimonianze dalla biblioteca di Petrarca: le annotazioni al De lingua latina di Varrone’, Quaderni Petrarcheschi 17–8, 829–56. ——— 2015: ‘Cum poeticis multis uerbis magis delecter quam utar: poetic citations and etymological enquiry in Varro’s De lingua Latina’, in Butterfield (ed.), 51–70. Pittà, A. (ed.) 2015: M. Terenzio Varrone, De vita populi Romani: Introduzione e commento, Pisa.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 134

14/11/2017 14:56:23

BIBLIOGRAPHY

135

Poucet, J. 1987: ‘Temps mythique et temps historique: les origines et les premiers siècles de Rome’, Gerion 5, 69–85. Prescendi, F. 2007: Décrire et comprendre le sacrifice: les réflexions des Romains sur leur propre religion à partir de la littérature antiquaire, Stuttgart. Prokoph, F. M. 2015: ‘Cornelius Nepos und ein Stück Literaturgeschichte: Indizien zur Chronologie der Imagines des Atticus und des Varro’, in B. Dunsch and F. M. Prokoph (eds), Geschichte und Gegenwart: Beiträge zu Cornelius Nepos aus Fachwissenschaft, Fachdidaktik und Unterrichtspraxis, Wiesbaden, 85–133. Purcell, N. 1995: ‘The Roman villa and the landscape of production’, in Cornell and Lomas (eds), 151–79. ——— 2003 (= 2005): ‘The way we used to eat: diet, community, and history at Rome’, AJP 124, 329–58. Repr. in B. K. Gold and J. F. Donahue (eds), Roman Dining: A Special Issue of AJP, Baltimore, 1–30. Rafferty, D. 2011: ‘Princeps senatus’, Melbourne Historical Journal 2, 1–22. Rapisarda, C. A. (ed.) 1991: Censorini De die natali liber ad Q. Caerellium, Pref., testo critico, trad., e comm., Bologna. Rawson, E. 1972: ‘Cicero the Historian and Cicero the Antiquarian’ JRS 62, 33–45. ——— 1978: ‘The introduction of logical organization in Roman Prose Literature,’ PBSR 46, 12–34 (= Roman Culture and Society: Collected Papers, Oxford, 1991, 324–51). ——— 1985: Intellectual Life in the Late Roman Republic, London. ——— 1991: ‘The first Latin annalists’, in Roman Culture and Society: Collected Papers, Oxford, 245–71. Repr. of Latomus 35, 1971, 689–717. Reischl, B. 1976: Reflexe griechischer Kulturentstehungslehren bei augusteischen Dichtern, Munich. Ribbeck, O. 1866: Prolegomena critica ad P. Vergilii Maronis opera maiora, Leipzig., repr. Hildesheim 1966. Rich, J. W., and J. H. C. Williams 1999: ‘Leges et iura p.R. restituit: a new aureus of Octavian and the settlement of 28–27 BC’, NC 159, 169–213. Riese, A. (ed.) 1865: M. Terenti Varronis Saturarum Menippearum reliquiae, Leipzig. Rinkewitz, W. 1981: Pastio villatica: Untersuchungen zur intensiven Hoftierhaltung in der römischen Landwirtschaft, Frankfurt. Riposati, B. (ed.) 1939: M. Terenti Varronis De vita populi romani. Fonti, esegesi, edizione critica dei frammenti, Milan. Riposati, B., and A. Marastoni 1974 : Bibliografia varroniana, Milan. Ritschl, F. W. 1877a: Kleine philologische Schriften [= Opuscula Philologica], vol. III, zur römischen Litteratur [= ad litteras Latinas spectantia], Leipzig, 418–505. Reprint of F. W. Ritschl, Die Schriftstellerei des M. Terentius Varro und die des Origenes, nach dem ungedruckten Katalog des Hieronymus, Bonn, 1847; reprinted as ‘Die Schriftstellerei des M. Terentius Varro’, RhM 6 1848, 481–560. ——— 1877b: ‘Ueber des Marcus Terentius Varro Imaginum sive Hebdomadum libri’, in Kleine philologische Schriften [= Opuscula Philologica], vol. III, zur römischen Litteratur [= ad litteras Latinas spectantia], Leipzig, 508–92.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 135

14/11/2017 14:56:23

136

BICS-60-2  2017

Rocca-Serra, G. (ed.) 1980: Censorinus, Le jour natal, Traduction annotée, Paris. Rösch-Binde, C. 1998: Vom “δεινὸς ἀνήρ” zum “diligentissimus investigator antiquitatis”: Zur komplexen Beziehung zwischen M. Tullius Cicero und M. Terentius Varro, Munich. Rohde, G. 1936: Die Kultsatzungen der römischen Pontifices, Berlin. Rolfe, J. C. (ed.) 1948–54: The Attic Nights of Aulus Gellius (revised edn), 3 vols, London and Cambridge, MA. ——— (ed.) 1961: The Attic Nights of Aulus Gellius (2nd edn), London and Cambridge, MA. ——— (ed.) 1984: Cornelius Nepos (revised edn), London and Cambridge, MA. Romano, E. 2003: ‘Il concetto di antico in Varrone’, in M. Citroni (ed.), Memoria e identità: La cultura romana costruisce la sua immagine, Firenze, 99–117. Rolle, A. 2017: Dall’Oriente a Roma: Cibele, Iside e Serapide nell’opera di Varrone, Pisa. Roselaar, S. T. 2010: Public Land in the Roman Republic, Oxford. Roth, C. L. 1861: ‘Zu Varro de lingua latina: Randbemerkungen’, Philologus 17, 175–6. Rousseau, P. 2009: ‘Language, morality, and cult: Augustine and Varro’, in P. Rousseau and M. Papoutsakis (eds), Transformations of Late Antiquity: Essays for Peter Brown, Farnham, 159–75. Rüpke, J. 1990: Domi militiae: Die religiöse Konstruktion des Krieges in Rom, Stuttgart. ——— 2005: ‘Varro’s tria genera theologiae: religious thinking in the late republic’, Ordia Prima 4, 107–29. ——— 2008: Fasti Sacerdotum: A Prosopography of Pagan, Jewish, and Christian Religious Officials in the City of Rome, 300 bc to ad 499 (revised English edn), Oxford. ——— 2011a: ‘Different colleges – Never mind!?’, in J. H. Richardson and F. Santangelo (eds), Priests and State in the Roman World, Stuttgart, 25–38. ——— 2011b: The Roman Calendar from Numa to Constantine: Time, History, and the Fasti, Chichester. ——— 2014: ‘Historicizing religion: Varro’s Antiquitates and history of religion in the late Roman republic’, HR 53, 246–68. Ryan, F. X. 1998: Rank and Participation in the Republican Senate, Stuttgart. Salvadore, M. (ed.) 2004: M. Terenti Varronis. Fragmenta omnia quae extant, II, De vita populi Romani libri IV, Hildesheim,-Zurich-New York. Sandberg, K., and C. Smith (eds) 2018: OMNIUM ANNALIUM MONUMENTA: Historical Writing and Historical Evidence in Republican Rome, Leiden. Sanio, F. D. 1867: Varroniana in den Schriften der römischen Juristen, Leipzig. Santalucia, B. 1994: Studi di diritto penale romano, Rome. Santangelo, F. 2013: ‘Priestly auctoritas in the Roman republic’, CQ 63, 743–63. Santini, C. 2008: ‘Virgil as Etruscologist in Servius’s commentary, or: from the author to the author’, in S. Casali and F. Stok (eds), Servio: stratificazioni esegetiche e modelli culturali / Servius: Exegetical Stratifications and Cultural Models, Brussels, 78–92.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 136

14/11/2017 14:56:24

BIBLIOGRAPHY

137

Saunders, T. J. 2001: ‘Dicaearchus’ historical anthropology’, in Fortenbaugh and Schütrumpf (eds), 237–54. Scheid, J. 1994. ‘Les archives de la pieté: reflexions sur les livres sacerdotaux’, in Le mémoire perdue: À la recherche des archives oubliées, publiques et privées, de la Rome antique, Paris, 173–85. ——— 2006: ‘Oral tradition and written tradition in the formation of sacred law in Rome’, in C. Ando and J. Rüpke (eds), Religion and Law in Classical and Christian Rome, Stuttgart, 14–33. Schiavone, A. 2012: The Invention of Law in the West, Cambridge, MA. Schmidt, P. L. 2000: ‘C. Suetonius Tranquillus’, in K. Sallmann (ed.), L’Âge de Transition: De la littérature romaine à la littérature chrétienne de 117 a 284 après J.-C. [= R. Herzog and P. L. Schmidt (gen. eds), Nouvelle histoire de la littérature latine, vol. IV], Paris, 14–58. Schneider, J. G. 1794: Scriptorum rei rusticae veterum Latinorum, Turin. Schrader, H. (ed.) 1882: Porphyrii quaestionum Homericarum ad Iliadem pertinentium reliquiae, fasc. 2, Leipzig. Schröder, B. J. 1999: Titel und Text, Berlin. Schulz, F. 1946: History of Roman Legal Science, Oxford. Schütrumpf, E. 2001: ‘Dikaiarchs Βίος Ἑλλάδος und die Philosophie des vierten Jahrhunderts’, in Fortenbaugh and Schütrumpf (eds), 255–77. Scott Perry, J. 2014: ‘The Lex Julia de senatu habendo: a view from the 1930s’, in L. L. Brice and D. Slootjes (eds), Aspects of Ancient Institutions and Geography: Studies in honor of R. J. A. Talbert, Leiden and Boston, 48–64. Seager, R. 2002: Pompey the Great: A Political Biography (2nd edn), London. Semi, F. (ed.) 1965: M. Terentius Varro, 4 vols, Venice. Shackleton Bailey, D. R. (ed.) 1965: Cicero’s Letters to Atticus, vol. I, 68-59 bc, letters 1–45 (Books I and II), Cambridge. ——— (ed.) 1967: Cicero’s Letters to Atticus, vol. VI, 44 bc, letters 355–426 (Books XIV– XVI), Cambridge. Shanzer, D. 2005: ‘Augustine’s Disciplines: Silent diutius Musae Varronis?’, in K. Pollmann and M. Vessey (eds), Augustine and the Disciplines, Oxford, 69–112. Shatzman, I. 1975: Senatorial Wealth and Roman Politics, Brussels. Sherk, R. K. 1969: Roman Documents from the Greek East: Senatus Consulta and Epistulae to the Age of Augustus, Baltimore. Silvermintz, D. 2010: ‘Plato’s supposed defense of the division of labor: a reexamination of the role of job specialization in the Republic’, History of Political Economy, 42, 747–72. Smith, C. 2006: The Roman clan: the gens from ancient ideology to modern anthropology, Cambridge. Smith, C. 2006a: ‘Adfectatio regni in the Roman republic’, in S. Lewis (ed.), Ancient Tyranny, Edinburgh, 49–64. Sommerstein, A. H. (ed.) 1996: Religion and Superstition in Latin Literature, Bari.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 137

14/11/2017 14:56:24

138

BICS-60-2  2017

Spencer, D. forthcoming: Varro’s Guide to Being Roman. Spengel, L., and A. Spengel (eds) 1885: M. Terenti Varronis De lingua Latina libri, Berlin. Stella Maranca, F. 1934: ‘Varrone giureconsulto?’, Annali del seminario giuridicoeconomico della R. Università di Bari, 167–87. Stok, F. (ed.) 2013: Totus scientia plenus: Percorsi dell’esegesi virgiliana antica, Pisa. Strzelecki, W. (ed.) 1967: C. Atei Capitonis fragmenta, Leipzig. Talbert, R. J. A. 1984a: The Senate of Imperial Rome, Princeton. ——— 1984b: ‘Augustus and the Senate’, G&R 31, 55–63. Tansey, P. 2000: ‘The princeps senatus in the last decades of the republic’, Chiron 30, 15–30. Tarver, T. 1996: ‘Varro, Caesar, and the Roman calendar: a study in late republican religion’, in Sommerstein (ed.), 39–57. Taylor, D. J. 1977: ‘Two notes on Varro’, AJP 98, 130–2. ——— 2015: ‘The new Varro and the structure of his De Lingua Latina’, in Butterfield (ed.), 19–31. Taylor, L. R. 1934: ‘Varro’s De gente populi Romani’, CPh 29, 221–9. Tellegen-Couperus, O. 2012: ‘Sacred law and civil law’, in O. Tellegen-Couperus (ed.), Law and Religion in the Roman Republic, Leiden and Boston, 147–64. Teuffel, W. S. 1873: A History of Roman Literature, English translation, London. Thibodeau, P. 2011: Playing the Farmer: Representations of Rural Life in Vergil’s Georgics, Berkeley. Thilo, G. (ed.) 1881–7: Servii grammatici qui feruntur in Vergilii carmina commentarii, 3 vols, Leipzig. Thomas, É. 1880: Scoliastes de Virgile: Essai sur Servius et son commentaire sur Virgile, Paris. Thompson, J. C. 2013: ‘The medieval manuscript tradition of Augustine’s works: an overview from 400 to 1200’, in K. Pollmann (ed.), The Oxford Guide to the Historical Reception of Augustine, vol. I, Oxford, 51–8. Timpanaro, S. 1986: Per la storia della filologia virgiliana antica, Rome. Todisco, E. 2007: ‘La res publica restituta e i Fasti Praenestini’, in M. Pani (ed.), Epigrafia e territorio, politica e società: Temi di antichità romane, vol. VII, Bari, 341–58. ——— 2016a: ‘Dal Commentario isagogico alla epistola ad Oppiano: riscritture politiche varroniane’, in M. Chelotti and M. Silvestrini (eds), 479–88. ——— 2016b: ‘Il frammento curia/cura nel de vita populi Romani di Varrone tra etimologia e politica’, in M. Chelotti and M. Silvestrini (eds), 489–97. ——— 2018a forthcoming: ‘Il contributo di Varrone alla conoscenza delle procedure di funzionamento del Senato’, in P. Buongiorno and. S. Lohsse (eds), Darstellung und Gebrauch der Senatus consulta in den literarischen Quellen der kaiserlichen Zeit, (Münster, 5–7 May 2015). ——— 2018b forthcoming: ‘Echi degli interventi augustei sul Senato nel quarto libro delle Elegie di Properzio (4.1.11-14)’, in Properzio tra Repubblica e Principato, XXI Convegno Internazionale (Assisi–Cannara, 30 maggio–1 giugno 2016 Atti Convegno properziano). © 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 138

14/11/2017 14:56:24

BIBLIOGRAPHY

139

Tortorici, E. 1993: ‘Curia Iulia’, in Lexicon Topographicum Urbis Romae, I, Rome, 332334. Traglia, A. (ed.) 1974: M. Terenzio Varrone, Opere, Turin (repr. 1992). ——— 1993: Varrone prosatore, in B. Amata (ed.), Cultura e lingue classiche 3, III Convegno di aggiornamento e di didattica (Palermo, 29 ottobre–1 novembre 1989), Rome, 693–885. Traversa, L. 2009–10: ‘Comunicazione e partecipazione politica: il Commentariolum petitionis’, AFLB 51–2, 115–52. Tuori, K. 2007: Ancient Roman Lawyers and Modern Legal Ideals: Studies on the Impact of Contemporary Concerns in the Interpretation of Ancient Roman Legal History (2nd edn), Frankfurt am Main. Turcan, R. (ed.) 1982: Firmicus Maternus: L’erreur des religions païennes, Paris. Tyrrell, W. B. 1978: A Legal and Historical Commentary to Cicero’s Oratio pro C. Rabirio perduellionis reo, Amsterdam. Untermann, J. 2000: Wörterbuch des Oskisch-Umbrischen, Heidelberg. Uría Varela, J., and R. Gutiérrez González 2011: ‘Vague boundaries: delimiting grammatical fragments in Charisius’, Eruditio Antiqua 3, 57–72. Vallat, D. 2012a: ‘Le Servius Danielis: introduction’, Eruditio Antiqua 4, 89–99. ——— 2012b: ‘Servius Danielis et les obtrectatores: éléments de polémique antivirgilienne’, Eruditio Antiqua 4, 247–87. ——— 2013: ‘Per transitum tangit: allusions, sens cachés et réception de Virgile dans le commentaire de Servius’, in Stok (ed.), 51–81. ——— 2014: ‘Une présence silencieuse: Festus et les commentaires virgiliens’, Eruditio Antiqua 6, 143–63. ——— 2015: ‘Conflits d’autorité: Virgile, Donat, Servius’, Eruditio Antiqua 7, 5–30. ——— 2016: ‘Les métamorphoses d’un commentaire: “Servius” et Virgile’, Rursus 9 (https://rursus.revues.org/1190). van Nuffelen, P. 2010: ‘Varro’s Divine Antiquities: Roman religion as an image of truth’, CPh 105, 162–88. Velaza, J. 2008: ‘Servius et l’Histoire Auguste: un problème de datations en chaîne?’, RPh 82, 147–56. Vessey, M. 2012: ‘The history of the book: Augustine’s City of God and post-Roman cultural memory’, in J. Wetzel (ed.), Augustine’s City of God: A Critical Guide, Cambridge, 14–32. Volk, K. forthcoming: ‘Varro and the Disorder of Things’, HSPh 110. Walde, A., and J. B. Hofmann 1965: Lateinisches etymologisches Wörterbuch (4th edn), 2 vols, Heidelberg. Wallace-Hadrill, A. 1997: ‘Mutatio morum: the idea of a cultural revolution’, in T. N. Habinek and A. Schiesaro (eds), The Roman Cultural Revolution, Cambridge, 3–22. ——— 2008: Rome’s Cultural Revolution, Cambridge. Watson, A. 1987: ‘The birth of the legal profession’, Michigan Law Review 85, 1071–82. ——— 1992: The State, Law and Religion: Pagan Rome, Athens.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 139

14/11/2017 14:56:24

140

BICS-60-2  2017

——— 1995: The Spirit of Roman Law, Athens. ——— 2007: ‘Law and society’, in Cairns and du Plessis (eds), 9–35. Webb, C. C. I. (ed.) 1909: Ioannis Saresberiensis […] Policratici, 2 vols, Oxford. Wehrli, F. (ed.) 1967: Die Schule des Aristoteles: Texte und Kommentare, vol. I, Dikaiarchos, (2nd edn) Basel and Stuttgart. White, K. D. 1970: Roman Farming, London. ——— 1973: ‘Roman agricultural writers I: Varro and his predecessors’, ANRW 1.4, 439– 97. White, S. A. 2001: ‘Principes sapientiae: Dicaearchus’ biography of philosophy’, in Fortenbaugh and Schütrumpf (eds), 195–236. Wibier, M. 2014: ‘The topography of the law book: common structures and modes of reading’, in L. Jansen (ed.), The Roman Paratext, Cambridge, 56–72. Wieacker, F. 1988: Römische Rechtsgeschichte: Quellenkunde, Rechtsbildung, Jurisprudenz und Rechtsliteratur, vol. I, Munich. Willems, P. 1878–85: Le Sénat de la République romaine: Sa composition et ses attributions, vol. II, Leuven. Wiseman, T. P. 1987: ‘Topography and rhetoric: the trial of Manlius’, in T. P. Wiseman, Roman Studies: Literary and Historical, Liverpool, 225–43. (Reprint of Historia 28, 1979, 32–50.) ——— 2002: ‘History, poetry, and Annales’, in D. S. Levene and D. P. Nelis (eds), Clio and the Poets, Leiden-Boston-Köln, 331–62. ——— 2009: Remembering the Roman People: Essays on Late-Republican Politics and Literature, Oxford. ——— 2016: ‘Review-discussion: Varro’s biography of the Roman people’, Histos 10, CXI–CXXVIII. Wisse, J., M. Winterbottom, and E. Fantham (eds) 2008: M. Tullius Cicero, De oratore Libri III: A Commentary on Book III, 96–230, Heidelberg. Woodman, A. J. 1988: Rhetoric in Classical Historiography, London. ——— 2008: ‘Cicero on historiography: De oratore 2.51–64’, CJ 104, 23–31. Yavetz, Z. 1974: ‘Existimatio, fama, and the Ides of March’, HSPh 78, 35–65. Zehnacker, H. 2003: ‘Quelques remarques sur le revers du nouvel aureus d’Octavien (28 av. J.-C.)’, Bulletin de la Société Française de Numismatique 58, 1–3. Zetzel, J. E. G. 1981: Latin Textual Criticism in Antiquity, New York. ——— 2013: ‘A contract on Ameria: law and legality in Cicero’s Pro Roscio Amerino’, AJP 134, 425–44.

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:16 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 140

14/11/2017 14:56:24

GENERAL INDEX For ancient authors, see also the Index Locorum Aelius Paetus Catus, Sextus 39, 40, 41 Aelius Stilo (Lucius Aelius Stilo Praeconinus) 67 Aelius Tubero, Quintus 39, 59 Antiochus of Ascalon 2, 67 Antistius Labeo, Marcus 39 Aristarchus of Samothrace 103 Arnobius 81, 106 Asinius Pollio, Gaius 59 Ateius Capito, Gaius 39, 41–42, 54 Atticus, Titus Pomponius 56–57, 61–75 passim (esp. 67–72) Augustine of Hippo (Aurelius Augustinus) 3, 6, 34, 42–44, 76–91 passim, 106 De civitate Dei 76–91 passim (esp. 79–82), 95 judgement on Varro 77–78, 80, 85 and n. 44, 87, 95

Ennius, Quintus 19, 29, 64 Etymology 9, 11, 15, 16, 40, 42, 119–20 Fabius Maximus Servilianus 39 Fabius Pictor, Quintus 9, 39, 40, 42, 46 Festus, Sextus Pompeius 41, 42, 45–46, 105, 119 Gellius, Aulus 42, 44–45, 49–60 passim, 62–67, 92, 105, 109 Granius Flaccus 39, 41, 46 Iulius Caesar, Lucius 39 Ius pontificium 38–42 Ius primae sententiae 53–54 Lactantius, Lucius Caecilius Firmianus 44, 81 Lex Aurelia de tribunicia potestate 56 Lex Cornelia de legibus solvendis 56 Lex Gabinia de senatu legatis dando 55 Lex Iulia de senatu habendo 53, 55 Lex Pompeia Licinia de tribunicia potestate 56 Lex Pompili regis 46 Lex Pupia de senatu diebus comitialibus non habendo 55

Brutus, Marcus Junius 39 Caesar, Gaius Iulius 5–6, 57–58, 69, 74–75 dedication of Ant. div. to 6, 44, 57 Cato, Marcus Porcius (the Elder) 9, 65, 106, 116 Catullus, Gaius Valerius 74 Censorinus 13–14, 20, 62 Cicero, Marcus Tullius 9, 34–35, 37–39, 42, 46, 64–66, 68–69, 78–79 De republica 68–69, 76 praise of Varro 1, 5, 34–35, 71, 77, 80–81, 91 Claudius Marcellus, Marcus 39 Claudius Pulcher, Appius 31, 39, 41

Macrobius Ambrosius Theodosius 42, 45, 46, 78, 79 Manilius, Manius 39, 47 Manlius, Marcus Manlius Capitolinus 72–75 Minucius Felix, Marcus 81 Mos maiorum 4 n. 16, 21 54, 58, 87 Mucius Scaevola, Publius 37 Mucius Scaevola, Quintus 37, 39, 41, 43–44, 46

Dicaearchus of Messana 3, 12, 13, 17, 20, 22, 24–28 Donatus, Aelius 93, 94, 95, 105 BICS-60-2  2017

141 © 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:28 +00:00 All use subject to https://about.jstor.org/terms

BICS 60-2 Nov14-2 FINAL-FINAL.indd 141

14/11/2017 14:56:24

BICS-60-2  2017

142

Nepos, Cornelius 62, 66, 73–74 Nonius Marcellus 46, 100, 105 Numa Pompilius 19, 41, 46, 47, 82, 90 Octavian (Gaius Octavius) 53–54, 59–60 Oppianus 49–60 passim (esp. 59–60) Pastio villatica 3, 11, 21–33 passim (esp. 29–33) Piso, Lucius Calpurnius Piso Frugi 9 Pliny the Elder (Gaius Plinius Secundus) 6, 102, 105, 116 Pompey (Gnaeus Pompeius Magnus) 5–6, 49–60 passim, 68, 75 Porphyry 25–28 Senate 4, 51–52, 54–55, 56, 57, 60, 67, 90 Seneca, Lucius Annaeus 78, 86 Servius, Maurus Servius Honoratus 78, 92–107 passim Servius Danielis 92–107 passim (esp. 93, 99) Sextus Papirius (pont. max.) 41 Suetonius (Gaius Suetonius Tranquillus) 102, 105 Sulla, Lucius Cornelius 56 Sulpicius Rufus, Publius 39 Tertullian (Quintus Septimius Florens Tertullianus) 45, 81–83 Tiberius Coruncanius (pont. max.) 42 Trebatius Testa, Gaius 39, 40–41 Tullus Hostilius 112 Twelve Tables 39, 41, 73 Valerius Messalla, Rufus 39 Varro, Marcus Terentius Annales 60–75 passim dating 71

Antiquitates rerum humanarum et divinarum 8, 79, 97 Antiquitates rerum humanarum 14 and n. 23, 35 n. 5 Antiquitates rerum divinarum 34–48 passim, 76–91 passim (esp. 79, 89) authority of 6, 59, 81–82, 85, 103, 105–6 De familiis Troianis 5, 70 De gente populi Romani 58, 71–72 De lingua Latina 6, 8–9, 108–122 passim (esp. 109, 120) De poetis 64–65 De re rustica 11–13, 21–33 passim De vita populi Romani 3–4, 6, 58 Epistolicae quaestiones 49–51 Epistula ad Oppianum 49–60 passim grammatical conjugation 114–16 grammatical declension 114–16 Imagines 69–70 legate to Pompey 68 librarian 57, 75 Logistorici 14 n. 24, 59 n. 69, 69 and n. 52, 96–97 methodology of 2–3, 8, 11, 15–16 past 13–17, 19–20 past traditions 3–6, 88 political activity of 5–6, 67 praetorship of 6, 75 theology 80–89 theologia tripertita 15, 43, 82–83, 85, 87, 89 tribunate of 67, 75 Varro Atacinus, Publius Terentius 94 Veranius 39, 46 Verrius Flaccus, Marcus 46, 64, 108 Virgil (Publius Vergilius Maro) 92–107 passim

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:28 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 142

14/11/2017 14:56:24

INDEX LOCORUM Appian of Alexandria Bellum civile 1.121 56 4.47 59 4.202 75

Caesar, Gaius Iulius Bellum civile 2.17.1–3

Catullus, Gaius Valerius 1.5–7 74

Ateius Capito, Gaius (Strzelecki) Fr 10 42 n. 53

Claudius Pulcher, Appius (Bremer) Fr 1 41 n. 50

Atticus, Titus Pomponius (FRHist 33) Annales Fr 1 71 Fr 2 63 Fr 4 63

Censorinus De die natali 2.2 69 21.1–2 13–14, 19 21.6 62

Augustine of Hippo (Aurelius Augustinus) De civitate Dei 4.22 43 n. 56, 81 4.27 43 n. 62, 44 and n. 64 4.31 82 6.2 77, 80, 84–85, 91 6.3 8, 72, 79 6.4 85 6.5 43, 83 6.8 85, 86 6.10 86 7.5 86 7.17: 3 7.22 80 7.23 88 7.27 88–9 7.28 87 7.33 87 7.35 44 n. 65, 90 8.1 83, 88 18.8 15 19.1 14 n. 24

BICS-60-2  2017

57

Cicero, Marcus Tullius Academica 1.9 1, 34, 62, 71, 72, 91 Brutus 27–28 63 60 65 72 64, 65 156 37 De domo 32

36 n. 14

De legibus 2.47 38 De officiis 1.129 118 De oratore 2.52–53 9

143 © 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:58 +00:00 All use subject to https://about.jstor.org/terms

BICS 60-2 Nov14-2 FINAL-FINAL.indd 143

14/11/2017 14:56:24

BICS-60-2  2017

144

Epistulae ad Atticum 4.14.1 68 4.16.2 68 13.12.3 68 13.25 68 13.25.3 68–69 16.9 59 16.11.3 66 Epistulae ad familiares 9.1 69 9.1–6 75 Philippics 5.53 54

14.7.9 53 16.8.1 50 16.12.5–8 109 17.21 63, 64–65 17.21.24–25 61, 72, 73 17.21.42–43 64, 45 17.21.48 57 17.24.45 65 20.11.4 64 Granius Flaccus (Bremer) Fr I.1

41

Jerome (Eusebius Sophronius Hieronymus) Epistulae 33 36

Dicaearchus of Messana (Mirhady) Fr 54 (= 48 Wehrli) 12, 24 Fr 55 (= 51 Wehrli) 12 n. 15, 24 Fr 56A (= 49 Wehrli) 25, 26–28

Lactantius Institutiones divinae 1.6.7

Dio Cassius 36.39 55 55.3.1–3 55

Livy (Titus Livius) Periochae 97 56

Fabius Pictor, Quintus (Bremer) Fr 3 40 and n. 43

Macrobius Ambrosius Theodosius Saturnalia 1.15.21 46 and n. 73 1.16.18 45 3.4.7 71 3.11.5 71

Festus, Sextus Pompeius (Lindsay) 81 118 204 45 n. 70 223 114 386 41 n. 50 Gellius, Aulus 1.24.3–4 65 2.28.3 45 n. 66 3.10.6 70 4.6.10 42 n. 53 4.10.1–4 53 6.9.11 59 7.12.5–6 41 and n. 45 10.15.3–7 40 and n. 43 13.12.5–6 4 13.12.6 67 13.13.4 57 14.7 49

44 n. 65, 90

Nepos, Cornelius (FRHist 45) Atticus Fr 18.1–2 Chronicon Fr 3

69 62

Nonius Marcellus (Müller) 137.3 (199.3 Lindsay) 67 145.25 (212.5 Linsday) 114 223 (330 Lindsay) 46 and n. 72 Pacuvius, Marcus (Ribbeck) 221 (= 237 W)

121

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 144

14/11/2017 14:56:24

INDEX LOCORUM

Plato Republic 369b 5ff

24

Plautus, Titus Maccius Amphitruo 275 113 Mercator 619–24 113 Persa 408 118 597 118 Truculentus 495 113 Pliny the Elder (Gaius Plinius Secundus) Historia Naturalis 7.115 59 35.11 69 8.199 100 29.7.14 116 Polybius 1.6.2 66 Porphyry De abstinentia 4.2.1–9

25, 26–28

Servius (Maurus Servius Honoratus) Commentarii in Aeneidem 1.54 94 1.172 98, 105 1.246 104 1.382 97, 104 1.449 101 1.532 104 2.166 70 4.167 98 4.427 106 5.145 94 5.409 101–2

145

6.304 102 8.128 97 8.233 105, 106 10.894 96 Commentarii in Georgica 3.24 102 3.446 100 3.273 99, 102 4.63 97 Servius Danielis Commentarii in Aeneidem 1.42 105 1.108 104 1.112 104 1.448 100 2.225 97 3.366 99 8.526 97 Commentarii in Georgica 1.11 96, 100 1.19 96 1.99 94 3.313 102 Solinus, Gaius Julius 1.27

62, 63

Suetonius (Gaius Suetonius Tranquillus) Augustus 35 55 Caesar 44.5 75 52.2 59 78.2 74 Trebatius Testa (Bremer) Fr II.5 Fr II.9

41 and n. 45 40 and n. 44

Valerius Maximus 8.15.8 56

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 145

14/11/2017 14:56:24

BICS-60-2  2017

146

Varro, Marcus Terentius Antiquitates rerum divinarum (Cardauns) Fr 2a 71 Fr 7 15 et n.25 Fr 8 14 n. 31 Fr 9 43 Fr 44 45 n. 69 Fr 78 45 n. 66 VIII App (e) 45 and n. 67 VIII App (f) 46 and n. 73 XIII App (a) 46 and n. 72 Antiquitates rerum humanarum (Mirsch) Fr 17.1 19 n. 39 Fr 19.1 13 Fr 22.2 14 De familiis Troianis (FRHist 52) Fr 3a 70 De gente populi Romani (Fraccaro) Lib. 1 testimonium 13–14 Fr 3 15 De lingua Latina 5.3 10 5.5 10–11, 20 5.7 15 5.7–8 3 5.7–9 17–18, 19 5.13 9 5.33 57 5.61 98 6.2: 2, 20 6.11 109 6.21 118 6.28 116 6.30 45 n. 68 6.86–95 117 6.86–87 52 6.88 52 6.89 52 6.95 52, 117 7.2 15 7.3 19 7.4 3, 11, 15–16

7.8 110–11, 112 7.36 100 7.50 113 7.52 119 7.60 113 7.70 113–14 7.96 109 7.109 68 7.110 9 8.1 9 8.1–23 120 8.5–6 9–10 8.43 110 8.50–51 120–21 8.51 6, 59 8.77 121 9.6 57 9.104 116 9.109 115 10.25 116 10.7 120 10.62 114–15 De re rustica 1.2.16 12 n. 15, 24 1.29 102 2.1.2 11, 2.1.3–5 16 2.1.3–5 12, 13, 24, 25, 26, 25, 28 2.1.11 25 2.1.19 99 3.1.1–8 29–31 3.1.1–4 12–13 3.1.5 33 3.2.3 33 3.2.16 32 3.3.1–8 31 3.4.3 33 3.5.1–7 33 3.9.19 33 3.10.16 97 3.16.3 12 n. 14 De vita populi Romani (Riposati) Fr 114 (= 425 S = 108 P) 58 Fr 123 (= 437 S = 117 P) 58 Fr 124 (= 438 S = 106 P) 58

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 146

14/11/2017 14:56:24

INDEX LOCORUM

Logistorici (Chappuis) Fr 77

14 n. 24

Saturae Menippeae (Buecheler) 84 14 268 14 453 67 577 100 Virgil (Publius Vergilius Maro) Aeneid 1.108–9

104

147

Georgics 3.272–79 99 Xenophon Cyropaedia 8.2.5–6 24 INSCRIPTIONS Corpus Inscriptionum Latinarum I2.24 121 I2.231 59 n. 66 I2.584 121

© 2017 Institute of Classical Studies University of London This content downloaded from 141.70.80.5 on Sat, 20 Apr 2024 08:56:58 +00:00 All use subject to https://about.jstor.org/terms BICS 60-2 Nov14-2 FINAL-FINAL.indd 147

14/11/2017 14:56:24