COLLOQUIAL AND LITERARY LATIN

What is colloquial Latin? What can we learn about it from Romanliterature, and how does an understanding of colloquial Latin enhanceour appreciation of literature? This book sets out to answer such ques-tions, beginning with examinations of how the term ‘colloquial’ hasbeen used by linguists and by classicists (and how its Latin equivalentswere used by the Romans) and continuing with exciting new researchon colloquial language in a wide range of Latin authors. Each chapteris written by a leading expert in the relevant area, and the materialpresented includes new editions of several texts. The introductorysection presents the first account in English of developments in thestudy of colloquial Latin over the last century, and throughout thebook findings are presented in clear, lucid and jargon-free language,making a major scholarly debate accessible to a broad range of studentsand non-specialists.

eleanor dickey is Associate Professor of Classics at the Univer-sity of Exeter. She is the author of numerous books and articleson the Greek and Latin languages and linguistics, including AncientGreek Scholarship: A Guide to Finding, Reading, and UnderstandingScholia, Commentaries, Lexica, and Grammatical Treatises, from theirBeginnings to the Byzantine Period (2007); Latin Forms of Address:From Plautus to Apuleius (2002); and Greek Forms of Address: FromHerodotus to Lucian (1996).

anna chahoud is Professor of Latin at Trinity College Dublin. Sheis the author of C. Lucilii Reliquiarum Concordantiae (1998), and ofarticles on Republican Latin and the grammatical tradition. She hascontributed to the New Oxford Dictionary of National Biography andis a member of the editorial board of The Bryn Mawr Classical Review.

COLLOQUIAL ANDLITERARY LATIN

edited by

ELEANOR DICKEY

and

ANNA CHAHOUD

cambridge university pressCambridge, New York, Melbourne, Madrid, Cape Town, Singapore, Sao Paulo, Delhi

Cambridge University PressThe Edinburgh Building, Cambridge cb2 8ru, UK

Published in the United States of America by Cambridge University Press, New York

www.cambridge.orgInformation on this title: www.cambridge.org/9780521513951

c⃝ Cambridge University Press 2010

This publication is in copyright. Subject to statutory exceptionand to the provisions of relevant collective licensing agreements,

no reproduction of any part may take place withoutthe written permission of Cambridge University Press.

First published 2010

Printed in the United Kingdom at the University Press, Cambridge

A catalogue record for this publication is available from the British Library

Library of Congress Cataloguing in Publication dataColloquial and literary Latin / edited by Eleanor Dickey, Anna Chahoud.

p. cm.Includes bibliographical references and index.

isbn 978-0-521-51395-11. Latin language – Style. 2. Latin language, Colloquial. 3. Latin literature – History and criticism.

4. Speech in literature. I. Dickey, Eleanor. II. Chahoud, Anna. III. Title.pa2311.c65 2010

477 – dc22 2009054015

isbn 978-0-521-51395-1 hardback

Cambridge University Press has no responsibility for the persistence oraccuracy of URLs for external or third-party internet websites referred to

in this publication and does not guarantee that any content on suchwebsites is, or will remain, accurate or appropriate.

In honour of

J. N. Adams

Contents

List of contributors page xAcknowledgements xiiForeword (David Langslow) xiii

part i theoretical framework

1 Introduction 3Eleanor Dickey

2 Colloquial language in linguistic studies 7James Clackson

3 Roman authors on colloquial language 12Rolando Ferri and Philomen Probert

4 Idiom(s) and literariness in classical literary criticism 42Anna Chahoud

5 Preliminary conclusions 65Eleanor Dickey

part ii early latin

6 Possessive pronouns in Plautus 71Wolfgang David Cirilo de Melo

7 Greeting and farewell expressions as evidence for colloquiallanguage: between literary and epigraphical texts 100Paolo Poccetti

8 Colloquial and literary language in early Roman tragedy 127Hilla Halla-aho and Peter Kruschwitz

vii

viii Contents

9 The fragments of Cato’s Origines 154John Briscoe

part iii classical latin

10 Hyperbaton and register in Cicero 163J. G. F. Powell

11 Notes on the language of Marcus Caelius Rufus 186Harm Pinkster

12 Syntactic colloquialism in Lucretius 203Tobias Reinhardt

13 Campaigning for utilitas: style, grammar and philosophy inC. Iulius Caesar 229Andreas Willi

14 The style of the Bellum Hispaniense and the evolution ofRoman historiography 243Jan Felix Gaertner

15 Grist to the mill: the literary uses of the quotidian in Horace,Satire 1.5 255Richard F. Thomas

16 Sermones deorum: divine discourse in Virgil’s Aeneid 266Stephen J. Harrison

part iv early principate

17 Petronius’ linguistic resources 281Martti Leiwo

18 Parenthetical remarks in the Silvae 292Kathleen M. Coleman

19 Colloquial Latin in Martial’s epigrams 318Nigel M. Kay

20 Current and ancient colloquial in Gellius 331Leofranc Holford-Strevens

Contents ix

21 Forerunners of Romance -mente adverbs in Latin prose andpoetry 339Brigitte L. M. Bauer

part v late latin

22 Late sparsa collegimus: the influence of sources on thelanguage of Jordanes 357Giovanbattista Galdi

23 The tale of Frodebert’s tail 376Danuta Shanzer

24 Colloquial Latin in the Insular Latin scholastic colloquia? 406Michael Lapidge

25 Conversations in Bede’s Historia Ecclesiastica 419Michael Winterbottom

Abbreviations 431References 441Subject index 473Index verborum 481Index locorum 489

Contributors

professor brigitte l. m. bauer, University of Texas at Austin

dr john briscoe, University of Manchester

professor anna chahoud, Trinity College Dublin

dr james clackson, Jesus College, Cambridge

professor kathleen m. coleman, Harvard University

dr wolfgang david cirilo de melo, Universiteit van Gent

professor eleanor dickey, University of Exeter

professor rolando ferri, Universita di Pisa

dr jan felix gaertner, Universitat Leipzig

professor giovanbattista galdi, University of Cyprus

dr hilla halla-aho, University of Helsinki

professor stephen j. harrison, Corpus Christi College, Oxford

dr leofranc holford-strevens, Oxford University Press

dr nigel m. kay, independent scholar

dr peter kruschwitz, University of Reading

professor david langslow, University of Manchester

professor michael lapidge, Clare College, Cambridge

dr martti leiwo, University of Helsinki

professor harm pinkster, University of Amsterdam

professor paolo poccetti, Universita di Roma 2 ‘Tor Vergata’

x

Contributors xi

professor jonathan powell, Royal Holloway, University of London

dr philomen probert, Wolfson College, Oxford

professor tobias reinhardt, Corpus Christi College, Oxford

professor danuta shanzer, The University of Illinois at Urbana-Champaign

professor richard f. thomas, Harvard University

professor andreas willi, Worcester College, Oxford

professor michael winterbottom, Corpus Christi College, Oxford

Acknowledgements

Thanks are due above all to Jim, from all of us, for being the wonderfulscholar and mentor that he is. Many of us, unused as we are to working onsuch topics without consulting him, obtained his assistance in disguise, sohe will no doubt find in these pages the results of many queries he respondedto with his characteristic generosity. As usual, thank you Jim! And hugethanks, obviously, are due from the editors to all the contributors, not onlyfor their chapters but also for their enthusiasm, support and promptness,not only with initial submission but also with responses to queries. Theeditors would also like to thank each other! We would also like to thankthe numerous friends of Jim’s who supported this project and would haveliked to contribute to it but were prevented from doing so either by thedistance between this volume’s topic and their own research or by the stricttime frame involved. These include Gualtiero Calboli, Trevor Evans, MarkJanse, John Lee, Robert Maltby, Roland Mayer, Stephen Oakley, JohnPenney, Olga Spevak and Tony Woodman. Lastly, our warmest thanks goto Iveta Adams, whose expertise and care went far beyond a copy-editor’sduties. Her contribution to this volume has been an invaluable and a veryspecial one.

xii

ForewordDavid Langslow

The present volume is to honour a man for outstanding contributions toscholarship, and to thank a friend for support and inspiration. Jim Adams isone of the very best and most important students of the Latin language whohave ever lived. I attempt to sketch some of his achievements below, butfirst let me say this: I know that I can speak for the editors and many othersbesides (not only the contributors to the present volume, but generationsof linguists, historians and classicists – not only Latinists – from manydifferent countries), when I say that we are grateful to Jim not only forbeing the pre-eminent scholar that he is, for opening up and showing theway on numerous new or neglected aspects of Latin, and for publishing hisfindings so quickly, with such clarity and in such abundance, but also forinspiring us, pointing us in the right direction, and helping us to be betterscholars. In my experience – and again I know that I speak for many – Jimhas for decades been generous and unfailing in his readiness to share hislearning, experience and approach, in answering questions, and in readingand discussing, and commenting and advising on, plans in germ and workin manuscript. For a linguist with work in draft, there are few things sobeneficial as having it read by Jim, because he sees straight through theproblems and tells you what the solutions are, and where to look to findthe evidence to prove it. And he is nearly always right – even when youare actually working on a language other than Latin. Jim is a willing andselfless academic mentor to anyone he thinks may be able to profit fromhis help, and it has been well said that he is too loyal a friend for his owngood. It is hard, if not impossible, truly to deserve what Jim has done forus and our subjects, let alone to reciprocate, given the range and depth ofknowledge and understanding necessary to give him equivalent help. Atleast this volume is a tangible token of how keenly appreciated his help hasbeen in so many of our endeavours over so many years.

Jim was born in Sydney in 1943. He was educated at North SydneyBoys’ High School, and graduated from the University of Sydney with

xiii

xiv Foreword

the University Medal for Latin (a special award, made by no means everyyear). A clearly influential teacher, of whom he has often spoken, was G. P.Shipp. After two years as a teaching fellow at his home university, he cameto the UK in 1967, and has stayed ever since. He completed his Oxforddoctorate, a philological commentary on Tacitus, Annals 14.1–54, in underthree years (as a Commonwealth Scholar, at Brasenose College), he heldthe Rouse Research Fellowship in Classics at Christ’s College, Cambridgefor two years, and was appointed to a lecturership in Greek and Latin atthe University of Manchester in 1972, at the age of twenty-nine.

Jim spent twenty-two years in Manchester, being promoted to SeniorLecturer in 1978, to Reader in 1982, and appointed to a personal chair in1993, the year after his election to a Fellowship of the British Academy.In Manchester, he was Chairman of the Departmental Board from 1983,and Head of Department from 1989. He took leave away from Manchesterin 1994–5 in order to take up a one-year Senior Research Fellowship atSt John’s College, Oxford. Soon after, he moved more permanently, to aprofessorship at the University of Reading in 1995, and finally, in 1998, toa Senior Research Fellowship at All Souls College, Oxford. Since 1995, hehas chaired to enormously good effect the supervisory committee of theDictionary of Medieval Latin from British Sources; it is thanks in no smallpart to him that this great project is now as far as possible assured of beingcompleted.

Jim’s subject has always been the Latin language, in the broadestsense – Latin in all its forms and varieties, from the beginning of ourrecord to the emergence of the Romance languages. This is always thebackdrop, no matter how precise his focus at any given moment. Thisfocus is normally very precise indeed, but a signal strength of his work liesin an unusually deft combination of the exhaustive analysis and presenta-tion of details with investigations and conclusions of enormous scope andscale. There are few texts or documents of any sort produced in Latin ofany sort, by speakers or would-be speakers of Latin of any period, place,register, ethnic origin or social class that he does not Know (I am happy tobe able at last to reflect in print the correct observation made to me by afellow PhD-examiner in the late 1990s that ‘Jim Adams knows Latin witha capital K’!).

He has published steadily, at a high rate since the first year of hisCambridge research fellowship (his first article, on a type of hyperbatonin Latin prose, is from 1971), prolifically since the early 1990s, when thearticles came even thicker and faster, and the big books started to appear:Pelagonius, Bilingualism, and Regional Diversification are each longer than

Foreword xv

Vulgar Latin Chronicle, Claudius Terentianus, and Latin Sexual Vocabularyput together – and not a syllable less terse. Since 1990 his research hasincluded (in addition to two jointly authored articles) participation infour large collaborative projects: on the publication and interpretation ofthe texts from Vindolanda (with their editors Alan Bowman and DavidThomas), and, between 1994 and 2005, on jointly organising and editingthe proceedings of three major conferences, respectively on the languageof Latin poetry (with Roland Mayer), bilingualism in ancient society (withMark Janse and Simon Swain), and the language of Latin prose (to honourMichael Winterbottom – with Tobias Reinhardt and Michael Lapidge).

Jim’s published work is characterised by the greatest acuity, rigour, effi-ciency, and good judgement. There is also often a breathtaking boldnessabout both the questions that he takes on and the comprehensiveness ofhis answers to them. The questions either have never occurred to peopleto address before because the necessary connections have not been made,or they have seemed intractable given the (supposed) state of the evidence,or they have been much discussed, and the contradictory half-answers arewell known, or they are (surely!) simply too large and difficult for any-one to tackle single-handed. As one reads Jim’s findings and discussion ofthem, one has the sense of problems and earlier, inadequate solutions beingtransformed or swept aside, and a feeling of finality in his conclusions. Hisstyle is refreshingly unselfregarding, his writing is lucid, terse, urgent. Theurgency of his oral delivery (often against a handout containing – for aone-hour paper – well over a hundred examples), which is transformed intoefficiency and finality on the printed page, is also notable: the lecture thatbecame the 1999 article on nominative personal pronouns was memorablycompared in discussion both to ‘scoring a century before lunch’ and to‘driving through the Ardennes’.

Theory is scarce, though not absent; terminology is traditional, at leastfrom the point of view of linguistics. Jim has always treated more theoreti-cally laden work and new terminology (of the ‘old wine in new bottles’ type)with a healthy scepticism, and has regarded with equal respect and purelyon grounds of merit and utility the fashionable and the little-known. Animportant consequence of the theoretical neutrality and straightforwardempiricism of his approach is that his early works may still be read withundiminished profit side by side with his most recent.

His publications address explicitly texts and authors; grammatical andlexical topics; and, especially more recently in a head-on fashion, someof the big questions facing the student of Latin and Romance linguistics,Latin literature, and Roman history. His studies of named literary texts

xvi Foreword

and authors are concerned with prose much more often than verse – earlyon, Tacitus, the Historia Augusta, Livy, Cicero, Lactantius; most recently,the pseudo-Caesarian Bellum Africum, Petronius, Mustio (the subject ofAdams 2005a) – but Ausonius, Martial, and Latin epic (especially Virgil)are the subjects of three early articles, Catullus and the Augustan poets area main focus of the 1999 article mentioned above, Plautus has always beena central point of reference (and is still so in Bilingualism and RegionalDiversification) – and, on the documentary side, let us not forget theremarkable poets of Bu Njem!

The grammatical domains for which Jim is best known are word orderand vocabulary. His lexical studies sometimes serve to establish the exis-tence, the form and the meaning of unnoticed or misunderstood words (agood number of veterinary terms, but also such ordinary words as the Latinfor ‘to canter’, are thus saved from lexicographical oblivion or misrepresen-tation). Other articles or chapters on vocabulary (as in the cases of wordsfor ‘put’ and ‘throw’, or of some of the anatomical terms transferred fromanimals to humans, or of some of the new words attested at Vindolanda)illustrate changes under way in Latin foreshadowing Romance usage. Most,however, provide object lessons in how to use the distribution of synony-mous or complementary words as evidence for their register, their social orconnotational meaning, so (e.g.) the articles on words for ‘woman’, ‘wife’,‘prostitute’, ‘kill’, and the seven chapters of the Latin Sexual Vocabulary,the publication of which prompted an article in the Sunday Times.

Jim’s interest in word order dates from the 1970s. It is the subject ofhis very first article, on hyperbaton (mentioned above), and 1976 saw thepublication of the still-important article on ‘a typological approach to Latinword order’. It has yielded along the way the 1991 article on the constructioninfinitive + habeo in late Latin and the origin of the Romance future (Fr.je chanterai, Ital. cantero, etc.), in which a detail of variable word orderis made in masterly fashion to throw light on an important problem ofhistorical morphology and semantics. In many respects, his work on wordorder supersedes that of the great French Latinist Jules Marouzeau (L’ordredes mots), and here more than anywhere embraces recent developments inlinguistic theory, notably in pragmatics. Jim’s crowning achievement in thestudy of Latin word order is to have modified (in a pair of long papers bothpublished in 1994) Eduard Fraenkel’s modification of Wackernagel’s Law byredefining the ‘second position’ as following not the first word of the clause(so Wackernagel) nor of the ‘colon’ (so Fraenkel) but rather a ‘preferentialhost’, one of a set of accented but not necessarily emphatic words to whichunstressed elements (including pronouns and weak forms of the verb ‘to

Foreword xvii

be’) are attracted. In a brilliant coda to the ‘unstressed pronouns’ paper, amost appealing and suggestive historical connection is made between theenclisis on the newly observed ‘host’ in Latin and the proclisis on the verbcharacteristic of Romance. This is, by any standards and for any language-family, an unusually illuminating and satisfying account of the history ofa syntactic pattern, from the prehistoric parent language to the modernvernacular languages.

In truth, however, although lexicography and word order are clearlymore prominent in Jim’s bibliography, there are few areas of the grammar(from spelling and phonology to sentence structure) that he has not coveredin his studies of (e.g.) the Anonymus Valesianus II, Claudius Terentianus,C. Novius Eunus, the Bath curse tablets, the ostraca from Bu Njem, theVindolanda letters, all of which amount to comprehensive, if selective,contrastive grammars of these texts. (I say ‘selective’ because that is howJim presents them: in fact, nothing much worth saying is left to say at theend.)

The big questions about Latin that have clearly interested Jim, andcontinue to interest him, include: (a) What are the limits of variationthat we can document within Latin, and against what sort of parameters –chronological, geographical, sociolinguistic, stylistic in the broadest sense –may we describe each variable? (b) What sorts of contact did Latin enjoywith other languages, and what are the grammatical effects, on Latin andon the other language, in each contact situation? (c) How did one language,Latin, evolve into the several Romance languages?

All three questions are addressed already in early publications. The evo-lution of Latin into Romance is illustrated in 1977 in the foreshadowingof French words in the Annales regni Francorum, and numerous aspects ofthe wider problem are discussed in a 1989 review article of Roger Wright’simportant book Late Latin and Early Romance (1982). Features of bilin-gualism and consequent linguistic interference are written up in the 1970sfirst for the influence of Greek on Egyptian Latin. Regional variation isinvestigated to begin with in the Latin of Egypt and Britain; sociolinguisticvariation is documented in features of the Latin of military documents, andin characteristics of female speech in Latin comedy; stylistic variation, andwhat may be inferred from it about authorship and chronological devel-opment, prompts some of Jim’s earliest articles, on Tacitus, the writers ofthe Historia Augusta, and Livy.

In pursuing these questions, Jim has for nearly forty years consistentlyboth opened up important new areas to philological study and shed freshlight on familar authors and texts. Many of the writers and documents

xviii Foreword

that he has studied were largely unknown to students of antiquity, and, butfor his work, would probably have remained so. Jim has shown that themost unpromising material, examined in the right way, can yield importantinsights of quite general relevance. His treatment of the high literary, thehighly technical, and the most ‘hopeless gibberish’ alike renders his workessential reading for all Latinists and Roman historians. In many cases, itis evident from the titles or tables of contents of his works that a vast arrayof literary and documentary evidence will be surveyed and appraised, butthe same is true also of, for example, Pelagonius, which treats an importantbut neglected chapter of Roman social history, offers an extraordinarynumber of instructive insights into the Latin language in various periodsand registers, and has much to say on Latin literary topics (cf. my reviewin BMCR 97.04.01). In discussions of international collaborative work ontechnical Latin, Jim has argued repeatedly and forcefully against restrictingoneself to the technical writers, since so much important evidence is tobe found in non-technical literary Latin and in inscriptions of all kinds:conversely, to characterise the language of, say, Cicero, Virgil or Ovidwith reference just to other literary authors risks no less yielding a partial,impoverished and misleading view.

If, more generally, we may readily agree that we ignore at our perilthe contribution that the fullest possible understanding of a languagecan make to the historical or literary interpretation of texts and theircontexts, it remains true that such an understanding of Latin is more easilyaspired to than achieved. It is also true that few have approached such anunderstanding so closely or made its implications so widely available asJim. A former colleague of mine rightly remarked that ‘Jim Adams is asclose as we can get to a native speaker of Latin.’ The wise philologist, orliterary critic or historian, knows to interrogate native speakers. Certainly,the extent to which ‘Adams, J. N.’ features in a bibliography is a goodindicator of the degree to which the author regards the Latin language asof relevance to the theme.

W. M. Lindsay ended his appreciation of the life and work of FranzSkutsch (CR 26 [1912] 238) with the words, ‘What shall we do now that ourprotagonist is gone?’ Thank Goodness, ours is still with us, and showingno sign of slowing down. May Jim’s forthcoming anthology and discussionof non-standard Latin texts and features be very far from his last words onthe Latin language!

part i

Theoretical framework

chapter 1

IntroductionEleanor Dickey

What is colloquial Latin? What is literary Latin? ‘Literary’ is a famouslycontested term, and ‘colloquial’ is no less fraught with difficulties. Notonly is its precise meaning unclear, but it is laden with value judgements:some consider it a pejorative term and others a positive one. The word hasbecome involved in the social struggle over the relative value of differentvarieties of language and as such has been given a wide range of differentimplications and connotations over the centuries, some complementaryand others contradictory. In order to use this word in scholarly discourse,one first needs not only to determine what it means, but also to explainhow one’s usage resembles and differs from that of others who have usedthe same term.

The Concise Oxford Dictionary (Fowler and Fowler 1995) defines ‘collo-quial’ as ‘belonging to or proper to ordinary or familiar conversation, notformal or literary’ while defining ‘literary’ as ‘of, constituting, or occupiedwith books or literature or written composition, esp. of the kind valued forquality of form . . . (of a word or idiom) used chiefly in literary works orother formal writing’. Such definitions tell us a number of different thingsabout the way these terms are normally used:– ‘colloquial’ and ‘literary’ refer to registers1, with literary being a higher,

more formal, register than colloquial;– they are defined in part by opposition to each other, as is often the case

with registers;– they are genre-dependent, each being proper to particular genres of

communication;2

– the distinction between them is connected to the distinction betweenspoken and written language.

1 For the concept of register see p. 10.2 In the linguistic, rather than the literary, sense of ‘genre’: a genre of communication is a type of

circumstance in which one might use language, such as a conversation, a formal lecture, a poem, anewspaper article, etc.

3

4 eleanor dickey

The difficulties with such definitions are numerous. First, they givethe impression that all linguistic features can be divided between thesetwo categories (i.e., whatever is not literary is colloquial), and yet theycannot.3 Many words and usages are register-neutral, usable in any varietyof language: it would be as silly to ask whether English ‘and’ is an elementof literary or of colloquial language as it would be to ask to which registerthe English present tense belongs. Moreover literary and colloquial arenot the only registers that exist; some words and usages, such as technicalterminology, belong to registers distinct from them both.

The connection between register and genre is likewise not as straightfor-ward as it seems. Ordinary familiar conversation is a genre that can spana wide variety of registers, in part because there is a connection betweenregister and social status: the ordinary conversational language of people ofhigh status tends to have more high-register characteristics than the ordi-nary conversational language of people of low status (indeed the language isconsidered high-register by speakers precisely because it is characteristic ofhigh-status speakers). The differences between these two extremes of con-versational language can be pronounced, to the extent that they may sharelittle that is not common to other registers of the language, and therefore itcan be difficult to say anything meaningful about conversational languageas a whole.

Recognition of this problem has led to restricted uses of ‘colloquial’,referring either to the conversational language of low-status people or thatof high-status people. When used in the first sense, ‘colloquial’ can beequivalent to ‘ungrammatical’ or otherwise ‘wrong’.4 When used in thesecond sense, however, it can become a kind of Holy Grail of languageusage, both for native speakers5 and for modern students of languageslike Latin or ancient Greek (since considerable social prestige has in somerecent periods been attached to a full command of those languages).

Restrictions of this type are useful, but they also cause some problems.Often authors use the term in one or the other restricted sense withoutindicating which one is meant, and this can cause considerable ambiguity.In addition, from any restriction of the term to a particular sort of conver-sational usage it follows that a substantial segment of the population either

3 Whether all language can be divided up among registers is a different question; probably it can.A passage of reasonable length will normally contain enough differentiatable linguistic features tomake it possible to classify it as literary, colloquial, or something else, and therefore it is not easy tosee how it could be wholly register-neutral.

4 See Samuel Johnson, quoted in Simpson and Weiner 1989: s.v. ‘colloquial’ 2: ‘ . . . to refine ourlanguage to grammatical purity, and to clear it from colloquial barbarisms’.

5 See Chapter 2.

Introduction 5

lacks a colloquial register entirely or commands it only with difficulty as apartially learned register.6 This result clashes with the normal meaning ofthe term ‘colloquial’ enough to cause a sense of internal contradiction forreaders.

On the other side of the register/genre question, most literary genresare capable of accommodating a range of registers. Indeed the interplay ofregisters is often part of what gives literature its richness. One cannot fullyappreciate the language of any literary genre unless one is able to considerthe possibility that it may involve register variation, and in many cases onemust recognise that literary works include some language belonging to lowregisters.

The equation of register/genre distinctions with those between speechand writing causes additional difficulties. Literary language (whether thismeans high-register language or the language of literary genres) is notsimply the same as written language, nor is colloquial language (whetherthis means low-register language or the language of some conversationalgenre(s)) the same as spoken language. The language of the Homericpoems is generally agreed to be a literary one – that is, elevated, remotefrom ordinary conversation, and used only for poetry – but also generallyagreed to have evolved in a society that had no knowledge of writing.Even in societies where the production of high-register works in literarygenres is likely to be aided by writing, their delivery is often oral. In ourculture, plays, lectures and speeches are delivered orally, often after beingcomposed in writing – but not always, as many successful speakers do notuse a written text and some plays are improvised or incorporate changesto the original script made during rehearsal and not written down. Thefact that a member of the audience often cannot tell from the languageused whether the giver of a speech or lecture is following a written text, orwhether a comedy sketch was written or improvised, shows that there is noclear and simple connection between linguistic register and spoken versuswritten language production.

In fact, any register can be produced orally: doctors, lawyers and otherspecialists use technical language as readily as politicians produce their ownspecial genre, and some academics have a habit of delivering in conversation

6 Not all conversation is conducted in the idiom that its speakers find easiest and most natural, becausethere are often advantages to using a type of language with which one is less familiar. In this contextone normally thinks of people of low-status origin who as adults attempt to acquire a higher-statusconversational register, e.g. the situation in Shaw’s Pygmalion or Moliere’s Bourgeois gentilhomme. Butthe reverse also occurs: people who find themselves surrounded by speakers of a lower conversationalregister may also try to acquire elements of that register, particularly if use of higher-register featuresgenerates hostility in their conversational partners.

6 eleanor dickey

sentences so grammatically complex that listeners wish they had beenwritten down. At first glance, however, it seems that the same cannot besaid for writing: the language of published books and periodicals, whilereasonably diverse, does not have the same range as that of orally deliveredspeech. But published language is only a subset of written language, and ifone turns to non-print media such as the Internet, much greater diversityin written genres becomes apparent. The development of e-mail and textmessaging has led to the blossoming of very informal written genres andto the evolution of new linguistic features attached to those genres. Arethese genres and their distinctive linguistic features literary because theyare found exclusively in written contexts, or are they colloquial becausethey belong to an informal register? Is the distinction between spoken andwritten language that seemed so striking in nineteenth- and twentieth-century Western society merely a temporary phenomenon that was bornwith the rise of publishers and editors and is now becoming obsolete asthey lose control over the distribution of written language? These are only afew of the problems with the spoken/written divide; in fact even before therise of electronic communication linguists were discarding a classificationof language based on spoken versus written format and replacing it withclassifications based on genre.7 It is therefore highly problematic to usethe written/spoken distinction to help understand the literary/colloquialdistinction.

The common usage of the term ‘colloquial’ is thus of little use to ascholar, but at the same time it is not possible to investigate colloquial Latinwithout understanding what it is or may be; a clear understanding of thequestion is essential for answering it. Let us therefore turn to three sourcesthat might be more help: the science of linguistics, the terminology usedby the Romans themselves to talk about variation in their language, andthe ways scholars have traditionally used the term ‘colloquial’ in discussingLatin and Greek.

7 See Biber (1988: esp. 52–3) and Chafe and Tannen (1987).