THE LANGUAGE OF THE FREEDMEN IN PETRONIUS' CENA TRIMALCHIONIS MNEMOSYNE BIBLIOTHECA CLASSICA BATA VA COLLEGERUNT A D. LEEMAN · H. W PLEKET · C . .J. RUI.JGH BIBLIOTHECAE FASCICULOS EDENDOS CURAVIT C. J. RUIJGH, KLASSIEK SEMINARIUM, OUDE TURFMARKT 129, AMSTERDAM SUPPLEMENTUM CENTESIMUM DECIMUM SEPTIMUM BRET BOYCE THE LANGUAGE OF THE FREEDMEN IN PETRONIUS' CENA TRIMALCHIONIS THE LANGUAGE OF THE FREEDMEN IN PETRONIUS' CENA TRIMALCHIONIS BY BRET BOYCE E.J. BRILL LEIDEN • NEW YORK • K0BENHA VN • KOLN 1991 The paper in this book meets the guidelines for permanence and durability of the Com mittee on Production Guidelines for Book Longevity of the Council on Library Resources. Library of Congress Cataloging-in-Publication Data Boyce, Bret. The language of the freedmen in Petronius' Cena Trimalchionis / by Bret Boyce. p. cm.-(Mnemosyne, bibliotheca classica Batava. Supplementum centesimum decimum septimum, ISSN 0169-8958; v. 117) Includes bibliographical references and index. ISBN 90-04-09431-8 1. Petronius Arbiter. Satyricon. Cena Trimalchionis. 2. Freedman-Rome-Language (New words, slang, etc.). 3. Latin language, Vulgar-Texts. 4. Freedmen in literature. 5. Speech in literature. I. Title. II. Series: Mnemosyne, bibliotheca classica Batava. Supplementum; 117. PA6558.A5B69 1991 873'.0l-dc20 91-18778 CIP ISSN 0169-8958 ISBN 90 04 09431 8 © Copyright 1991 by E. J. Brill, Leiden, The Netherlands All r(~hts reserved. No part of this book may be reproduced or translated in any form, by print, photoprint, microfi"lm, microfiche or any other means without written permission from the publisher Authorization to photocopy items for internal or personal J. use is granted by E. Brill provided that the appropriate fees are paid directly to Copyright Clearance Center, 2 7 Congress Street, SALEA1 MA OJ9 70, USA. Fees are subject to change. PRINTED IN THE NETHERLANDS CONTENTS Preface and Acknowledgements ............................................................... vi Chapter One: Introduction ...................................................................... 1 Chapter Two: The Language of the Freedmen ........................................... 36 Phonology .................................................................................... 36 Morphology .................................................................................. 46 Lexicon ........................................................................................ 54 Syntax ......................................................................................... 61 Conclusions .................................................................................. 73 Chapter Three: Language and Characterization of the Freedmen .................... 76 Dama ........................................................................................... 76 Seleucus ....................................................................................... 77 Phileros ....................................................................................... 78 Ganymedes ................................................................................... 79 Echion ......................................................................................... 81 Niceros ........................................................................................ 85 Habinnas ...................................................................................... 87 Hermeros ...................................................................................... 90 Trimalchio .................................................................................... 94 Bibliography ...................................................................................... 103 Index ................................................................................................ 109 PREFACE AND ACKNOWLEDGEMENTS Despite the voluminous literature on the language of the freedmen in Petronius' Cena Trimalchionis the subject has never been treated at length in English, and has not received a fully comprehensive consideration in any language since H. L. W. Nelson's 1947 dissertation Petronius en zijn 'vulgair' latijn. Although Nelson's work was an invaluable contribution to the subject, introducing a new rigor into the study of Petronius' "vulgar" Latin, it sometimes undervalued the popular component in the language of the freedmen. The present study is aimed at reexamining the popular characteristics of the language of the freedmen, as well as offering a new analysis of Petronius' characterization of his various freedmen speakers through their language. I would like to thank all those who assisted me in preparing this study, which is a revised form of my doctoral dissertation submitted at Brown University in 1989. My dissertation adviser, Professor David Konstan, and my readers, Professors Michael C. J. Putnam and William F. Wyatt of Brown University provided invaluable help and encouragement. Professors Alan L. Boegehold and David Pingree of Brown University, and Ann Vasaly of Boston University also provided many helpful suggestions. I would also like to thank a number of others who offered kind advice and prompt bibliographical assistance: Professors John Bodel of Harvard University, Siegmar Dopp of the Ruhr-Universitat Bochum, John Patrick Lynch of the University of California at Santa Cruz, H. L. W. Nelson of the Rijksuniversiteit Utrecht, and Gareth Schmeling of the University of Florida at Gainesville. The reference and interlibrary ~oan staff of the Brown University Library provided indispensable assistance. I thank Messrs. Julian Deahl and Hans van der Meij as well as the rest of the production and editorial staff at Brill for their help in publishing this work. I would also like to thank Bruno Rubio, my family, and all my friends for their support. I am of course responsible for all errors that remain. CHAPTER.ONE INTRODUCTION The Satyrica of Petronius occupies a unique position in the history of ancient literature. The oldest novel of which we possess any substantial remains, it has defied the attempts of critics to assign it to a generic category: variously classed as love-romance, picaresque novel, Menippean satire, travelogue, mime, and burlesque of epic, it is all and none of these, the product of an author steeped in the literary tradition who appears nonetheless deliberately to undertake to shatter the classical norms of stylistic, generic, and topical propriety. Most importantly for our present inquiry, the work of Petronius, and the Cena Trimalchionis in particular, represents, as Erich Auerbach has pointed out in an illuminating essay, "the ultimate limit to which realism attained in antiquity": 1 Petronius is the first extant author fully to attempt "to imitate a random, everyday, contemporary milieu with its sociological background, and to have his characters speak their jargon without recourse to any form of stylization." 2 Homer's Thersites, although a member of the lower orders, speaks the same artificial, stylized and literary dialect used to convey both the narrative and the speeches of the noble characters: there is no attempt to distinguish or characterize speakers according to class or regional disparities by dialect in the epic. In the language of Trimalchio and his freedmen guests, on the other hand, we note numerous lexical and morphological features which serve to distinguish sharply their language not only from the literary norms of silver Latin, but also from the cultivated colloquial Umgangssprache of the narrator Encolpius and the other educated characters of the novel. For Auerbach the key to this difference between Homer and Petronius lies in the inviolate ancient principle of the separation of styles: "Everything commonly realistic, everything pertaining to everyday life, must not be treated on any level except the comic, which admits no problematic probing." 3 This intuition is no doubt basically correct: the high literary genres of tragedy and epic present no realistic treatment of everyday life remotely comparable to that found in comic and satiric literature.4 1 Auerbach (1953) 31. 2 Ibid. 30. 3 Ibid. 31. 4 We may take exception, however, to Auerbach's assertion that Petronius' work "admits no problematic probing," that it is incapable of seriousness, that it contains "nothing which might help us understand the action in terms of its economic and political context" (!). See the criticisms of Auerbach's position in Castorina (1973) 29-32. John Bodel's recent study (Bodel 1984) of the freedmen in Petronius has in fact amply demonstrated with detailed parallels from the historical 2 CHAPTERO NE In what follows we shall investigate the formal, grammatical, and stylistic devices used by Petronius to characterize the freedmen at Trimalchio' s banquet by their speech. At this point a word about terminology is in order. By '(cultivated) colloquial,' 'conversational,' or 'urbane' language we mean the everyday spoken language of the educated upper classes of society. We also apply this term to the written sources which reflect and are in fact our only source of evidence for this educated spoken idiom, such as Cicero's letters, always attempting to bear in mind the fact that the literary sources do not always faithfully represent the spoken idiom, because they are always susceptible to the influence of the stylized literary language.5 By 'vulgar' or 'popular' language we mean the language of the lower classes, in the case of Rome the plebs, which generally lacked access to a liberal education and formed the vast majority of the population of the Empire. We shall apply the term 'general colloquial' to those features which belong simultaneously to cultivated or urbane colloquial speech and to the popular language. The term 'vulgar Latin' is somewhat infelicitous. In the first place, the word 'vulgar' has a pejorative connotation in its ordinary usage; yet the expression vulgaris sermo has been in use at least since Cicero and has by now acquired the consecrated status of a technical term. When we employ the expression 'vulgar Latin' it will be this purely technical sense that is intended, without any pejorative nuance. The term 'vulgar Latin' has further been criticized because it has been applied to such a wide variety of different phenomena: Romance philologists tend to use it to designate a hypothetically reconstructed 'Proto-Romance,' while Latinists generally employ it in reference to the spoken Latin language as known to us (however imperfectly) from actual attestations, primarily in non-literary sources. It is naturally in this latter sense that we shall use the term, although the evidence of the Romance languages will also have to be taken into account. Ernst Pulgram has gone so far as to advocate discarding the term 'vulgar Latin' entirely and replacing it with 'spoken Latin,' since scholars have too often tended to obscure the distinction between the spoken language and the attempts of sub-literary writers to use the classical language while lapsing involuntarily into their native spoken idiom.6 However, Pulgram 's caveats, although valuable, do not really solve the problem, and it is not completely true, as Pulgram states, that Plautus, Petronius, and • Apuleius "give us honest spoken Latin."7 He is more accurate when he says that "we do not know how people of various periods really talked Latin, except by inference and deduction."8 Our sources of spoken or lower-class Latin are either evidence that the portrayals of Trimalchio's freedmen guests, and in particular Hermeros, correspond quite accurately to the economic realities of the first century A. D. and that Petronius does in fact undertake a serious exploration of the psychological effects of slavery and the immutability and inescapability of the civil status of freedmen in the early Roman empire. 5 The fundamental study of the cultivated colloquial language (German Umgangssprache) is Hofmann (1951). 6 Pulgram (1950) and (1958) 311-323. 7 Pulgram (1958) 314. 8 Ibid., 315. INTRODUCTION 3 literary writers attempting to imitate popular modes of speech, or popular writers attempting to imitate literary modes of writing, such as the Church Fathers, Gregory of Tours, the pilgrim Egeria (Aetheria) etc. The vast majority of scholars have not chosen to follow Pulgram but continue to use the term vulgar Latin in spite of its problematic nature: as Carlo Battisti has said, "II termine latino e volgare, ormai cosi radicato nei nostri studi, da non poter essere estirpato; quindi lo subiamo, prendendo nota della sua inadeguatezza ed imprecisione. •'9 Given this general acceptance of the term it would be perverse to reject it; therefore, yielding to convention we adopt it, trying at all times to pay close attention to chronology and to the reservations that we have just mentioned.10 Before proceeding one further preliminary note is in order. In the present study we shall assume that the usual Neronian dating of the Satyrica, supported overwhelmingly by the internal evidence and accepted by the vast majority of modem scholars, is correct. We further consider the identification of the author of the Satyrica with the T. Petronius Niger who was consul suffectus in 62 A. D. and a member of Nero's court to be highly plausible. 11 In any case the old theory of Marmorale (1948) which treats the work as a product of the third century A. D. is to be decisively rejected. The present chapter will begin with a survey of the history of the literary representation of the diction of the lower classes and of foreigners in the works of Petronius' predecessors. We shall then trace the progress of scholarship on the question of 'vulgar' Latin in the work of Petronius himself. Finally, we shall outline the main problems we propose to address in examining the language of the freedmen in the Cena Trimalchionis. A brief survey of Greek and Roman literature will largely confirm the above mentioned thesis of Auerbach that realistic literary depiction of the speech of commoners (and, we may add, of barbarians) is confined to the comic mode. The 'higher' literary genres, which Aristotle defines as those which portray men as "better than in real life" (Poetics 1448a) and for which only "serious" (Ta onouoaia 1448b) style and subject-matter are appropriate, generally eschew any attempt to reproduce the solecisms of the lower classes and of foreigners. We have already noted that this is consistently true for Homer. In the work of the tragedians there are likewise few deviations from this rule. The Persae (471 B. C.) 9 Battisti (1949) 23. 10 For further discussion of the term 'vulgar Latin' see, in addition to the authorities cited above, Nelson (1951); Schmeck (1955) 5-21; Silva Neto (1957) 11-37; Viiliniinen (1967) 3-6; Hermann (1967) 13-18. For some attempts by modem linguists to develop a general theory of the relationship between language and social class see William Labov, Sociolinguistic Patterns (Philadelphia 1972) and Basil Bernstein, Class, Codes and Control: Theoretical Studies Towards a Sociology of Language (2nd. ed., New York 1974). 11 See the definitive study of K. F. C. Rose, The Date and Author of the Satyr icon (Mnemosyne Supplement 16) (Leiden 1971); also, Sullivan (1968) 22-33. The recent argument for a Flavian date by Rene Martin, "Quelques remarques concemant la date du Satyricon," REL (1975) 182-224 has been answered by Bodel (1984) 8-10. 4 CHAPfERONE of Aeschylus has been claimed by some scholars as a possible exception. W. Headlam has suggested that the frequent Ionic forms to be found in this play (e. g. 'laovEc; 178, 563; foi:::Ev 659 e:(Ei:::EivwoEv 761 etc.) are to be explained as a deliberate attempt by the author to characterize the Persian speakers as Asiatics by having them use the dialect of the Ionic colonies in Asia Minor.12 In addition, we find a few foreign expressions, such as l3a~~~v (657-8) and the exclamation 6a, "said by the schol. (M) on 117, 122 to be a nEpo1i:::ov Sp~vriµa." 13 We cannot, however, be certain to what extent these forms may simply be due to the early date of the Persae, now generally considered to be Aeschylus' first extant work, since the early Attic literary language was heavily influenced by Ionic.14 In any case, even if we accept Headlam's view, the play contains only an occasional Ionicism, thus at most a slight tinge of exotic 'Asiatic' coloration, but no real deviation from the high literary style deemed appropriate to tragedy, and certainly no consistent attempt to make the Persian characters speak Ionic, much less the broken patois of foreigners poorly acquainted with Greek. In the other plays of Aeschylus, there is even less of an attempt made at realistic reproduction of linguistic anomalies. For example, at Choephori 563-564 Orestes, about to announce his own death to Clytemnestra, declares that he and Pylades will assume the Pamassian dialect of Phocis to conceal their identity; yet his actual addresses to Clytemnestra (674-690, 700-706) are delivered in pure Attic without the slightest trace of dialect forms. Sophocles exhibits if anything even greater severity than Aeschylus in adhering to the classical doctrine of stylistic propriety. With Euripides the situation is slightly different. As has often been noted, Euripides is much less reluctant than his predecessors to give lower-class characters a significant role in his dramas. Moreover, just as in general he admits far more licence in language and meter than his predeces~ors, so in particular he admits a much larger number of colloquialisms.15 These colloquialisms, however, affect the speech of the noble characters as well as of the commoners; there is little attempt to indicate the social class of a speaker by linguistic peculiarities. Indeed there is nothing in Euripides that can truly be described as 'vulgar' or lower-class diction. His characters still all speak an artificial stylized literary dialect, even if the parameters of that dialect have been somewhat expanded in comparison to his predecessors. The language of the Phrygian in Orestes 1369 ff., for example, has little in common with the language of the Phrygian in Timotheus' Persae (to be discussed below): although a comic effect and perhaps the impression of the 12 CR 12 (1898) 189-190. 13 Broadhead, H. D. The Persae of Aeschylus (Cambridge 1960) xxx. 14 See A. Sidgwick, Aeschylus: Persae (Oxford, 1903) 44. 15 See especially P. T. Stevens, Colloquial Expressions in Euripides (Hermes Einzelschriften, Heft 38) (Wiesbaden 1976).